Разработка_моделей_ЭТАПЫ_3_4.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"collapsed_sections":[],"toc_visible":true,"authorship_tag":"ABX9TyPtJ8y7PfzdBoRSlhBnZWy5"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"widgets":{"application/vnd.jupyter.widget-state+json":{"f216f87e1d6543c6b8f6be9fd31548d1":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_550aa8ff89a145d696563cd6e3aebd01","IPY_MODEL_a87f5f67dd3b4863a818b2913cf1a9d7","IPY_MODEL_77c3c6ebae7444f7bcd22e698942a5e5"],"layout":"IPY_MODEL_4aca4516143b4478989a27b9658449e1"}},"550aa8ff89a145d696563cd6e3aebd01":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_004d3630614f44c39bb06709a98a9ea6","placeholder":"​","style":"IPY_MODEL_93121f2cf43a4fce9be594d114dfcfcb","value":"100%"}},"a87f5f67dd3b4863a818b2913cf1a9d7":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_6c56d363ef8240d09dddc2ac408ecda0","max":500,"min":0,"orientation":"horizontal","style":"IPY_MODEL_424784ce47434cc08605a1d8098dfd9a","value":500}},"77c3c6ebae7444f7bcd22e698942a5e5":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_dd034d45b3c54a0988967d6e036b8031","placeholder":"​","style":"IPY_MODEL_551a6a59c9824f129fe104d434506fd1","value":" 500/500 [01:18&lt;00:00,  3.20it/s, train_auc=100.00%, skipped=0.20%]"}},"4aca4516143b4478989a27b9658449e1":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"004d3630614f44c39bb06709a98a9ea6":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"93121f2cf43a4fce9be594d114dfcfcb":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"6c56d363ef8240d09dddc2ac408ecda0":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"424784ce47434cc08605a1d8098dfd9a":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"dd034d45b3c54a0988967d6e036b8031":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"551a6a59c9824f129fe104d434506fd1":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"f2e6f9809d5647958a069043e125f0e0":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_7538f441cd474c209b44698cbb584869","IPY_MODEL_8793c78889454339b845cd41afe1958b","IPY_MODEL_fd96142026814e2389f226d1401f75a4"],"layout":"IPY_MODEL_88fe8519247a4b36af28d8afd56f43e8"}},"7538f441cd474c209b44698cbb584869":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_fc7ae99971394a0ebeb3d8ace49bfcc3","placeholder":"​","style":"IPY_MODEL_e1e6d626578644bbb2fa9b6961cae4d1","value":"100%"}},"8793c78889454339b845cd41afe1958b":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_5411bd2d7abd4c3c9f12b5952365d677","max":500,"min":0,"orientation":"horizontal","style":"IPY_MODEL_27eec663955b40e2980e60500f810d8b","value":500}},"fd96142026814e2389f226d1401f75a4":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_405804a98729480694569712dfa8b5a8","placeholder":"​","style":"IPY_MODEL_8d37dd178f814380ad460e89ad34d27b","value":" 500/500 [00:23&lt;00:00, 23.44it/s, train_auc=99.99%, skipped=0.11%]"}},"88fe8519247a4b36af28d8afd56f43e8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"fc7ae99971394a0ebeb3d8ace49bfcc3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e1e6d626578644bbb2fa9b6961cae4d1":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"5411bd2d7abd4c3c9f12b5952365d677":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"27eec663955b40e2980e60500f810d8b":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"405804a98729480694569712dfa8b5a8":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"8d37dd178f814380ad460e89ad34d27b":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b2a8cf3a7e1e4b2bab42c1016dcba651":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_4dccdd4c2a224ee6a05b2966b5370b4d","IPY_MODEL_1997e91222a747fd94f2e027c936aeec","IPY_MODEL_83915256a5a4444fa7eec1882d549743"],"layout":"IPY_MODEL_c7fea9886a76480690ed1b420eb99ced"}},"4dccdd4c2a224ee6a05b2966b5370b4d":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_96d7def756cf45a6914ab84fafe06d45","placeholder":"​","style":"IPY_MODEL_74bcb9b9c4f74c03b1eb792cd3ca8e55","value":"100%"}},"1997e91222a747fd94f2e027c936aeec":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_3546206a61da4de9bc34a710c81e688f","max":500,"min":0,"orientation":"horizontal","style":"IPY_MODEL_de6ae8914b824dda968e103aef2fb12f","value":500}},"83915256a5a4444fa7eec1882d549743":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_046f2f7bdc604c509cd35046fe1c4d90","placeholder":"​","style":"IPY_MODEL_ced7dacb1d644485880fd9de17bc5946","value":" 500/500 [00:10&lt;00:00, 47.69it/s, train_auc=99.98%, skipped=0.13%]"}},"c7fea9886a76480690ed1b420eb99ced":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"96d7def756cf45a6914ab84fafe06d45":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"74bcb9b9c4f74c03b1eb792cd3ca8e55":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"3546206a61da4de9bc34a710c81e688f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"de6ae8914b824dda968e103aef2fb12f":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"046f2f7bdc604c509cd35046fe1c4d90":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ced7dacb1d644485880fd9de17bc5946":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"6f53e5b8ccb04934ac4aa271e6814b3e":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_89d8bba95fa1424ebd5de8c2b0823f53","IPY_MODEL_dfc358cdf19944ab9284003faf50a4f9","IPY_MODEL_48f72d2951f24c12a5f4d5cb0b55bd00"],"layout":"IPY_MODEL_b6246f2dc9ab416187b7a887a0ec644c"}},"89d8bba95fa1424ebd5de8c2b0823f53":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e7ce64781bb44c169c04ab20b80d3488","placeholder":"​","style":"IPY_MODEL_420839ad85a447c2aeb97857a404aa99","value":"100%"}},"dfc358cdf19944ab9284003faf50a4f9":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_0d70d7e0e07645a4b779addd6f198bac","max":500,"min":0,"orientation":"horizontal","style":"IPY_MODEL_dc461efabf9345ff880372430894d9bb","value":500}},"48f72d2951f24c12a5f4d5cb0b55bd00":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_722493f48cda4f4aba7cda26608ef9fb","placeholder":"​","style":"IPY_MODEL_25ded155fa4d4437921ece0a67efeacf","value":" 500/500 [00:07&lt;00:00, 57.93it/s, train_auc=99.97%, skipped=0.34%]"}},"b6246f2dc9ab416187b7a887a0ec644c":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e7ce64781bb44c169c04ab20b80d3488":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"420839ad85a447c2aeb97857a404aa99":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0d70d7e0e07645a4b779addd6f198bac":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"dc461efabf9345ff880372430894d9bb":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"722493f48cda4f4aba7cda26608ef9fb":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"25ded155fa4d4437921ece0a67efeacf":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b2175aa5b41545e0a0548f80c943e765":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_0fabdd6ed6b24617a3c04588957dde21","IPY_MODEL_bea3b418152d4a638834731e47037a82","IPY_MODEL_bf99d386bcab4e089f3f80776f0d7cc8"],"layout":"IPY_MODEL_537e2acdbe28445c905615884ab1f566"}},"0fabdd6ed6b24617a3c04588957dde21":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_a70faaff7d7842b79bbd8436be1c3c18","placeholder":"​","style":"IPY_MODEL_63b8aaecfa0e41d7b269cf54433efb2d","value":"100%"}},"bea3b418152d4a638834731e47037a82":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_73d8fde024b64890a279ba6deb08525b","max":500,"min":0,"orientation":"horizontal","style":"IPY_MODEL_ff8726818c674deba469fb649cbce022","value":500}},"bf99d386bcab4e089f3f80776f0d7cc8":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_639858ddd1bd4262a50e9da87c1f4f3c","placeholder":"​","style":"IPY_MODEL_155cdde2a04248c1bcdf0f2111895802","value":" 500/500 [00:05&lt;00:00, 53.24it/s, train_auc=99.98%, skipped=0.47%]"}},"537e2acdbe28445c905615884ab1f566":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a70faaff7d7842b79bbd8436be1c3c18":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"63b8aaecfa0e41d7b269cf54433efb2d":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"73d8fde024b64890a279ba6deb08525b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ff8726818c674deba469fb649cbce022":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"639858ddd1bd4262a50e9da87c1f4f3c":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"155cdde2a04248c1bcdf0f2111895802":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"01a50f309a074618a7ce9200c8d7b83b":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_3f3bc3b6a60a410a8fba3f89ab3a6700","IPY_MODEL_01a85be2f26f42449e501058ab967881","IPY_MODEL_3c94349ee03a4630aefcdbd7aa9e5659"],"layout":"IPY_MODEL_5d5ad7d9c2464e86ab6285740894ce0b"}},"3f3bc3b6a60a410a8fba3f89ab3a6700":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_3311475c6c2c41fb85c794ac73e2f0f4","placeholder":"​","style":"IPY_MODEL_9dd75561063c4463ab0730dc0b1dad05","value":"100%"}},"01a85be2f26f42449e501058ab967881":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_3e73e067d1b448258b3b9d6fbe0589bf","max":500,"min":0,"orientation":"horizontal","style":"IPY_MODEL_91aef3d5282f46a38e0f78d1634a17b0","value":500}},"3c94349ee03a4630aefcdbd7aa9e5659":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_c6a6a6d624d242398b953853fc350220","placeholder":"​","style":"IPY_MODEL_40c54eedd0c5481dbb9e3a08d431317f","value":" 500/500 [00:59&lt;00:00,  8.58it/s, train_auc=100.00%, skipped=0.02%]"}},"5d5ad7d9c2464e86ab6285740894ce0b":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"3311475c6c2c41fb85c794ac73e2f0f4":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9dd75561063c4463ab0730dc0b1dad05":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"3e73e067d1b448258b3b9d6fbe0589bf":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"91aef3d5282f46a38e0f78d1634a17b0":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"c6a6a6d624d242398b953853fc350220":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"40c54eedd0c5481dbb9e3a08d431317f":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"04928ddd4ac24b5392223d3501ed46ca":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_6d18324b5865442192ad606b4cab110e","IPY_MODEL_d9928b4873b344729aafcbdcd5ad5a93","IPY_MODEL_b14ff5a3028d4b0bb963582b547cafa6"],"layout":"IPY_MODEL_b1583bd219cd4c16b3ea8629dfb792bd"}},"6d18324b5865442192ad606b4cab110e":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_0deb722ae8ea474a92fca24d9afbec08","placeholder":"​","style":"IPY_MODEL_df04df190fe64103a5cef6796a1ffa1d","value":"100%"}},"d9928b4873b344729aafcbdcd5ad5a93":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_e01ad7cce5414efd8d9864aff5774a60","max":15,"min":0,"orientation":"horizontal","style":"IPY_MODEL_1a9da9d624ae4dbdaf8faaac31198275","value":15}},"b14ff5a3028d4b0bb963582b547cafa6":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_2d5e92206b6d41eea4202aebb8721dd3","placeholder":"​","style":"IPY_MODEL_c46dd76b5dda4e7a9e2b85328d691fbb","value":" 15/15 [05:24&lt;00:00, 22.37s/it]"}},"b1583bd219cd4c16b3ea8629dfb792bd":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"0deb722ae8ea474a92fca24d9afbec08":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"df04df190fe64103a5cef6796a1ffa1d":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"e01ad7cce5414efd8d9864aff5774a60":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"1a9da9d624ae4dbdaf8faaac31198275":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"2d5e92206b6d41eea4202aebb8721dd3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"c46dd76b5dda4e7a9e2b85328d691fbb":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"ba19ba6d26804281859004131ea5c303":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_4a761abd0d4d4dec992b5259f2a8b8e4","IPY_MODEL_54d2783412794142b07b72fd21b58c0c","IPY_MODEL_20ccf552c6a34b5bbb4c846d2e52de5f"],"layout":"IPY_MODEL_76932ce9ec914e21bdc47c72ceba5653"}},"4a761abd0d4d4dec992b5259f2a8b8e4":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e77dc869ec4e441ab854e581e10d7bc1","placeholder":"​","style":"IPY_MODEL_7a4ac19e05a549f5a2e60ee46a30c3c2","value":"100%"}},"54d2783412794142b07b72fd21b58c0c":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_11ff27baecef47b6ad978763913a54ae","max":15,"min":0,"orientation":"horizontal","style":"IPY_MODEL_9f8906ba962d4f9c852bf25db269b145","value":15}},"20ccf552c6a34b5bbb4c846d2e52de5f":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_91dea137df8540a78a2243c5fefa485f","placeholder":"​","style":"IPY_MODEL_ad15c24ba0cf4bcbb8f2dfa9e21e4b44","value":" 15/15 [03:54&lt;00:00, 17.32s/it]"}},"76932ce9ec914e21bdc47c72ceba5653":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"e77dc869ec4e441ab854e581e10d7bc1":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7a4ac19e05a549f5a2e60ee46a30c3c2":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"11ff27baecef47b6ad978763913a54ae":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9f8906ba962d4f9c852bf25db269b145":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"91dea137df8540a78a2243c5fefa485f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ad15c24ba0cf4bcbb8f2dfa9e21e4b44":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"a9dfac0af13140afb31fba779cde470b":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_d4988bbdb533474490cc6a60f6c1c027","IPY_MODEL_a485092a06084c808ce1feb49366a209","IPY_MODEL_06e5a66e8be54c33be93f4dd3d887b3c"],"layout":"IPY_MODEL_e5f09affcfa14066b9ff0e184cc7e24f"}},"d4988bbdb533474490cc6a60f6c1c027":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_cf2a9c6f0e17427781e76955166ee329","placeholder":"​","style":"IPY_MODEL_1e0c7d5af18d4cb6bda9091e8f729fc8","value":"100%"}},"a485092a06084c808ce1feb49366a209":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_a8e16620671140989a5506f63b34fcca","max":15,"min":0,"orientation":"horizontal","style":"IPY_MODEL_4a31b50b4b364b97ab9b787573f16527","value":15}},"06e5a66e8be54c33be93f4dd3d887b3c":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_cd49916be24f4f65a2982f12e849bf5a","placeholder":"​","style":"IPY_MODEL_74b142ccebe24c16b5816bed42643c59","value":" 15/15 [03:07&lt;00:00, 14.37s/it]"}},"e5f09affcfa14066b9ff0e184cc7e24f":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"cf2a9c6f0e17427781e76955166ee329":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"1e0c7d5af18d4cb6bda9091e8f729fc8":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"a8e16620671140989a5506f63b34fcca":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"4a31b50b4b364b97ab9b787573f16527":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"cd49916be24f4f65a2982f12e849bf5a":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"74b142ccebe24c16b5816bed42643c59":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"d17201ffb8b843619d6f737bd0c3c98a":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_dffece2752d74c439a244f776f1d9bb6","IPY_MODEL_5216639b0ed542dcb6c7b05e0b403563","IPY_MODEL_7897cb7e1de347cfa77b24b606b87226"],"layout":"IPY_MODEL_10721637d4874205a1f75b2ddd459acd"}},"dffece2752d74c439a244f776f1d9bb6":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_06a8edc134af46c385c2ccb1346f3049","placeholder":"​","style":"IPY_MODEL_b27fb7519c364da2b46e40b5414eae38","value":"100%"}},"5216639b0ed542dcb6c7b05e0b403563":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_a1bdef50d82a4f04ae9630f710932e0a","max":15,"min":0,"orientation":"horizontal","style":"IPY_MODEL_72d9a78270644827bfb8267e0a8a1af7","value":15}},"7897cb7e1de347cfa77b24b606b87226":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_5d2ddf3224ec4318b793de72e1db78e3","placeholder":"​","style":"IPY_MODEL_4138ef5f27f840bc88b17a5196eefed6","value":" 15/15 [05:29&lt;00:00, 22.36s/it]"}},"10721637d4874205a1f75b2ddd459acd":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"06a8edc134af46c385c2ccb1346f3049":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"b27fb7519c364da2b46e40b5414eae38":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"a1bdef50d82a4f04ae9630f710932e0a":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"72d9a78270644827bfb8267e0a8a1af7":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"5d2ddf3224ec4318b793de72e1db78e3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"4138ef5f27f840bc88b17a5196eefed6":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}}}}},"cells":[{"cell_type":"markdown","source":[">>[Импорт библиотек](#scrollTo=f5iBcowrvgTT)\n","\n",">>[Загрузка файлов](#scrollTo=w33Da8I_TMCc)\n","\n",">>[ЭТАП 3. Возможности Collaborative Filtering](#scrollTo=_GgFB5ueMdb6)\n","\n",">>>[(3.1) CF для активного пула читателей и книг](#scrollTo=qX9GAiF-kF5c)\n","\n",">>>>[Вводные](#scrollTo=8UPostUc1ull)\n","\n",">>>>[Самые активные пользователи](#scrollTo=hrMeH5bDkOQv)\n","\n",">>>>[Книги, которые читают](#scrollTo=OSKqhUxMU0en)\n","\n",">>>>[Эксперименты с моделью](#scrollTo=H83Z-lY1U_yq)\n","\n",">>>>>[Код для экспериментов](#scrollTo=jU-IvOP6sAFK)\n","\n",">>>>>[Фильтрация книг - как меняется датасет?](#scrollTo=dgBs7jwGtnIg)\n","\n",">>>>>[Модель для \"активных\" книг](#scrollTo=6wFeDJ4yt6LQ)\n","\n",">>>>>[Модель для \"активных пользователей\"](#scrollTo=AhpArqly0CXS)\n","\n",">>>[(3.2) Дублирующиеся книги - баг или фича?](#scrollTo=b3_ggQzldhpx)\n","\n",">>>>[Вклад каждой книги в историю читателя](#scrollTo=crQ2D2KiotJ9)\n","\n",">>>>[Нормализация весов и обучение модели](#scrollTo=O2MLwV5roJav)\n","\n",">>>>[Оценка результатов](#scrollTo=ZOlCoMTBvSR1)\n","\n",">>>>[Тест на платформе](#scrollTo=mMzHDkTE-xxz)\n","\n",">>>[(3.3) Рекомендации новых авторов](#scrollTo=km5Txxb6Mx4A)\n","\n",">>>>[Вводные](#scrollTo=ss6O8CLxX7_a)\n","\n",">>>>[Добавляем авторов в датасет](#scrollTo=Par5y4ezYbNa)\n","\n",">>>>[Популярные и любимые авторы](#scrollTo=_6GyJBe0jHEz)\n","\n",">>>>[Обучаем модель предсказаний](#scrollTo=_Gx8F6uUrAQ5)\n","\n",">>>>[Общий вывод](#scrollTo=kDyla-lA2YgU)\n","\n",">>[ЭТАП 4. Интерпретация предсказаний](#scrollTo=BdHBbcx0dP_q)\n","\n",">>>[(4.1) Загрузка предсказаний](#scrollTo=gxZYkcA4gLgD)\n","\n",">>>[(4.2) Какие книги мы рекомендуем?](#scrollTo=0biIxLfHhZkz)\n","\n",">>>[(4.3) Общий Hit-Rate](#scrollTo=0V0oFUVqdfeK)\n","\n",">>>[(4.4) Какие книги мы угадываем?](#scrollTo=32qk-UpSheoL)\n","\n",">>>[(4.5) Пользователи, которых мы угадываем](#scrollTo=6ntNZeQ0dd2k)\n","\n",">>>[(4.6) Неявные дубликаты книг](#scrollTo=5Annr7d30Pb7)\n","\n"],"metadata":{"colab_type":"toc","id":"ccU-Jk6sKmp0"}},{"cell_type":"markdown","metadata":{"id":"f5iBcowrvgTT"},"source":["## Импорт библиотек"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":6566,"status":"ok","timestamp":1663074053830,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"DqShdtHevczV","outputId":"fdae1421-34df-474f-bc7d-08eb77c36b66"},"outputs":[{"output_type":"stream","name":"stdout","text":["Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Collecting implicit\n","  Downloading implicit-0.6.1-cp37-cp37m-manylinux2014_x86_64.whl (18.6 MB)\n","\u001b[K     |████████████████████████████████| 18.6 MB 335 kB/s \n","\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from implicit) (1.21.6)\n","Requirement already satisfied: scipy>=0.16 in /usr/local/lib/python3.7/dist-packages (from implicit) (1.7.3)\n","Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.7/dist-packages (from implicit) (4.64.1)\n","Installing collected packages: implicit\n","Successfully installed implicit-0.6.1\n"]}],"source":["  # Библиотека с API для рекомендательных моделей\n","!pip install implicit"]},{"cell_type":"code","execution_count":null,"metadata":{"executionInfo":{"elapsed":1732,"status":"ok","timestamp":1663074055555,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"xCPgrUJEEgaP","colab":{"base_uri":"https://localhost:8080/"},"outputId":"eaad42be-6124-4882-abda-5062061c2c8a"},"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/implicit/gpu/__init__.py:14: UserWarning: CUDA extension is built, but disabling GPU support because of 'Cuda Error: no CUDA-capable device is detected (/project/./implicit/gpu/utils.h:71)'\n","  f\"CUDA extension is built, but disabling GPU support because of '{e}'\",\n"]}],"source":["import pandas as pd\n","import numpy as np\n","from scipy.sparse import csr_matrix\n","import scipy.sparse as sparse\n","from sklearn.model_selection import train_test_split\n","from tqdm.notebook import tqdm\n","\n","# Сохранение и загрузка моделей\n","import pickle\n","\n","# Графики\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","\n","# Библиотека для рекомендаций Implicit\n","from implicit.als import AlternatingLeastSquares\n","from implicit.bpr import BayesianPersonalizedRanking\n"]},{"cell_type":"markdown","metadata":{"id":"baCp74M-WNxy"},"source":["Немного модернизируем решение, предложенное в Baseline.\n","\n","- В исходную функцию для подсчета метрик добавим **количество угаданных книг**\n","\n","- В функцию перевода датафрейма в csr-матрицу добавим `словари`, чтобы не использовать внешние переменные"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Xr1uX1EL6ap0"},"outputs":[],"source":["###############################################\n","# Исходные функции для модели с рекомендациями\n","##############################################\n","\n","def contest_metric(df_solution, df_grd, print_report=True):\n","\n","  \"\"\"\n","  Возвращает метрики Recall, Precision, F1-score и количество угаданных книг  \n","  \"\"\"\n","\n","  pred = set(df_solution['chb'] + '_' + df_solution['sys_numb'].values)\n","  true = set(df_grd['chb'] + '_' + df_grd['sys_numb'].values) \n","  recall = len(pred.intersection(true)) / len(true)\n","  precision = len(pred.intersection(true)) / (20 * len(df_grd['chb'].unique()))\n","  \n","  f1_score = 2 * (precision * recall) / (precision + recall)\n","  true_pairs_count = len(pred.intersection(true))\n","\n","  if print_report == True:\n","    print(f\"Угаданных книг: {true_pairs_count}\")\n","    print(f\"Recall: {round(recall, 5)}\")\n","    print(f\"Precision: {round(precision, 5)}\")\n","    print(f\"F1-score: {round(f1_score, 5)}\")  \n","\n","  return [true_pairs_count, recall, precision, f1_score]\n","\n","\n","def df_to_sparse(df, user_index_dict, item_index_dict):\n","  \"\"\"\n","  Создает sparse-матрицу из датафрейма\n","\n","  Данные о пользователях в столбце 'chb', данные о книгах - в столбце 'sys_numb`\n","  Внешние словари используются для получения ID каждого пользователя и книги\n","  \"\"\"\n","  row = []\n","  col = []\n","  data = []\n","\n","  for line in df.itertuples():\n","    row.append(user_index_dict[line.chb])\n","    col.append(item_index_dict[line.sys_numb])\n","    data.append(1)\n","  \n","  return csr_matrix((data, (row, col)))\n","\n","\n","##################################################################\n","# Подготовка данных для обучения\n","##################################################################\n","\n","def recs_split_random(df, test_size):\n","  \"\"\"\n","  Возвращает train и test, разделенные c помощью np.random\n","  \"\"\"\n","\n","  # Используем np.random для рандомизации индексов\n","  df['random'] = np.random.random(size=len(df))\n","\n","  # Создаем маски индексов для train и test датасетов\n","  # test_size - размер валидационной выборки\n","  train_mask = df['random'] <  (1 - test_size)\n","  test_mask = df['random'] >= (1 - test_size)\n","  \n","  # Разделяем датасет на train и test, используя маски индексов\n","  train = df[train_mask]\n","  test = df[test_mask]\n","  return train, test\n","\n","\n","def split_and_prepare_recs_data(data, user_col, item_col, test_size,\n","                                weight_col='items', \n","                                print_report=True,\n","                                delete_test_duplicates=True):\n","  \n","  \"\"\"\n","  Разделяет на выборки, готовит данные для рекомендательных моделей \n","  и выводит количество неизвестных user и item в тесте\n","\n","  Возвращает train, test, полный датасет и веса\n","  Если в датасете только уникальные пары user-item, то все веса = 1\n","  \"\"\"\n","\n","  train_data, test_data = recs_split_random(data, test_size)\n","  all_data = data\n","  \n","  # Если создаем матрицу для implicit feedback (0, 1)\n","  if weight_col == 'items':\n","    # Отбираем столбцы с данными о пользователях и объектов рекомендации\n","    # Добавляем данные c частотой каждой пары user-item в поле 'items'\n","    train_data = (train_data.groupby([user_col, item_col])\n","                  .size()\n","                  .to_frame('items')\n","                  .reset_index())\n","    \n","    all_data = (data.groupby([user_col, item_col])\n","                  .size()\n","                  .to_frame('items')\n","                  .reset_index()) \n","    \n","  # Удаляем дубликаты в тестовых данных\n","  if delete_test_duplicates == True:    \n","      test_data = (test_data.groupby([user_col, item_col])\n","                    .size()\n","                    .to_frame('items')\n","                    .reset_index())  \n","\n","  # Нормализуем данные в поле weight_col\n","  # С параметром 'items' получим столбец из единиц\n","  train_weight = np.log2(train_data[weight_col] + 1)  \n","  all_weight = np.log2(all_data[weight_col] + 1)\n"," \n","\n","  # Отдельные датасеты для пар user-item \n","  train_data = train_data[[user_col, item_col]]\n","  test_data = test_data[[user_col, item_col]]\n","  all_data = all_data[[user_col, item_col]]\n","\n","  # Cписки уникальных пользователей для train и test\n","  train_users = np.sort(train_data[user_col].unique())\n","  test_users = np.sort(test_data[user_col].unique())\n","\n","  # Списки уникальных items для train и test\n","  train_items= np.sort(train_data[item_col].unique())\n","  test_items= np.sort(test_data[item_col].unique())\n","\n","  # Новые users и items, незнакомые для модели\n","  # они не попадут в обучение\n","  cold_start_items = set(test_items) - set(train_items)\n","  cold_start_users = set(test_users) - set(train_users)\n","\n","  if print_report:\n","    print(\"=> Количество пользователей\")\n","    print(f\"Общий датасет: {len(unique_transactions[user_col].unique())}\")\n","    print(f\"Обучающая выборка: {len(train_users)}\")\n","    print(f\"Валидационная выборка: {len(test_users)}\")\n","    print()    \n","    print(\"=> Проблема cold start для модели\")\n","    print(\"Неизвестных пользователей:\", len(cold_start_users))\n","    print(\"Неизвестных объектов:\", len(cold_start_items))\n","  \n","  # Набор датасетов для обучения рекомендательных моделей\n","  recs_datasets = {\n","      'train_user_item': train_data,\n","      'test_user_item': test_data,\n","      'all_user_item': all_data,\n","      'train_weight': train_weight,\n","      # 'test_weight': test_weight,\n","      'all_weight': all_weight\n","  }\n","  return recs_datasets\n","\n","def get_user_item_matrix(user_items_df, user_col, item_col, \n","                         user_item_weight_list):\n","\n","  \"\"\"Генерирует СSR-матрицу user-items в формате индексов для обучения моделей\n","  \n","  Возвращает:\n","  -------\n","  csr-матрица user-items с индексами\n","  словари для обратного перевода индексов user и item в исходные названия\n","  \"\"\"\n","  \n","  # Cловари для маппинга индексов ID -> читатель/книга\n","  index_to_user = pd.Series(np.sort(np.unique(user_items_df[user_col])))\n","  index_to_item = pd.Series(np.sort(np.unique(user_items_df[item_col])))\n","\n","  # Словари для обратного маппинга\n","  user_to_index = pd.Series(data=index_to_user.index, \n","                            index=index_to_user.values)\n","  item_to_index = pd.Series(data=index_to_item.index, \n","                            index=index_to_item.values)\n","\n","  # Добавляем индексы в обучающий датасет\n","  user_items_df_indexed = user_items_df.copy()\n","  user_items_df_indexed['user_id'] = user_items_df[user_col].map(user_to_index)\n","  user_items_df_indexed['item_id'] = user_items_df[item_col].map(item_to_index)\n","\n","  # Данные для перевода датасета в CSR матрицу\n","  data = user_item_weight_list\n","  rows = user_items_df_indexed['user_id']\n","  cols = user_items_df_indexed['item_id']\n","\n","  # CSR матрица user(строки)-books(столбцы)\n","  user_items_csr = csr_matrix((data, (rows, cols)), \n","                              shape=(len(user_to_index), len(item_to_index)))\n","\n","  return user_items_csr, index_to_user, index_to_item\n","\n","\n","########################################################################\n","# Получение рекомендаций с помощью моделей implicit\n","########################################################################\n","\n","def get_implicit_recommendations(trained_model_imp, user_items_matrix, \n","                                 index_to_user_map, index_to_item_map, \n","                                 top_n=20, filter_liked_items=True):  \n","  \"\"\"\n","  Генерирует top_n рекомендаций с обученной моделью библиотеки implicit\n","\n","  Возвращает:\n","  ----------\n","  Датафрейм, где каждая строка - список top_n рекомендаций для пользователя\n","  индексы - названия users, \n","  данные - названия items из словарей index_to_user_map, index_to_item_map\n","  \"\"\"\n","\n","  # Получаем список рекомендаций - матрица с индексами книг и читателей\n","  recs_matrix = trained_model_imp.recommend_all(\n","      user_items=user_items_matrix, \n","      N=top_n, \n","      filter_already_liked_items=filter_liked_items\n","  )\n","\n","  # Создаем датафрейм\n","  # Переводим индексы в исходные названия user и item\n","  recs_df = (\n","      pd.DataFrame(recs_matrix, index=index_to_user_map)\n","      .apply(lambda c: c.map(index_to_item_map))\n","  )\n","\n","  return recs_df\n","\n","\n","def get_avg_recs_metrics(recs_matrix, valid_user_items_df,\n","                               user_col, item_col, \n","                               print_report=True, round_n=4):\n","  \"\"\"\n","  Считает и выводит усредненные метрики для матрицы рекомендаций\n","\n","  Параметры:\n","  ---------\n","  recs_matrix - матрица предсказаний, строка = список рекомендаций для пользователя\n","  valid_user_items_df - валидационный датасет \n","  \"\"\"\n","  # Создаем словарь из валидационного датасета\n","  # ключ = user_name, значения = item_name\n","  user_items_dict = (\n","      valid_user_items_df\n","      .groupby(user_col)[item_col]\n","      .apply(set).to_dict()\n","  )\n","  \n","  # Ищем пересечение множества пользователей, на которых была обучена модель\n","  # с пользователями в валидационном датасете\n","  train_users = recs_matrix.index.to_list()\n","  valid_users = valid_user_items_df[user_col].unique()\n","  users_to_compare = set(train_users) & set(valid_users)\n","\n","  # Считаем усредненные метрики по всем пользователям\n","  # hit-rate, precision, recall и f1-score\n","  hitrate = round(\n","      np.mean([int(len(set(recs_matrix.loc[u]) & user_items_dict[u]) > 0) for u in users_to_compare]), round_n)\n","\n","  precision = round(\n","      np.mean([\n","      len(set(recs_matrix.loc[u]) & user_items_dict[u]) / \n","      len(recs_matrix.loc[u]) for u in users_to_compare]), 5)\n","\n","  recall = round(\n","      np.mean([\n","      len(set(recs_matrix.loc[u]) & user_items_dict[u]) / \n","      len(user_items_dict[u]) for u in users_to_compare]), 5)\n","\n","  f1_score = round(\n","      2 * (precision * recall) / (precision + recall), 5)\n","\n","\n","  if print_report:\n","    print(f\"Hit-Rate_avg: {hitrate}\")\n","    print(f\"Recall_avg: {precision}\")\n","    print(f\"Precision_avg: {recall}\")\n","    print(f\"F1-score_avg: {f1_score}\")\n","    \n","  return [hitrate, precision, recall, f1_score]\n","\n","\n","def matrix_recs_to_pairs(recs_matrix_df, \n","                         user_col_name, item_col_name):\n","  \"\"\"\n","  Переводит матрицу рекомендаций в датафрейм с парами user-item\n","  \n","  \n","  Параметры:\n","  --------\n","  recs_matrix_df - каждая строка [u]- список рекомендованных \n","                   items для пользователя [u]  \n","  user_col_name - название столбца для user\n","  item_col_name - название столбца для item\n","\n","  Возвращает датафрейм с двумя столбцами.\n","  \"\"\"\n","  # Переводим 2D набор items в 1D список\n","  pairs_df= pd.DataFrame(recs_matrix_df.values.ravel(), \n","                         columns=[item_col_name])\n","  \n","  # Количество столбцов = количество рекомендаций\n","  top_n = recs_matrix_df.shape[1]\n","\n","  # Добавляем индексы пользователей по порядку\n","  # одинаковый индекс для каждых top_n строк\n","  pairs_df[user_col_name] = pairs_df.index  // top_n\n","\n","  \n","  # Маппинг индексов в имена пользователей\n","  # используем матрицу рекомендаций \n","  pairs_df[user_col_name] = pairs_df[user_col_name].map(\n","      (recs_matrix_df.reset_index()\n","       .rename(columns={'index':user_col_name})[user_col_name].to_dict())\n","  )\n","\n","  return pairs_df\n","\n"]},{"cell_type":"markdown","metadata":{"id":"w33Da8I_TMCc"},"source":["## Загрузка файлов"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":20843,"status":"ok","timestamp":1663074076390,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"XTFHRtl68d40","outputId":"f7860c92-7af1-44ad-c718-de7d79eb09e8"},"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive\n"]}],"source":["# Доступ к файлам на colab\n","import glob\n","import os\n","\n","# Библиотека для работы с google drive\n","from google.colab import drive\n","\n","# Монтируем диск\n","drive.mount('/content/drive')\n","\n","# Смотрим файлы в директории\n","# !ls \"/content/drive/My Drive/Colab Notebooks\""]},{"cell_type":"code","execution_count":null,"metadata":{"id":"7lfTaeUtQncv"},"outputs":[],"source":["# Меняем домашнюю директорию\n","# для доступа к датасетам\n","os.chdir(\"/content/drive/My Drive/Colab Notebooks\")"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"8mfp-NaFD6KW"},"outputs":[],"source":["# Cчитывание данных\n","\n","# читатели\n","users = pd.read_csv(\n","    './users.csv', sep=';', index_col=None, \n","     dtype={'age': str, 'chb': str, 'chit_type': str, 'gender': str})\n","\n","# книги\n","items = pd.read_csv(\n","    'items.csv', sep=';', index_col=None, \n","     dtype={'author': str, 'bbk': str, 'izd': str, \n","           'sys_numb': str, 'title': str, 'year_izd': str})\n","\n","# история книг в библиотеке\n","transactions = pd.read_csv(\n","    'train_transactions_extended.csv', \n","    sep=';', index_col=None,\n","    dtype={'chb': str, 'date_1': str, 'is_printed': str, \n","            'is_real': str, 'source': str, 'sys_numb': str, 'type': str})"]},{"cell_type":"markdown","metadata":{"id":"_GgFB5ueMdb6"},"source":["## ЭТАП 3. Возможности Collaborative Filtering"]},{"cell_type":"markdown","metadata":{"id":"qX9GAiF-kF5c"},"source":["### (3.1) CF для активного пула читателей и книг"]},{"cell_type":"markdown","metadata":{"id":"8UPostUc1ull"},"source":["#### Вводные\n","\n","\n"]},{"cell_type":"markdown","metadata":{"id":"alkSbllD12jz"},"source":["`Collaborative Filtering` рассчитан на \"активную\" аудиторию:\n","  - рекомендации генерируются с помощью  информации о других пользователях и книгах.\n","\n","У нас **новая система**, и она только начала пополняться данными об активности читателей и их предпочтениях:\n","\n","  - Именно поэтому мы не можем получить `высокое качество` предсказаний\n","\n","  - Если взять читателей и книги только из **\"активного\" пула**, покажет ли CF более высокие результаты?\n","\n","Тогда для остальных пользователей мы сможем подобрать другой подход к рекомендациям и в итоге предложить **гибридное** решение:)"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"7v0OgvW-Pgi4"},"outputs":[],"source":["# Удаляем дубликаты\n","unique_transactions = transactions[['chb', 'sys_numb']].drop_duplicates()"]},{"cell_type":"markdown","metadata":{"id":"hrMeH5bDkOQv"},"source":["#### Самые активные пользователи\n","\n","\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":318},"executionInfo":{"elapsed":10,"status":"ok","timestamp":1663057866690,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"tEgDVJuklTTs","outputId":"dc205382-9ca3-40a6-8bb5-2243be6d050a"},"outputs":[{"output_type":"stream","name":"stdout","text":["Количество уникальных книг, взятых каждым пользователем:\n"]},{"output_type":"execute_result","data":{"text/plain":["              books_count\n","chb                      \n","100000641403            3\n","100000644359           11\n","100000665127           17\n","100000676191            3\n","100000679200            6\n","100000681262           34\n","100000683677            2"],"text/html":["\n","  <div id=\"df-84da7a73-1ef8-448a-bb53-b196eb505f18\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>books_count</th>\n","    </tr>\n","    <tr>\n","      <th>chb</th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>100000641403</th>\n","      <td>3</td>\n","    </tr>\n","    <tr>\n","      <th>100000644359</th>\n","      <td>11</td>\n","    </tr>\n","    <tr>\n","      <th>100000665127</th>\n","      <td>17</td>\n","    </tr>\n","    <tr>\n","      <th>100000676191</th>\n","      <td>3</td>\n","    </tr>\n","    <tr>\n","      <th>100000679200</th>\n","      <td>6</td>\n","    </tr>\n","    <tr>\n","      <th>100000681262</th>\n","      <td>34</td>\n","    </tr>\n","    <tr>\n","      <th>100000683677</th>\n","      <td>2</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-84da7a73-1ef8-448a-bb53-b196eb505f18')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-84da7a73-1ef8-448a-bb53-b196eb505f18 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-84da7a73-1ef8-448a-bb53-b196eb505f18');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":36}],"source":["# Создаем новый датафрейм с активностью пользователей\n","user_books_count = (\n","     unique_transactions.groupby('chb')\n","    .count().rename(columns={'sys_numb':'books_count'}))\n","\n","print(\"Количество уникальных книг, взятых каждым пользователем:\")\n","user_books_count.head(7)"]},{"cell_type":"markdown","metadata":{"id":"XKQ_qaACn4NQ"},"source":["- Вспомним, какая активность у большинства наших *читателей*"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":300},"executionInfo":{"elapsed":469,"status":"ok","timestamp":1663057873375,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"w14kjUlelTqs","outputId":"db38d6ff-e73e-4c19-c353-9a58ac2b69f9"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["        books_count\n","count  16753.000000\n","mean      13.669313\n","std       70.121205\n","min        2.000000\n","25%        3.000000\n","50%        6.000000\n","75%       12.000000\n","max     6671.000000"],"text/html":["\n","  <div id=\"df-06e45b40-e151-45b7-94a0-46d45b83de5d\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>books_count</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>16753.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>13.669313</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>70.121205</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>2.000000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>3.000000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>6.000000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>12.000000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>6671.000000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-06e45b40-e151-45b7-94a0-46d45b83de5d')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-06e45b40-e151-45b7-94a0-46d45b83de5d button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-06e45b40-e151-45b7-94a0-46d45b83de5d');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":37}],"source":["user_books_count.describe()"]},{"cell_type":"markdown","metadata":{"id":"2_7ttWalqDGY"},"source":["1. 50% пользователей прочитали меньше 6 книг (среди них могут быть и новички, и те, кто в целом редко читает книги)\n","\n","  - за нижнюю границу возьмем 6 книг\n","\n","2. 25% пользователей взяли больше 12 книг, но среди них есть и рекордсмены с тысячами\n","\n","  - попробуем найти границу для рекордсменов, которые скачивают все подряд)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":2,"status":"ok","timestamp":1663057874469,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"O3SFyhUolT5B","outputId":"8a8fa3e4-719d-4c17-a65e-07865c3dfb8d"},"outputs":[{"output_type":"stream","name":"stdout","text":["99% пользователей прочитали <  111 книг\n","90% пользователей прочитали <  25 книг\n"]}],"source":["print(\"99% пользователей прочитали < \",\n","      f\"{int(user_books_count.books_count.quantile(0.99))} книг\")\n","\n","print(\"90% пользователей прочитали < \",\n","      f\"{int(user_books_count.books_count.quantile(0.90))} книг\")"]},{"cell_type":"markdown","metadata":{"id":"ZIFDxkf1smlQ"},"source":["- Попробуем взять верхнюю границу в `111 книг`"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":410},"executionInfo":{"elapsed":439,"status":"ok","timestamp":1663057876575,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"YFH58vKVlUGa","outputId":"21ae7e65-1114-4091-df6e-3feb6790677b"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 576x432 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["filter_111 = 'books_count > 6 and books_count < 111'\n","\n","plt.figure(figsize=(8, 6))\n","user_books_count.query(filter_111).hist(ax=plt.gca())\n","plt.title(\n","    label=\"Пул активных пользователей - \" + f\"{user_books_count.query(filter_111).shape[0]} читателей\", \n","    size=16)\n","plt.xlabel(\"Количество взятых книг (> 6 и < 111)\", fontsize=14)\n","plt.ylabel(\"Количество читателей\", fontsize=14)\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"ATnYsplYs-pL"},"source":["**Вывод**\n","\n","1. Мы отобрали пользователей, которые прочитали от 6 до 111 книг (~половина)\n","\n","  - попробуем обучить модель с рекомендациями на этом пуле\n","\n","2. Мало книг не значит, что читатель неактивный, возможно, он только недавно пришел в библиотеку\n","\n","  - пока нам это неважно, важно количество книг для качественных предсказаний\n","\n","  - во время сегментации пользователей учтем этот нюанс!)"]},{"cell_type":"markdown","metadata":{"id":"OSKqhUxMU0en"},"source":["#### Книги, которые читают"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":300},"executionInfo":{"elapsed":759,"status":"ok","timestamp":1663057888039,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"vtXq6xwIFcNq","outputId":"ec03e45a-15c2-436d-dfb8-7c7035db7f5f"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["         users_count\n","count  194666.000000\n","mean        1.176384\n","std         0.578743\n","min         1.000000\n","25%         1.000000\n","50%         1.000000\n","75%         1.000000\n","max        21.000000"],"text/html":["\n","  <div id=\"df-b5426f53-b53d-4dba-baf5-8021f733237d\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>users_count</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>194666.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>1.176384</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>0.578743</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>1.000000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>1.000000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>1.000000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>1.000000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>21.000000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b5426f53-b53d-4dba-baf5-8021f733237d')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-b5426f53-b53d-4dba-baf5-8021f733237d button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-b5426f53-b53d-4dba-baf5-8021f733237d');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":40}],"source":["unique_transactions = transactions[['chb', 'sys_numb']].drop_duplicates()\n","\n","# Создаем новый датафрейм с количеством  читателей\n","# для каждой книги\n","books_popularity = (\n","    unique_transactions\n","    .groupby('sys_numb')\n","    .chb.count().to_frame()\n","    .rename(columns={'chb':'users_count'}))\n","\n","books_popularity.describe()"]},{"cell_type":"markdown","metadata":{"id":"MBgShTDfI-p4"},"source":["- Большинство книг читали только один раз, мы уже делали этот вывод в исследовании\n","\n","- Посмотрим, сколько у нас останется книг в датасете, если их отфильтровать"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":410},"executionInfo":{"elapsed":554,"status":"ok","timestamp":1663057890370,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"BVGmvEN4JQES","outputId":"c62efd0b-d82c-4064-f75f-7dd5358cafdd"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 576x432 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["plt.figure(figsize=(8, 6))\n","books_popularity.query('users_count > 1').hist(ax=plt.gca())\n","plt.title(\n","    label=\"Книги, которые нашли несколько читателей\", \n","    size=16)\n","plt.xlabel(\"Количество читателей\", fontsize=14)\n","plt.ylabel(\"Количество книг\", fontsize=14)\n","plt.show();"]},{"cell_type":"markdown","metadata":{"id":"8y6Vk5Q7KIk_"},"source":["**Вывод**\n","\n","1. Пул \"активных\" книг ~12% от общего датасета с транзациями:\n","\n"," - ~25 000 книг из 194 000 имеют больше 1-го читателя\n","\n","2. По идее, это минимальный набор книг, которые имеет смысл включать в обучение с подходом Collaborative Filtering."]},{"cell_type":"markdown","metadata":{"id":"H83Z-lY1U_yq"},"source":["#### Эксперименты с моделью"]},{"cell_type":"markdown","metadata":{"id":"jU-IvOP6sAFK"},"source":["##### Код для экспериментов"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"-K4SYSktsLlc"},"outputs":[],"source":["############################################################################\n","# Функции для экспериментов\n","############################################################################\n","\n","def implicit_model_experiment(model_imp, dataset):\n","  \n","  \"\"\"Обучает рекомендательную модель и возвращает результаты на тесте\"\"\"\n","\n","  # Словарь с предобработанными датасетами для обучения\n","  recs_datasets = split_and_prepare_recs_data(dataset,\n","                                              'chb', 'sys_numb', \n","                                              test_size=0.2, \n","                                              print_report=False)\n","\n","  # Формируем матрицы и словари с маппингом\n","  user_books_csr, users_dict, books_dict= get_user_item_matrix(\n","      recs_datasets['train_user_item'], \n","      'chb', 'sys_numb', \n","      recs_datasets['train_weight'])                                                       \n","\n","  # Обучаем модель   \n","  model_imp.fit((user_books_csr * 40).astype('double'), show_progress=True)\n","\n","  # Получаем список top-20 рекомендаций для каждого читателя\n","  recs_top_20 = get_implicit_recommendations(model_imp, \n","                                          user_books_csr,\n","                                          users_dict, books_dict,\n","                                          filter_liked_items=True)                                 \n","  \n","  # Считаем усредненные показатели модели\n","  results = get_avg_recs_metrics(\n","      recs_top_20, \n","      recs_datasets['test_user_item'],\n","      'chb', 'sys_numb', \n","      print_report=False)\n","  \n","  return results\n"]},{"cell_type":"markdown","metadata":{"id":"dgBs7jwGtnIg"},"source":["##### Фильтрация книг - как меняется датасет?"]},{"cell_type":"markdown","metadata":{"id":"780wSamu-__M"},"source":["Посмотрим, как будет меняться набор пользователей и книг в датасете с разной степенью отсева \"непопулярных\" книг"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":144},"executionInfo":{"elapsed":775,"status":"ok","timestamp":1663058154697,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"h0CbTy63J9Tx","outputId":"c3fa3fe8-f7f9-468a-ecd8-395480237637"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["               users > 0  users > 1  users > 2  users > 3  users > 4  \\\n","кол-во книг       194666      24597       5813       1997        861   \n","кол-во users       16753      12547       7682       4703       2907   \n","books_count >          2          2          2          2          2   \n","\n","               users > 5  \n","кол-во книг          429  \n","кол-во users        1829  \n","books_count >          2  "],"text/html":["\n","  <div id=\"df-3880b916-8563-4e09-8f1d-decb84f3d2d6\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>users &gt; 0</th>\n","      <th>users &gt; 1</th>\n","      <th>users &gt; 2</th>\n","      <th>users &gt; 3</th>\n","      <th>users &gt; 4</th>\n","      <th>users &gt; 5</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>кол-во книг</th>\n","      <td>194666</td>\n","      <td>24597</td>\n","      <td>5813</td>\n","      <td>1997</td>\n","      <td>861</td>\n","      <td>429</td>\n","    </tr>\n","    <tr>\n","      <th>кол-во users</th>\n","      <td>16753</td>\n","      <td>12547</td>\n","      <td>7682</td>\n","      <td>4703</td>\n","      <td>2907</td>\n","      <td>1829</td>\n","    </tr>\n","    <tr>\n","      <th>books_count &gt;</th>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3880b916-8563-4e09-8f1d-decb84f3d2d6')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-3880b916-8563-4e09-8f1d-decb84f3d2d6 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-3880b916-8563-4e09-8f1d-decb84f3d2d6');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":47}],"source":["book_filters_df = pd.DataFrame(index=['кол-во книг', 'кол-во users', 'books_count >'])\n","\n","for min_users in [0, 1, 2, 3, 4, 5]:\n","\n","  # Задаем фильтр для книг\n","  # по min количеству пользователей, которые их читали\n","  filter_books = 'users_count > @min_users'\n","   \n","  # Фильтруем датасет\n","  dataset = (unique_transactions\n","             .loc[lambda x: x.sys_numb.isin(\n","                 books_popularity.query(filter_books).index)]\n","             ) \n"," \n","  book_filters_df[f\"users > {min_users}\"] = [\n","      # количество уникальных пользователей и книг\n","      dataset.sys_numb.nunique(),\n","      dataset.chb.nunique(),\n","\n","      # минимальное количество книг в истории пользователей\n","      user_books_count[user_books_count.index.isin(dataset.chb)].books_count.min()\n","  ]\n","\n","book_filters_df"]},{"cell_type":"markdown","metadata":{"id":"SQHcP_2FhOOV"},"source":["**Вывод**:\n","\n","1. Самый большой `отсев книг ~ в 7 раз`:\n","  - если оставим книги, у которых есть хотя бы несколько читателей (users > 1)\n","2. В датасете останется `50% пользователей`:\n","  - если оставим книги, у которых уже больше 2-х читателей (users > 2)\n","\n","Посмотрим, как такие выборки отразятся на качестве модели!"]},{"cell_type":"markdown","metadata":{"id":"6wFeDJ4yt6LQ"},"source":["##### Модель для \"активных\" книг"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"rpNQsSp2ugEx"},"outputs":[],"source":["# Датафрейм для результатов эксперимента\n","results_df = pd.DataFrame(index=['Hit-Rate_avg', 'Recall_avg', 'Precision_avg', 'F1-score_avg'])\n","\n","# Создаем модель BRP\n","model_bpr = BayesianPersonalizedRanking(\n","    factors=200, learning_rate=0.3,  \n","    iterations=500,\n","    regularization=0.01)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":335,"referenced_widgets":["f216f87e1d6543c6b8f6be9fd31548d1","550aa8ff89a145d696563cd6e3aebd01","a87f5f67dd3b4863a818b2913cf1a9d7","77c3c6ebae7444f7bcd22e698942a5e5","4aca4516143b4478989a27b9658449e1","004d3630614f44c39bb06709a98a9ea6","93121f2cf43a4fce9be594d114dfcfcb","6c56d363ef8240d09dddc2ac408ecda0","424784ce47434cc08605a1d8098dfd9a","dd034d45b3c54a0988967d6e036b8031","551a6a59c9824f129fe104d434506fd1","f2e6f9809d5647958a069043e125f0e0","7538f441cd474c209b44698cbb584869","8793c78889454339b845cd41afe1958b","fd96142026814e2389f226d1401f75a4","88fe8519247a4b36af28d8afd56f43e8","fc7ae99971394a0ebeb3d8ace49bfcc3","e1e6d626578644bbb2fa9b6961cae4d1","5411bd2d7abd4c3c9f12b5952365d677","27eec663955b40e2980e60500f810d8b","405804a98729480694569712dfa8b5a8","8d37dd178f814380ad460e89ad34d27b","b2a8cf3a7e1e4b2bab42c1016dcba651","4dccdd4c2a224ee6a05b2966b5370b4d","1997e91222a747fd94f2e027c936aeec","83915256a5a4444fa7eec1882d549743","c7fea9886a76480690ed1b420eb99ced","96d7def756cf45a6914ab84fafe06d45","74bcb9b9c4f74c03b1eb792cd3ca8e55","3546206a61da4de9bc34a710c81e688f","de6ae8914b824dda968e103aef2fb12f","046f2f7bdc604c509cd35046fe1c4d90","ced7dacb1d644485880fd9de17bc5946","6f53e5b8ccb04934ac4aa271e6814b3e","89d8bba95fa1424ebd5de8c2b0823f53","dfc358cdf19944ab9284003faf50a4f9","48f72d2951f24c12a5f4d5cb0b55bd00","b6246f2dc9ab416187b7a887a0ec644c","e7ce64781bb44c169c04ab20b80d3488","420839ad85a447c2aeb97857a404aa99","0d70d7e0e07645a4b779addd6f198bac","dc461efabf9345ff880372430894d9bb","722493f48cda4f4aba7cda26608ef9fb","25ded155fa4d4437921ece0a67efeacf","b2175aa5b41545e0a0548f80c943e765","0fabdd6ed6b24617a3c04588957dde21","bea3b418152d4a638834731e47037a82","bf99d386bcab4e089f3f80776f0d7cc8","537e2acdbe28445c905615884ab1f566","a70faaff7d7842b79bbd8436be1c3c18","63b8aaecfa0e41d7b269cf54433efb2d","73d8fde024b64890a279ba6deb08525b","ff8726818c674deba469fb649cbce022","639858ddd1bd4262a50e9da87c1f4f3c","155cdde2a04248c1bcdf0f2111895802"]},"executionInfo":{"elapsed":221106,"status":"ok","timestamp":1663058380222,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"t1r4dAOTJXR_","outputId":"e15ccfc0-64bb-461f-ae9a-58fcf8a576f0"},"outputs":[{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/500 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"f216f87e1d6543c6b8f6be9fd31548d1"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/500 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"f2e6f9809d5647958a069043e125f0e0"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/500 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"b2a8cf3a7e1e4b2bab42c1016dcba651"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/500 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"6f53e5b8ccb04934ac4aa271e6814b3e"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/500 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"b2175aa5b41545e0a0548f80c943e765"}},"metadata":{}},{"output_type":"execute_result","data":{"text/plain":["               books_cnt > 0  books_cnt > 1  books_cnt > 2  books_cnt > 3  \\\n","Hit-Rate_avg         0.02250        0.09950        0.14420        0.18790   \n","Recall_avg           0.00127        0.00625        0.00864        0.01186   \n","Precision_avg        0.00817        0.05826        0.10203        0.14555   \n","F1-score_avg         0.00220        0.01129        0.01593        0.02193   \n","\n","               books_cnt > 4  \n","Hit-Rate_avg         0.23000  \n","Recall_avg           0.01540  \n","Precision_avg        0.18228  \n","F1-score_avg         0.02840  "],"text/html":["\n","  <div id=\"df-6e49eca2-451e-4357-b71c-2e6689692557\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>books_cnt &gt; 0</th>\n","      <th>books_cnt &gt; 1</th>\n","      <th>books_cnt &gt; 2</th>\n","      <th>books_cnt &gt; 3</th>\n","      <th>books_cnt &gt; 4</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>Hit-Rate_avg</th>\n","      <td>0.02250</td>\n","      <td>0.09950</td>\n","      <td>0.14420</td>\n","      <td>0.18790</td>\n","      <td>0.23000</td>\n","    </tr>\n","    <tr>\n","      <th>Recall_avg</th>\n","      <td>0.00127</td>\n","      <td>0.00625</td>\n","      <td>0.00864</td>\n","      <td>0.01186</td>\n","      <td>0.01540</td>\n","    </tr>\n","    <tr>\n","      <th>Precision_avg</th>\n","      <td>0.00817</td>\n","      <td>0.05826</td>\n","      <td>0.10203</td>\n","      <td>0.14555</td>\n","      <td>0.18228</td>\n","    </tr>\n","    <tr>\n","      <th>F1-score_avg</th>\n","      <td>0.00220</td>\n","      <td>0.01129</td>\n","      <td>0.01593</td>\n","      <td>0.02193</td>\n","      <td>0.02840</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6e49eca2-451e-4357-b71c-2e6689692557')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-6e49eca2-451e-4357-b71c-2e6689692557 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-6e49eca2-451e-4357-b71c-2e6689692557');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":49}],"source":["# Обучаем модель на пулах книг с разной \"активностью\"\n","for min_users in [0, 1, 2, 3, 4]:\n","\n","  # Задаем фильтр для книг\n","  # по min количеству пользователей, которые их читали\n","  filter_books = 'users_count > @min_users'\n","   \n","  # Фильтруем датасет\n","  filtered_dataset = (unique_transactions\n","                      .loc[lambda x: x.sys_numb.isin(\n","                          books_popularity.query(filter_books).index)]).copy()\n","  \n","  # Обучаем модель\n","  new_result = implicit_model_experiment(model_bpr, filtered_dataset)\n","  \n","  # Сохраняем результат в общей таблице\n","  results_df[f'books_cnt > {min_users}'] = new_result\n","\n","# Выводим результаты эксперимента\n","results_df"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"IZIHlUhXwTI8","colab":{"base_uri":"https://localhost:8080/","height":144},"executionInfo":{"status":"ok","timestamp":1663060296084,"user_tz":-180,"elapsed":289,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"}},"outputId":"bdb87dd0-f7e0-4d17-fbf7-a240e6401c86"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["               users > 0  users > 1  users > 2  users > 3  users > 4  \\\n","кол-во книг       194666      24597       5813       1997        861   \n","кол-во users       16753      12547       7682       4703       2907   \n","books_count >          2          2          2          2          2   \n","\n","               users > 5  \n","кол-во книг          429  \n","кол-во users        1829  \n","books_count >          2  "],"text/html":["\n","  <div id=\"df-c243dc83-e9c4-4ec2-9b4a-6678dbbb0865\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>users &gt; 0</th>\n","      <th>users &gt; 1</th>\n","      <th>users &gt; 2</th>\n","      <th>users &gt; 3</th>\n","      <th>users &gt; 4</th>\n","      <th>users &gt; 5</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>кол-во книг</th>\n","      <td>194666</td>\n","      <td>24597</td>\n","      <td>5813</td>\n","      <td>1997</td>\n","      <td>861</td>\n","      <td>429</td>\n","    </tr>\n","    <tr>\n","      <th>кол-во users</th>\n","      <td>16753</td>\n","      <td>12547</td>\n","      <td>7682</td>\n","      <td>4703</td>\n","      <td>2907</td>\n","      <td>1829</td>\n","    </tr>\n","    <tr>\n","      <th>books_count &gt;</th>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","      <td>2</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c243dc83-e9c4-4ec2-9b4a-6678dbbb0865')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-c243dc83-e9c4-4ec2-9b4a-6678dbbb0865 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-c243dc83-e9c4-4ec2-9b4a-6678dbbb0865');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":51}],"source":["book_filters_df"]},{"cell_type":"markdown","metadata":{"id":"iyxvzv0w9NMW"},"source":["**Вывод**\n","\n","1. Наша 1-я гипотеза подтвердилась! \n","\n","  - Если фокусироваться на активном пуле книг, то метод Сolloborative Filtering дает более высокие результаты\n","\n","2. Метод `BPR` дает хорошее качество на менее \"разреженных\" матрицах и очень быстро обучается.\n","\n","3. Можно попробовать отсеять книги с `users_count > 2` => уже получим 25% Hit Rate.\n"]},{"cell_type":"markdown","metadata":{"id":"AhpArqly0CXS"},"source":["##### Модель для \"активных пользователей\""]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206,"referenced_widgets":["01a50f309a074618a7ce9200c8d7b83b","3f3bc3b6a60a410a8fba3f89ab3a6700","01a85be2f26f42449e501058ab967881","3c94349ee03a4630aefcdbd7aa9e5659","5d5ad7d9c2464e86ab6285740894ce0b","3311475c6c2c41fb85c794ac73e2f0f4","9dd75561063c4463ab0730dc0b1dad05","3e73e067d1b448258b3b9d6fbe0589bf","91aef3d5282f46a38e0f78d1634a17b0","c6a6a6d624d242398b953853fc350220","40c54eedd0c5481dbb9e3a08d431317f"]},"executionInfo":{"elapsed":94616,"status":"ok","timestamp":1663060422082,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"zQPWphUa768_","outputId":"371fdad6-aa75-4b5c-d065-4c9ffb9848ca"},"outputs":[{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/500 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"01a50f309a074618a7ce9200c8d7b83b"}},"metadata":{}},{"output_type":"execute_result","data":{"text/plain":["               books_cnt > 0  books_cnt > 1  books_cnt > 2  books_cnt > 3  \\\n","Hit-Rate_avg         0.02250        0.02100        0.14420        0.18790   \n","Recall_avg           0.00127        0.00114        0.00864        0.01186   \n","Precision_avg        0.00817        0.00827        0.10203        0.14555   \n","F1-score_avg         0.00220        0.00200        0.01593        0.02193   \n","\n","               books_cnt > 4  \n","Hit-Rate_avg         0.23000  \n","Recall_avg           0.01540  \n","Precision_avg        0.18228  \n","F1-score_avg         0.02840  "],"text/html":["\n","  <div id=\"df-a7ce60a6-e08c-4c81-bedf-066302ffae41\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>books_cnt &gt; 0</th>\n","      <th>books_cnt &gt; 1</th>\n","      <th>books_cnt &gt; 2</th>\n","      <th>books_cnt &gt; 3</th>\n","      <th>books_cnt &gt; 4</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>Hit-Rate_avg</th>\n","      <td>0.02250</td>\n","      <td>0.02100</td>\n","      <td>0.14420</td>\n","      <td>0.18790</td>\n","      <td>0.23000</td>\n","    </tr>\n","    <tr>\n","      <th>Recall_avg</th>\n","      <td>0.00127</td>\n","      <td>0.00114</td>\n","      <td>0.00864</td>\n","      <td>0.01186</td>\n","      <td>0.01540</td>\n","    </tr>\n","    <tr>\n","      <th>Precision_avg</th>\n","      <td>0.00817</td>\n","      <td>0.00827</td>\n","      <td>0.10203</td>\n","      <td>0.14555</td>\n","      <td>0.18228</td>\n","    </tr>\n","    <tr>\n","      <th>F1-score_avg</th>\n","      <td>0.00220</td>\n","      <td>0.00200</td>\n","      <td>0.01593</td>\n","      <td>0.02193</td>\n","      <td>0.02840</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a7ce60a6-e08c-4c81-bedf-066302ffae41')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-a7ce60a6-e08c-4c81-bedf-066302ffae41 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-a7ce60a6-e08c-4c81-bedf-066302ffae41');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":52}],"source":["# Обучаем модель на группах пользователей с разной \"активностью\"\n","for min_books in [1]:\n","\n","  # Задаем фильтр для книг\n","  # по min количеству пользователей, которые их читали\n","  filter_users = 'books_count > @min_books and books_count < 111'\n","   \n","  # Фильтруем датасет\n","  filtered_dataset = (unique_transactions\n","                      .loc[lambda x: x.chb.isin(\n","                          user_books_count.query(filter_users).index)]).copy()\n","  \n","  # Обучаем модель\n","  new_result = implicit_model_experiment(model_bpr, filtered_dataset)\n","  \n","  # Сохраняем результат в общей таблице\n","  results_df[f'books_cnt > {min_books}'] = new_result\n","\n","# Выводим результаты эксперимента\n","results_df"]},{"cell_type":"markdown","metadata":{"id":"b3_ggQzldhpx"},"source":["### (3.2) Дублирующиеся книги - баг или фича?"]},{"cell_type":"markdown","metadata":{"id":"crQ2D2KiotJ9"},"source":["#### Вклад каждой книги в историю читателя"]},{"cell_type":"markdown","metadata":{"id":"wQuDJ7oiejOe"},"source":["В процессе подготовки данных мы фиксировали частоту появления (веса) каждой книги для каждого пользователя: \n","\n","  - Такой подход был предложен разработчиками RankFM при тестировании их моделей в сравнении с ALS в Implicit.\n","\n","  - Это актуально, например, если бы мы исследовали покупку товаров)\n","\n","Так как мы преварительно удаляли все дубликаты user-book, то веса были равны 1.\n","\n","Интересно, а как выглядела бы картина, если бы мы оставили исходные данные?"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":473},"executionInfo":{"elapsed":2221,"status":"ok","timestamp":1663060514080,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"zhhBQtSlhOmR","outputId":"4bf95fa9-ba16-4fc0-f0d7-4a67921108c0"},"outputs":[{"output_type":"stream","name":"stdout","text":["Книги, которые перечитывали много раз:\n"]},{"output_type":"execute_result","data":{"text/plain":["                         chb  read_times  \\\n","sys_numb                                   \n","RSL07000449201  300001090427          44   \n","RSL01004357029  200000987642          42   \n","RSL01010175347  300000850004          35   \n","RSL01006724299  200000957291          35   \n","RSL01002745675  300000850004          26   \n","...                      ...         ...   \n","RSL01004063667  100001136730           1   \n","RSL01004063954  100001123223           1   \n","RSL01004063987  100001136730           1   \n","RSL01004064132  200001143192           1   \n","RSL60000349909  200000844116           1   \n","\n","                                                            title  \n","sys_numb                                                           \n","RSL07000449201                                        отсутствует  \n","RSL01004357029  Переписка В. А. Жуковского и А. П. Елагиной, 1...  \n","RSL01010175347  Система органов государственной власти в Респу...  \n","RSL01006724299  Коллекционеры России, XVII - начало XX вв. : э...  \n","RSL01002745675  Правопорядок в условиях формирования правового...  \n","...                                                           ...  \n","RSL01004063667  Пациенты диспансеров, страдающие шизофренией с...  \n","RSL01004063954  Провоспалительные и противовоспалительные цито...  \n","RSL01004063987  Фармакогностическое изучение сухоцвета однолет...  \n","RSL01004064132  Оптимизация санитарно-эпидемиологического надз...  \n","RSL60000349909                                        отсутствует  \n","\n","[229002 rows x 3 columns]"],"text/html":["\n","  <div id=\"df-23031f22-336d-4f5d-b28e-d1e5f8634995\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>chb</th>\n","      <th>read_times</th>\n","      <th>title</th>\n","    </tr>\n","    <tr>\n","      <th>sys_numb</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>RSL07000449201</th>\n","      <td>300001090427</td>\n","      <td>44</td>\n","      <td>отсутствует</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004357029</th>\n","      <td>200000987642</td>\n","      <td>42</td>\n","      <td>Переписка В. А. Жуковского и А. П. Елагиной, 1...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01010175347</th>\n","      <td>300000850004</td>\n","      <td>35</td>\n","      <td>Система органов государственной власти в Респу...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01006724299</th>\n","      <td>200000957291</td>\n","      <td>35</td>\n","      <td>Коллекционеры России, XVII - начало XX вв. : э...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002745675</th>\n","      <td>300000850004</td>\n","      <td>26</td>\n","      <td>Правопорядок в условиях формирования правового...</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004063667</th>\n","      <td>100001136730</td>\n","      <td>1</td>\n","      <td>Пациенты диспансеров, страдающие шизофренией с...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004063954</th>\n","      <td>100001123223</td>\n","      <td>1</td>\n","      <td>Провоспалительные и противовоспалительные цито...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004063987</th>\n","      <td>100001136730</td>\n","      <td>1</td>\n","      <td>Фармакогностическое изучение сухоцвета однолет...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004064132</th>\n","      <td>200001143192</td>\n","      <td>1</td>\n","      <td>Оптимизация санитарно-эпидемиологического надз...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL60000349909</th>\n","      <td>200000844116</td>\n","      <td>1</td>\n","      <td>отсутствует</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>229002 rows × 3 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-23031f22-336d-4f5d-b28e-d1e5f8634995')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-23031f22-336d-4f5d-b28e-d1e5f8634995 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-23031f22-336d-4f5d-b28e-d1e5f8634995');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":53}],"source":["print(\"Книги, которые перечитывали много раз:\")\n","books_read_twice = (transactions\n","                    .groupby(['chb', 'sys_numb'])\n","                    .size()\n","                    .to_frame('read_times')\n","                    .reset_index())\n","\n","(books_read_twice\n"," .set_index('sys_numb')\n"," .join(items.set_index('sys_numb')['title'])\n"," .sort_values(by='read_times', ascending=False))"]},{"cell_type":"markdown","metadata":{"id":"BQQPyxGxmwIZ"},"source":["- Какой контраст! Никто не перечитывал оптимизацию санитарно санитарно-эпидемиологического надзора и историю шизовфреников. Это радует:)"]},{"cell_type":"markdown","metadata":{"id":"su0W7GIFjm4X"},"source":["**Вывод**\n","\n","1. У нас есть пул читателей, которые заказывали одну и ту же книгу по несколько раз:\n","\n","  - возможно, заканчивался срок возврата книги\n","  - это любимые книги читателя\n","  - или какой-то сбой при загрузке книг\n","\n","\n","2. В прошлой имплементации удаление дубликатов снижало качество рекомендаций\n","\n","  - Можно попробовать провести обучение еще раз, но в этот раз с нормализацией\n"]},{"cell_type":"markdown","metadata":{"id":"O2MLwV5roJav"},"source":["#### Нормализация весов и обучение модели"]},{"cell_type":"markdown","metadata":{"id":"9dwzhAjXoSHT"},"source":["Мы уже включили нормализацию в набор готовых функций, посмотрим как она работает:"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":255},"executionInfo":{"elapsed":299,"status":"ok","timestamp":1663060522315,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"UTu5JmtEejOe","outputId":"e999a766-9576-4d94-afaf-1a19a0519572"},"outputs":[{"output_type":"stream","name":"stdout","text":["Показатели после нормализации с np.log2:\n"]},{"output_type":"display_data","data":{"text/plain":["213726    5.491853\n","86257     5.426265\n","80884     5.169925\n","144763    5.169925\n","144698    4.754888\n","            ...   \n","79217     1.000000\n","79218     1.000000\n","79219     1.000000\n","79220     1.000000\n","229001    1.000000\n","Name: read_times, Length: 229002, dtype: float64"]},"metadata":{}}],"source":["print(\"Показатели после нормализации с np.log2:\")\n","\n","display(np.log2(books_read_twice['read_times'] + 1).sort_values(ascending=False))"]},{"cell_type":"markdown","metadata":{"id":"xgijqyIYqIKf"},"source":[]},{"cell_type":"markdown","metadata":{"id":"Ud0T9rrcpVuK"},"source":["- Мы значительно уменьшили вес \"часто\" читаемых книг с 40 до 5 \n","\n","- Теперь они не будут перевешивать значимость других книжек, которые еще никто ниразу не взял повторно\n","\n","Построим модель! Может быть теперь мы получим более высокое качество?"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":2011,"status":"ok","timestamp":1663060528966,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"xINCePw0qO5D","outputId":"2697c30c-ce14-4a3b-e665-5f5bac3e6ad6"},"outputs":[{"output_type":"stream","name":"stdout","text":["=> Количество пользователей\n","Общий датасет: 16753\n","Обучающая выборка: 16666\n","Валидационная выборка: 12373\n","\n","=> Проблема cold start для модели\n","Неизвестных пользователей: 87\n","Неизвестных объектов: 32125\n"]}],"source":["# Словарь с предобработанными датасетами для обучения\n","# Дубликаты не удаляем\n","recs_dict = split_and_prepare_recs_data(transactions, \n","                                        'chb', 'sys_numb', 0.2)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":701,"status":"ok","timestamp":1663060532492,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"aOnABMsPq0eW","outputId":"6ace2344-89a6-4b4c-8264-4b8fc48d0250"},"outputs":[{"output_type":"stream","name":"stdout","text":["Размер sparse-матрицы user-books: (16666, 162541)\n"]}],"source":["# Формируем матрицы и словари с маппингом\n","user_books_csr, users_dict, books_dict= get_user_item_matrix(recs_dict['train_user_item'], \n","                                                             'chb', 'sys_numb', \n","                                                             recs_dict['train_weight'])                                                       \n","\n","print(\"Размер sparse-матрицы user-books:\", user_books_csr.shape)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":105,"referenced_widgets":["04928ddd4ac24b5392223d3501ed46ca","6d18324b5865442192ad606b4cab110e","d9928b4873b344729aafcbdcd5ad5a93","b14ff5a3028d4b0bb963582b547cafa6","b1583bd219cd4c16b3ea8629dfb792bd","0deb722ae8ea474a92fca24d9afbec08","df04df190fe64103a5cef6796a1ffa1d","e01ad7cce5414efd8d9864aff5774a60","1a9da9d624ae4dbdaf8faaac31198275","2d5e92206b6d41eea4202aebb8721dd3","c46dd76b5dda4e7a9e2b85328d691fbb"]},"executionInfo":{"elapsed":325053,"status":"ok","timestamp":1663060857544,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"RKENnmEcq0eX","outputId":"addcaa04-52ae-4125-e140-25c6d7eff8f5"},"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/implicit/utils.py:29: UserWarning: OpenBLAS detected. Its highly recommend to set the environment variable 'export OPENBLAS_NUM_THREADS=1' to disable its internal multithreading\n","  \"OpenBLAS detected. Its highly recommend to set the environment variable \"\n"]},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/15 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"04928ddd4ac24b5392223d3501ed46ca"}},"metadata":{}}],"source":["alpha_value = 40\n","\n","# Обучаем модель\n","imp_model = AlternatingLeastSquares(factors=200, random_state=777)\n","imp_model.fit((user_books_csr * alpha_value).astype('double'))"]},{"cell_type":"markdown","metadata":{"id":"ZOlCoMTBvSR1"},"source":["#### Оценка результатов"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":35077,"status":"ok","timestamp":1663060916712,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"eTYhYLJSq0eX","outputId":"e4c4bbe4-a7b7-47e4-9e8a-aa921d46886b"},"outputs":[{"output_type":"stream","name":"stdout","text":["Hit-Rate_avg: 0.0285\n","Recall_avg: 0.0016\n","Precision_avg: 0.01051\n","F1-score_avg: 0.00278\n"]},{"output_type":"execute_result","data":{"text/plain":["[0.0285, 0.0016, 0.01051, 0.00278]"]},"metadata":{},"execution_count":58}],"source":["# Получаем список top-20 рекомендаций для каждого читателя\n","recs_als_imp = get_implicit_recommendations(imp_model, \n","                                            user_books_csr,\n","                                            users_dict, books_dict,\n","                                            filter_liked_items=True)                                 \n","\n","# Выводим усредненные показатели модели\n","get_avg_recs_metrics(recs_als_imp, recs_dict['test_user_item'],\n","                           'chb', 'sys_numb')"]},{"cell_type":"markdown","metadata":{"id":"lhMDQa2vzGJn"},"source":["А если мы оставим все книги, которые попали в TOP-20 (параметр `filter_liked_items=False`)?\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":32240,"status":"ok","timestamp":1663060948941,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"C5JtnBEaq0eY","outputId":"796715c1-e15b-4e43-d169-0807e2bbb993"},"outputs":[{"output_type":"stream","name":"stdout","text":["Hit-Rate_avg: 0.256\n","Recall_avg: 0.02139\n","Precision_avg: 0.1115\n","F1-score_avg: 0.03589\n"]},{"output_type":"execute_result","data":{"text/plain":["[0.256, 0.02139, 0.1115, 0.03589]"]},"metadata":{},"execution_count":59}],"source":["# Получаем список top-20 рекомендаций для каждого читателя\n","recs_als_imp = get_implicit_recommendations(imp_model, \n","                                            user_books_csr,\n","                                            users_dict, books_dict,\n","                                            filter_liked_items=False)                                 \n","\n","# Выводим усредненные показатели модели\n","get_avg_recs_metrics(recs_als_imp, recs_dict['test_user_item'],\n","                           'chb', 'sys_numb')"]},{"cell_type":"markdown","metadata":{"id":"3aIAqua_zf_7"},"source":["**Вывод**\n","\n","- Ох как взлетело качество! Hit-Rate = 25% против 2%!\n","\n","- Все потому, что многие пользователи нашей библиотеки часто обращаются к одним и тем же книгам.\n","\n","\"+\" для хороших предсказаний, \"-\" для хороших рекомендаций =)"]},{"cell_type":"markdown","metadata":{"id":"mMzHDkTE-xxz"},"source":["#### Тест на платформе\n","\n"]},{"cell_type":"markdown","metadata":{"id":"Aq6kmXED--Oz"},"source":["Сейчас мы узнаем актуальны ли дубликаты для тестовой платформы.\n","\n","Данные у нас уже подготовлены, осталось обучить модель на полной выборке."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":427,"status":"ok","timestamp":1663060949356,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"AEWuJOsrAGip","outputId":"d21327f8-93a0-4dc9-e87c-f82ef7e1fe36"},"outputs":[{"output_type":"stream","name":"stdout","text":["Размер sparse-матрицы user-books: (16753, 194666)\n"]}],"source":["# Формируем матрицы и словари с маппингом\n","user_books_csr, users_dict, books_dict= get_user_item_matrix(recs_dict['all_user_item'], \n","                                                             'chb', 'sys_numb', \n","                                                             recs_dict['all_weight'])                                                       \n","\n","print(\"Размер sparse-матрицы user-books:\", user_books_csr.shape)"]},{"cell_type":"markdown","metadata":{"id":"3tlIq6LdAjGR"},"source":["- В полной выборке у нас добавилось ~40 000 позиций книг"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":49,"referenced_widgets":["ba19ba6d26804281859004131ea5c303","4a761abd0d4d4dec992b5259f2a8b8e4","54d2783412794142b07b72fd21b58c0c","20ccf552c6a34b5bbb4c846d2e52de5f","76932ce9ec914e21bdc47c72ceba5653","e77dc869ec4e441ab854e581e10d7bc1","7a4ac19e05a549f5a2e60ee46a30c3c2","11ff27baecef47b6ad978763913a54ae","9f8906ba962d4f9c852bf25db269b145","91dea137df8540a78a2243c5fefa485f","ad15c24ba0cf4bcbb8f2dfa9e21e4b44"]},"executionInfo":{"elapsed":235140,"status":"ok","timestamp":1663061184494,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"DpBcfUa1AGip","outputId":"355b9e31-3617-49be-b96d-3a0c10ff9ee2"},"outputs":[{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/15 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"ba19ba6d26804281859004131ea5c303"}},"metadata":{}}],"source":["alpha_value = 40\n","\n","# Обучаем модель\n","imp_model = AlternatingLeastSquares(factors=200, random_state=777)\n","imp_model.fit((user_books_csr * alpha_value).astype('double'))"]},{"cell_type":"markdown","metadata":{"id":"dr1hvyB_A5_1"},"source":["- Не забываем отключить фильтр `filter_liked_items`"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":35755,"status":"ok","timestamp":1663063178120,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"zR5B6Wt3AGiq","outputId":"fc1de222-f4db-44f9-c319-d1fea51c6165"},"outputs":[{"output_type":"stream","name":"stdout","text":["Сохранено 335060 предсказаний\n"]}],"source":["# Получаем список top-20 рекомендаций для каждого читателя\n","recs_als_imp = get_implicit_recommendations(imp_model, \n","                                            user_books_csr,\n","                                            users_dict, books_dict,\n","                                            filter_liked_items=False)      \n","    \n","# Переводим матрицу рекомендаций в формат пар user-book\n","solution = matrix_recs_to_pairs(recs_als_imp, 'chb', 'sys_numb') \n","\n","# Сохраняем для теста на платформе\n","solution.to_csv(\"predictions_als_imp_duplicates.csv\", index=False, sep=';')                   \n","\n","print(f\"Сохранено {len(solution)} предсказаний\")"]},{"cell_type":"markdown","metadata":{"id":"nzcT6NshBjfF"},"source":["**Вывод**\n","\n","1. Платформа выдала нам результат в 0.0028 \n","\n","  - делаем вывод, что дубликатов в тесте не так много)\n","\n","2. Когда мы выясним, откуда у пользователей по 1000 повторных прочтений книг\n","  - можно  использовать как  фичу \"книги, которые перечитывают\":\n","  - а пока для нас это только баг:)"]},{"cell_type":"markdown","metadata":{"id":"km5Txxb6Mx4A"},"source":["### (3.3) Рекомендации новых авторов"]},{"cell_type":"markdown","metadata":{"id":"ss6O8CLxX7_a"},"source":["#### Вводные"]},{"cell_type":"markdown","metadata":{"id":"f8R3xlBlXTgl"},"source":["Попробуем построить такую же систему рекомендаций на базе Collaborative Filtering, но для авторов:)\n","\n","- Сможем ли мы открыть для читателей новых авторов?\n","\n","- Могут ли авторы помочь в предсказании новых книг?"]},{"cell_type":"markdown","metadata":{"id":"Par5y4ezYbNa"},"source":["#### Добавляем авторов в датасет\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":243},"executionInfo":{"elapsed":1759,"status":"ok","timestamp":1663063552844,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"blJwmf_xYfPN","outputId":"9297bb94-0792-4929-d3b6-8ba62a1885f0"},"outputs":[{"output_type":"stream","name":"stdout","text":["размер датасета до объединения: 229002\n","после добавления авторов: 229002\n"]},{"output_type":"execute_result","data":{"text/plain":["         sys_numb           chb                             author\n","0  RSL01000000123  100000934918          Боронов, Адина Айылчиевич\n","1  RSL01000000244  300001043470        Дедов, Александр Викторович\n","2  RSL01000000259  300001090044  Добротворская, Карина Анатольевна\n","3  RSL01000000457  100000728030            Летягин, Лев Николаевич\n","4  RSL01000000462  200001091160         Литвин, Алексей Васильевич"],"text/html":["\n","  <div id=\"df-115153af-08e6-4c30-b97b-b1dee9673785\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sys_numb</th>\n","      <th>chb</th>\n","      <th>author</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>RSL01000000123</td>\n","      <td>100000934918</td>\n","      <td>Боронов, Адина Айылчиевич</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>RSL01000000244</td>\n","      <td>300001043470</td>\n","      <td>Дедов, Александр Викторович</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>RSL01000000259</td>\n","      <td>300001090044</td>\n","      <td>Добротворская, Карина Анатольевна</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>RSL01000000457</td>\n","      <td>100000728030</td>\n","      <td>Летягин, Лев Николаевич</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>RSL01000000462</td>\n","      <td>200001091160</td>\n","      <td>Литвин, Алексей Васильевич</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-115153af-08e6-4c30-b97b-b1dee9673785')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-115153af-08e6-4c30-b97b-b1dee9673785 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-115153af-08e6-4c30-b97b-b1dee9673785');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":67}],"source":["unique_transactions = transactions[['chb', 'sys_numb']].drop_duplicates()\n","print(\"размер датасета до объединения:\", unique_transactions.shape[0])\n","\n","# Создаем временные индексы для объединения items и transactions\n","# Добавляем в transactions авторов книг\n","unique_transactions = (unique_transactions\n","                       .set_index('sys_numb')\n","                       .join(items.set_index('sys_numb')['author'])\n","                       .reset_index()\n","                      )\n","\n","print(\"после добавления авторов:\", unique_transactions.shape[0])\n","\n","unique_transactions.head(5)               "]},{"cell_type":"markdown","metadata":{"id":"eSbFdwT3goaT"},"source":["- Сколько новых имен!)"]},{"cell_type":"markdown","metadata":{"id":"_6GyJBe0jHEz"},"source":["#### Популярные и любимые авторы"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":290,"status":"ok","timestamp":1663063566649,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"pwa603lW-m00","outputId":"85b47f5a-5f36-42f5-a341-90b0e152777c"},"outputs":[{"output_type":"stream","name":"stdout","text":["Разнообразие авторов в библиотеке: 172622\n","Авторы, которых уже читают: 104748\n"]}],"source":["print(\"Разнообразие авторов в библиотеке:\", items.author.nunique())\n","print(\"Авторы, которых уже читают:\", unique_transactions.author.nunique())"]},{"cell_type":"markdown","metadata":{"id":"f1xD8y07hTuC"},"source":["- Половина авторов библиотеки уже известна читателям, классно!"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":602,"status":"ok","timestamp":1663063569075,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"gk47v5RDhblu","outputId":"c1030229-86cd-47fc-a0ea-cb24e0938e49"},"outputs":[{"output_type":"stream","name":"stdout","text":["Самые популярные авторы:\n","\n"]},{"output_type":"execute_result","data":{"text/plain":["author\n","Чуковский, Корней Иванович 1882-1969          42\n","Библия                                        43\n","Лосев, Алексей Федорович                      43\n","Россия. Государственный контроль              43\n","Лейкин, Николай Александрович 1841-1906       45\n","Романюк, Сергей Константинович                46\n","Юнг, Карл Густав 1875-1961                    46\n","Российская Федерация Законы                   47\n","Москва, город                                 51\n","Нестеров, Михаил Петрович 1959-               52\n","Пушкин, Александр Сергеевич 1799-1837         55\n","Сарабьянов, Дмитрий Владимирович              65\n","Нащокина, Мария Владимировна                  80\n","Серова, Марина Сергеевна                     100\n","Струнников, Сергей Николаевич 1907-1944      120\n","Зверев, Сергей Иванович 1964-                165\n","Россия. Законы и постановления               174\n","Конволют                                     587\n","отсутствует                                17267\n","none                                       44714\n","Name: chb, dtype: int64"]},"metadata":{},"execution_count":69}],"source":["print(\"Самые популярные авторы:\\n\")\n","unique_transactions.groupby('author').chb.count().sort_values().tail(20)"]},{"cell_type":"markdown","metadata":{"id":"4JbEMMujiJzr"},"source":["- Законы и постановления России в 3 раза популярнее чем А.С. Пушкин..Эхх\n","\n","- В топе детективы и боевики Сергея Зверева, интересно)\n","\n","У нас очень много книг неизвестных авторов `отсутствует` и `none`. Придется их удалить из выборки."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":716,"status":"ok","timestamp":1663063573827,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"TjtAz6iukuOB","outputId":"05a07b3c-0821-4055-b5f2-6b13932d0c98"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["author\n","Алексеев, Сергей Сергеевич 1924-2013        41\n","Чуковский, Корней Иванович 1882-1969        42\n","Шахов, Максим Анатольевич 1964-             42\n","Лосев, Алексей Федорович                    43\n","Россия. Государственный контроль            43\n","Библия                                      43\n","Лейкин, Николай Александрович 1841-1906     45\n","Юнг, Карл Густав 1875-1961                  46\n","Романюк, Сергей Константинович              46\n","Российская Федерация Законы                 47\n","Москва, город                               51\n","Нестеров, Михаил Петрович 1959-             52\n","Пушкин, Александр Сергеевич 1799-1837       55\n","Сарабьянов, Дмитрий Владимирович            65\n","Нащокина, Мария Владимировна                80\n","Серова, Марина Сергеевна                   100\n","Струнников, Сергей Николаевич 1907-1944    120\n","Зверев, Сергей Иванович 1964-              165\n","Россия. Законы и постановления             174\n","Конволют                                   587\n","Name: chb, dtype: int64"]},"metadata":{},"execution_count":70}],"source":["unique_transactions = unique_transactions[~unique_transactions.author.isin(['отсутствует', 'none'])]\n","unique_transactions.groupby('author').chb.count().sort_values().tail(20)"]},{"cell_type":"markdown","metadata":{"id":"_0LX7vXolRdC"},"source":["- Так то лучше!\n","\n","- Посмотрим, а есть ли у наших читателей любимые авторы?"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":550,"status":"ok","timestamp":1663063576024,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"6GfjSRYgArn_","outputId":"d1dd6799-b619-4d7b-c7b3-16a5cbb2def2"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["chb           author                                    \n","100000953820  Плеханов, Георгий Валентинович (1856-1918)     21\n","100001064348  Москва, город                                  21\n","300000841170  Лейкин, Николай Александрович 1841-1906        24\n","300001015451  Тамоников, Александр Александрович 1959-       24\n","100001091809  Мышкинское уезд. земское собрание              25\n","100000953820  Конволют                                       25\n","300001015451  Серегин, Михаил Георгиевич                     25\n","300001132797  Рославльское уездное земское собрание          25\n","300001015451  Шитов, Владимир Кузьмич                        28\n","              Казанцев, Кирилл                               29\n","100000836042  Человеколюбивое общество (Петербург)           29\n","300001030736  Тульское губ. земское собрание                 30\n","300001015451  Дышев, Андрей Михайлович                       36\n","              Устинова, Татьяна Витальевна 1968-             37\n","              Шахов, Максим Анатольевич 1964-                42\n","              Нестеров, Михаил Петрович 1959-                52\n","300000758346  Конволют                                       78\n","100000704313  Россия. Законы и постановления                 82\n","300001015451  Серова, Марина Сергеевна                       99\n","              Зверев, Сергей Иванович 1964-                 165\n","Name: sys_numb, dtype: int64"]},"metadata":{},"execution_count":71}],"source":["unique_transactions.groupby(['chb', 'author']).sys_numb.count().sort_values().tail(20)"]},{"cell_type":"markdown","metadata":{"id":"-arcMMiym29G"},"source":["1. Похоже, популярность авторам делают их поклонники:\n","\n","  - Помним, что у загадочного мужчины (300001015451) было ~6000 выдач книг.\n","\n","2. Многие читают пачками книги из одного собрания:)\n","\n","Интересно, какой набор книг скрывается под секретным \"Конволют\"? И читает ли загадочный 300000758346 что-то еще?"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":424},"executionInfo":{"elapsed":10,"status":"ok","timestamp":1663063576025,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"9Dei5IWrlwdq","outputId":"fe2228ca-c6bb-4b6a-a43d-0b985d1f6be0"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["              sys_numb           chb  \\\n","7455    RSL01000643335  300000758346   \n","29609   RSL01001839439  300000758346   \n","29666   RSL01001842091  300000758346   \n","30585   RSL01001873071  300000758346   \n","30586   RSL01001885398  300000758346   \n","...                ...           ...   \n","184443  RSL01009880936  300000758346   \n","184447  RSL01009880973  300000758346   \n","184449  RSL01009881128  300000758346   \n","184452  RSL01009881172  300000758346   \n","184457  RSL01009881252  300000758346   \n","\n","                                               author  \n","7455    Скворцов, Дмитрий Иванович 1861-не ранее 1907  \n","29609               Иоанн Кронштадтский св. 1829-1908  \n","29666                    Магнуссен, Владимир Петрович  \n","30585                           Берио, Шарль Огюст де  \n","30586                                  Богдашев, П.Н.  \n","...                                               ...  \n","184443                                       Конволют  \n","184447                                       Конволют  \n","184449                                       Конволют  \n","184452                                       Конволют  \n","184457                                       Конволют  \n","\n","[1265 rows x 3 columns]"],"text/html":["\n","  <div id=\"df-a2a40f47-986c-4cb2-b088-438ff2baff28\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sys_numb</th>\n","      <th>chb</th>\n","      <th>author</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>7455</th>\n","      <td>RSL01000643335</td>\n","      <td>300000758346</td>\n","      <td>Скворцов, Дмитрий Иванович 1861-не ранее 1907</td>\n","    </tr>\n","    <tr>\n","      <th>29609</th>\n","      <td>RSL01001839439</td>\n","      <td>300000758346</td>\n","      <td>Иоанн Кронштадтский св. 1829-1908</td>\n","    </tr>\n","    <tr>\n","      <th>29666</th>\n","      <td>RSL01001842091</td>\n","      <td>300000758346</td>\n","      <td>Магнуссен, Владимир Петрович</td>\n","    </tr>\n","    <tr>\n","      <th>30585</th>\n","      <td>RSL01001873071</td>\n","      <td>300000758346</td>\n","      <td>Берио, Шарль Огюст де</td>\n","    </tr>\n","    <tr>\n","      <th>30586</th>\n","      <td>RSL01001885398</td>\n","      <td>300000758346</td>\n","      <td>Богдашев, П.Н.</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>184443</th>\n","      <td>RSL01009880936</td>\n","      <td>300000758346</td>\n","      <td>Конволют</td>\n","    </tr>\n","    <tr>\n","      <th>184447</th>\n","      <td>RSL01009880973</td>\n","      <td>300000758346</td>\n","      <td>Конволют</td>\n","    </tr>\n","    <tr>\n","      <th>184449</th>\n","      <td>RSL01009881128</td>\n","      <td>300000758346</td>\n","      <td>Конволют</td>\n","    </tr>\n","    <tr>\n","      <th>184452</th>\n","      <td>RSL01009881172</td>\n","      <td>300000758346</td>\n","      <td>Конволют</td>\n","    </tr>\n","    <tr>\n","      <th>184457</th>\n","      <td>RSL01009881252</td>\n","      <td>300000758346</td>\n","      <td>Конволют</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>1265 rows × 3 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a2a40f47-986c-4cb2-b088-438ff2baff28')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-a2a40f47-986c-4cb2-b088-438ff2baff28 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-a2a40f47-986c-4cb2-b088-438ff2baff28');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":72}],"source":["unique_transactions.query('chb == \"300000758346\"')"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":424},"executionInfo":{"elapsed":10,"status":"ok","timestamp":1663063576026,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"uFAc0C0Zoarx","outputId":"43d0063e-f0f6-41c3-c089-289b536347ba"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["              sys_numb                                              title  \\\n","13      RSL01010972888  Русский язык в школе = Russian language at sch...   \n","70      RSL01010988967  Инновационная экономика и современный менеджме...   \n","788     RSL01002912116                                        : Поэма еды   \n","860     RSL01005711499  Описание японских рукописей, ксилографов и ста...   \n","2034    RSL01010984580  Вестник дополнительного образования : журнал :...   \n","...                ...                                                ...   \n","343139  RSL01003885385  Описание рукописей Церковно-археологического м...   \n","343140  RSL01003885383  Описание рукописей Церковно-археологического м...   \n","343494  RSL01003793198   : Настоятели Московского Большого Успенского ...   \n","348959  RSL01003561528            : О человеке и его отношениях к природе   \n","352948  RSL01009946619  Краткое начертание естественной истории животн...   \n","\n","          author                                  izd year_izd            bbk  \n","13      Конволют                                 none     2022           none  \n","70      Конволют                                 none     2022           none  \n","788     Конволют                                 none     1928  Ш5(2=Р)53-651  \n","860     Конволют                                 none     1963           none  \n","2034    Конволют                                 none     2022           none  \n","...          ...                                  ...      ...            ...  \n","343139  Конволют                                 none     1875           none  \n","343140  Конволют                                 none     1879           none  \n","343494  Конволют  Тип. Свято-Троицкой Сергиевой лавры     1912           none  \n","348959  Конволют                              [б. и.]     1847            Е,0  \n","352948  Конволют                                 none     1827           none  \n","\n","[937 rows x 6 columns]"],"text/html":["\n","  <div id=\"df-d26c8103-e9c9-4e7e-83ec-15a0ab266e0f\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sys_numb</th>\n","      <th>title</th>\n","      <th>author</th>\n","      <th>izd</th>\n","      <th>year_izd</th>\n","      <th>bbk</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>13</th>\n","      <td>RSL01010972888</td>\n","      <td>Русский язык в школе = Russian language at sch...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>2022</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>70</th>\n","      <td>RSL01010988967</td>\n","      <td>Инновационная экономика и современный менеджме...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>2022</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>788</th>\n","      <td>RSL01002912116</td>\n","      <td>: Поэма еды</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>1928</td>\n","      <td>Ш5(2=Р)53-651</td>\n","    </tr>\n","    <tr>\n","      <th>860</th>\n","      <td>RSL01005711499</td>\n","      <td>Описание японских рукописей, ксилографов и ста...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>1963</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>2034</th>\n","      <td>RSL01010984580</td>\n","      <td>Вестник дополнительного образования : журнал :...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>2022</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>343139</th>\n","      <td>RSL01003885385</td>\n","      <td>Описание рукописей Церковно-археологического м...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>1875</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>343140</th>\n","      <td>RSL01003885383</td>\n","      <td>Описание рукописей Церковно-археологического м...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>1879</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>343494</th>\n","      <td>RSL01003793198</td>\n","      <td>: Настоятели Московского Большого Успенского ...</td>\n","      <td>Конволют</td>\n","      <td>Тип. Свято-Троицкой Сергиевой лавры</td>\n","      <td>1912</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>348959</th>\n","      <td>RSL01003561528</td>\n","      <td>: О человеке и его отношениях к природе</td>\n","      <td>Конволют</td>\n","      <td>[б. и.]</td>\n","      <td>1847</td>\n","      <td>Е,0</td>\n","    </tr>\n","    <tr>\n","      <th>352948</th>\n","      <td>RSL01009946619</td>\n","      <td>Краткое начертание естественной истории животн...</td>\n","      <td>Конволют</td>\n","      <td>none</td>\n","      <td>1827</td>\n","      <td>none</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>937 rows × 6 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d26c8103-e9c9-4e7e-83ec-15a0ab266e0f')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-d26c8103-e9c9-4e7e-83ec-15a0ab266e0f button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-d26c8103-e9c9-4e7e-83ec-15a0ab266e0f');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":73}],"source":["items.query('author == \"Конволют\"')"]},{"cell_type":"markdown","metadata":{"id":"QMjbfvnkoqgm"},"source":["- Поэма еды!\n","\n","-<img src=\"https://rusneb.ru/local/tools/exalead/thumbnail.php?url=000199_000009_002912116&width=360&height=460\">\n"]},{"cell_type":"markdown","metadata":{"id":"_6uT93pgrAWV"},"source":["**Вывод**\n","\n","1. Было бы интересно научиться открывать для читателей новых авторов:)\n","\n","  - Ведь если понравится автор, это еще +30 любимых книг:) \n","\n","2. Новизна авторов - +1 фактор к оценке и тестированию рекомендательной системы\n"]},{"cell_type":"markdown","metadata":{"id":"_Gx8F6uUrAQ5"},"source":["#### Обучаем модель предсказаний"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":1333,"status":"ok","timestamp":1663067050703,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"bUTvuHJ6Ghzm","outputId":"0aebcab0-e326-4aa6-e039-053d2e521dc3"},"outputs":[{"output_type":"stream","name":"stdout","text":["=> Количество пользователей\n","Общий датасет: 16265\n","Обучающая выборка: 15968\n","Валидационная выборка: 8782\n","\n","=> Проблема cold start для модели\n","Неизвестных пользователей: 297\n","Неизвестных объектов: 12371\n"]}],"source":["# Словарь с предобработанными датасетами для обучения\n","recs_dict = split_and_prepare_recs_data(\n","    # Удаляем дубликаты\n","    unique_transactions[['chb', 'author']].drop_duplicates(), \n","    'chb', 'author', 0.15)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":67,"referenced_widgets":["a9dfac0af13140afb31fba779cde470b","d4988bbdb533474490cc6a60f6c1c027","a485092a06084c808ce1feb49366a209","06e5a66e8be54c33be93f4dd3d887b3c","e5f09affcfa14066b9ff0e184cc7e24f","cf2a9c6f0e17427781e76955166ee329","1e0c7d5af18d4cb6bda9091e8f729fc8","a8e16620671140989a5506f63b34fcca","4a31b50b4b364b97ab9b787573f16527","cd49916be24f4f65a2982f12e849bf5a","74b142ccebe24c16b5816bed42643c59"]},"executionInfo":{"elapsed":188211,"status":"ok","timestamp":1663067238913,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"Z_rvp0aEGhzm","outputId":"a01a594e-2b0d-4bdf-bb24-ff1c18512ec4"},"outputs":[{"output_type":"stream","name":"stdout","text":["Размер sparse-матрицы user-authors: (15968, 92375)\n"]},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/15 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"a9dfac0af13140afb31fba779cde470b"}},"metadata":{}}],"source":["# Формируем матрицы и словари с маппингом\n","user_authors_csr, users_dict, authors_dict= get_user_item_matrix(\n","    recs_dict['train_user_item'], \n","    'chb', 'author', \n","    recs_dict['train_weight'])                                                       \n","\n","print(\"Размер sparse-матрицы user-authors:\", user_authors_csr.shape)\n","\n","alpha_value = 40\n","\n","# Обучаем модель\n","imp_model = AlternatingLeastSquares(factors=200, random_state=777)\n","imp_model.fit((user_authors_csr * alpha_value).astype('double'))\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":22159,"status":"ok","timestamp":1663067261055,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"w-891LXmPk05","outputId":"0fd1f385-2cb7-41f1-9d1f-e129d7bdc3b9"},"outputs":[{"output_type":"stream","name":"stdout","text":["Hit-Rate_avg: 0.0511\n","Recall_avg: 0.00282\n","Precision_avg: 0.02611\n","F1-score_avg: 0.00509\n"]}],"source":["# Получаем список top-20 рекомендаций для каждого читателя\n","recs_als_imp = get_implicit_recommendations(imp_model, \n","                                            user_authors_csr,\n","                                            users_dict, authors_dict,\n","                                            filter_liked_items=True)                                 \n","\n","# Выводим усредненные показатели модели\n","_ = get_avg_recs_metrics(recs_als_imp, recs_dict['test_user_item'],\n","                         'chb', 'author')"]},{"cell_type":"markdown","metadata":{"id":"kDyla-lA2YgU"},"source":["#### Общий вывод\n","\n","Теперь мы можем использовать созданную cистему Collaborative Filtering для рекомендаций новых авторов!\n","\n","\n","1. Такой подход будет намного интереснее для \"активных\" читателей =)\n","\n","  - Если наша модель будет \"успешно\" рекомендовать книги уже известных авторов, то пользователям это быстро наскучит)) \n","\n","2. Мы получили примерно такое же качество модели, как и для книг, HitRate ~5% для всех данных\n","\n","3. Можно так же, как мы тестировали, сузить фокус:\n","\n","  - на половине читателей -> 15% Hit Rate\n","  - на самых активных читателях и получить 30% HitRate\n","\n","По мере развития библиотеки и нашей системы, будет появляться все больше данных для успешного применения Collaborative Filtering.\n","\n","Для новых и не сильно активных пользователей (пока у нас таких большинство) нужен другой подход к рекомендациям!"]},{"cell_type":"markdown","metadata":{"id":"BdHBbcx0dP_q"},"source":["## ЭТАП 4. Интерпретация предсказаний"]},{"cell_type":"markdown","metadata":{"id":"Xxqd5-zLRXLe"},"source":["Выясним, для каких пользователей мы делаем больше хороших предсказаний, а для каких не можем угадать даже одну книгу:)\n","\n","В качестве **предсказаний** и реальных (**тестовых**) данных можно загрузить `любые результаты` экспериментов в блоке \"Загрузка предсказаний\""]},{"cell_type":"markdown","metadata":{"id":"gxZYkcA4gLgD"},"source":["### (4.1) Загрузка предсказаний"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"KR5NaH_RcbG-"},"outputs":[],"source":["# Загружаем ранее сохраненные предсказания и тестовую выборку\n","# all_users_predictions = pd.read_csv(\"recs_train_als_alpha_40.csv\", sep=';')"]},{"cell_type":"markdown","source":["Если сохраненных предсказаний с одного из экспериментов нет, обучим модель"],"metadata":{"id":"23CK8TXWchgZ"}},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":49,"referenced_widgets":["d17201ffb8b843619d6f737bd0c3c98a","dffece2752d74c439a244f776f1d9bb6","5216639b0ed542dcb6c7b05e0b403563","7897cb7e1de347cfa77b24b606b87226","10721637d4874205a1f75b2ddd459acd","06a8edc134af46c385c2ccb1346f3049","b27fb7519c364da2b46e40b5414eae38","a1bdef50d82a4f04ae9630f710932e0a","72d9a78270644827bfb8267e0a8a1af7","5d2ddf3224ec4318b793de72e1db78e3","4138ef5f27f840bc88b17a5196eefed6"]},"executionInfo":{"elapsed":364524,"status":"ok","timestamp":1663067634440,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"outputId":"c2cbf941-c651-4adb-e2c0-cddd5bedd1bb","id":"gskk87ZFcqW1"},"outputs":[{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/15 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"d17201ffb8b843619d6f737bd0c3c98a"}},"metadata":{}}],"source":["# Словарь с предобработанными датасетами для обучения\n","recs_dict = split_and_prepare_recs_data(\n","    transactions[['chb', 'sys_numb']].drop_duplicates(), \n","    'chb', 'sys_numb', 0.2,\n","    print_report=False)\n","\n","# Формируем матрицы и словари с маппингом\n","user_books_csr, users_dict, books_dict= get_user_item_matrix(\n","  recs_dict['train_user_item'], \n","  'chb', 'sys_numb', \n","  recs_dict['train_weight'])  \n","\n","alpha_value = 40\n","\n","# Обучаем модель\n","imp_model = AlternatingLeastSquares(factors=200, random_state=777)\n","imp_model.fit((user_books_csr * alpha_value).astype('double'))\n","\n","# Получаем список top-20 рекомендаций для каждого читателя\n","recs_als_imp = get_implicit_recommendations(imp_model, \n","                                            user_books_csr,\n","                                            users_dict, books_dict,\n","                                            filter_liked_items=True)    "]},{"cell_type":"code","source":["# Выводим усредненные показатели модели\n","_ = get_avg_recs_metrics(recs_als_imp, recs_dict['test_user_item'],\n","                         'chb', 'sys_numb')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"YKCG_MOFgziO","executionInfo":{"status":"ok","timestamp":1663067638961,"user_tz":-180,"elapsed":4533,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"}},"outputId":"e7b814bb-0e10-4a10-954d-4c8d14f1ff9a"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Hit-Rate_avg: 0.0305\n","Recall_avg: 0.00176\n","Precision_avg: 0.01214\n","F1-score_avg: 0.00307\n"]}]},{"cell_type":"code","source":["# Сохраняем предсказания в виде пар читатель-книга\n","all_users_predictions = matrix_recs_to_pairs(recs_als_imp, 'chb', 'sys_numb')\n","\n","# Тестовый датасет для оценки результатов\n","test_data = recs_dict['test_user_item']"],"metadata":{"id":"MZQqnX0TfJHM"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["Теперь посмотрим, какие сильные и слабые стороны у нашей модели!"],"metadata":{"id":"1jElJ_EjeoHA"}},{"cell_type":"markdown","metadata":{"id":"0biIxLfHhZkz"},"source":["### (4.2) Какие книги мы рекомендуем?"]},{"cell_type":"markdown","metadata":{"id":"Yn-1FrGsajE6"},"source":["Посмотрим предсказания, которые мы сгенерировали с помощью ALS метода \n"," - получили 0.00296 на валидационной выборке, 0.0058 на тесте платформы\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":455},"executionInfo":{"elapsed":297,"status":"ok","timestamp":1663067796842,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"BYNNWnSJgs9s","outputId":"9a03b746-4c87-4396-de5f-9310ee2d7ab7"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["                recs_pop\n","sys_numb                \n","RSL01000000123         1\n","RSL01005411964         1\n","RSL01005411652         1\n","RSL01005411090         1\n","RSL01005410671         1\n","...                  ...\n","RSL01002884697       573\n","RSL01003345484       606\n","RSL01000958257       681\n","RSL01005241876       991\n","RSL01008586079      1464\n","\n","[68009 rows x 1 columns]"],"text/html":["\n","  <div id=\"df-82ef8fc8-8c29-483f-9738-befee1adfcc9\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>recs_pop</th>\n","    </tr>\n","    <tr>\n","      <th>sys_numb</th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>RSL01000000123</th>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01005411964</th>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01005411652</th>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01005411090</th>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01005410671</th>\n","      <td>1</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002884697</th>\n","      <td>573</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003345484</th>\n","      <td>606</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01000958257</th>\n","      <td>681</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01005241876</th>\n","      <td>991</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01008586079</th>\n","      <td>1464</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>68009 rows × 1 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-82ef8fc8-8c29-483f-9738-befee1adfcc9')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-82ef8fc8-8c29-483f-9738-befee1adfcc9 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-82ef8fc8-8c29-483f-9738-befee1adfcc9');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":118}],"source":["# Для каждой книги\n","# считаем количество пользователей, которым ее порекомендовали\n","predicted_books = (\n","     all_users_predictions.groupby('sys_numb')\n","    .chb.count().sort_values().to_frame('recs_pop'))\n","\n","predicted_books"]},{"cell_type":"markdown","metadata":{"id":"XHUqyClQJ3Ey"},"source":["- Интересно! Есть книги, которые мы рекомендовали всего раз, а какие-то почти тысяче пользователей)\n","\n","- Посмотрим, что это за книги, и бьются ли они с самыми популярными в датасете)"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"bPcnkPWuKIbg"},"outputs":[],"source":["predicted_books = (\n","    predicted_books.join(books_popularity)\n","    .join(items.set_index('sys_numb')[['author', 'title', 'bbk']])\n",")"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":770},"executionInfo":{"elapsed":5,"status":"ok","timestamp":1663067801816,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"IYirFklIM_vW","outputId":"7a9f197d-13e1-4d8a-beba-31beae80012f"},"outputs":[{"output_type":"stream","name":"stdout","text":["Самые часто читаемые книги и их популярность в рекомендациях:\n"]},{"output_type":"execute_result","data":{"text/plain":["                recs_pop  users_count                            author  \\\n","sys_numb                                                                  \n","RSL01009819710       162           21                              none   \n","RSL01009824134       146           18                              none   \n","RSL01009824143       387           16                              none   \n","RSL01009819687       110           16                              none   \n","RSL01003187450       182           15                              none   \n","RSL01002890388       357           15                              none   \n","RSL01009819689       121           15                              none   \n","RSL01009817128       178           15                              none   \n","RSL01000741775       173           14  Сарабьянов, Дмитрий Владимирович   \n","RSL01009819690       154           14                              none   \n","RSL01009815656       172           14                              none   \n","RSL01002884697       573           14           Азимов, Айзек 1920-1992   \n","RSL01004271448       247           14     Злочевский, Гарольд Давидович   \n","RSL01009817125       124           13                              none   \n","RSL01006826042       121           13                              none   \n","RSL01009824080       116           13                              none   \n","RSL01004353863       182           13       Казусь, Игорь Александрович   \n","RSL01009824102       219           13                              none   \n","RSL01000669023       161           12      Нащокина, Мария Владимировна   \n","RSL01000570299       149           12                              none   \n","\n","                                                            title  \\\n","sys_numb                                                            \n","RSL01009819710      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01009824134     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009819687      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01003187450  Купеческая Москва : образы ушедшей российской ...   \n","RSL01002890388  Архитектура Москвы, 1920-1960 = Moscow archite...   \n","RSL01009819689      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01009817128      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01000741775                             Модерн : История стиля   \n","RSL01009819690      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01009815656     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01002884697  Слова на карте = Words on the map : геогр. наз...   \n","RSL01004271448  Старинные усадьбы и дачи Подмосковья : библиог...   \n","RSL01009817125     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01006826042  Москва : Памятники архитектуры XVIII - первой ...   \n","RSL01009824080      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01004353863  Советская архитектура 1920-х годов: организаци...   \n","RSL01009824102     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01000669023  Московский модерн : Проблема западноевроп. вли...   \n","RSL01000570299  Зодчие Москвы времени эклектики, модерна и нео...   \n","\n","                                                              bbk  \n","sys_numb                                                           \n","RSL01009819710                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824134                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009819687                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01003187450  Т3(2)5-285.9Москва.я43\\nТ3(2)5-752.39Москва.я4...  \n","RSL01002890388                              Н0(2Р=Р)92Москва07я23  \n","RSL01009819689                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009817128                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01000741775                            Щ103(0)5-239\"модерн\".я6  \n","RSL01009819690                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009815656                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01002884697                                             Д8я212  \n","RSL01004271448  Т3(2Р-4Мо)я1\\nТ3(2)0-752.29Московская область. я1  \n","RSL01009817125                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01006826042                                               none  \n","RSL01009824080                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01004353863                              Н0(2)71,0\\nН2л.г(2),0  \n","RSL01009824102                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01000669023              Н0(2Р=Р)92Москва05,0\\nН0(2Р=Р)5-239,0  \n","RSL01000570299               Н0(2Р=Р)92Москва.я21\\nН0(2Р=Р)5-8я21  "],"text/html":["\n","  <div id=\"df-2a24808b-24d8-4ae9-ae34-93ae8bff2906\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>recs_pop</th>\n","      <th>users_count</th>\n","      <th>author</th>\n","      <th>title</th>\n","      <th>bbk</th>\n","    </tr>\n","    <tr>\n","      <th>sys_numb</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>RSL01009819710</th>\n","      <td>162</td>\n","      <td>21</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824134</th>\n","      <td>146</td>\n","      <td>18</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009819687</th>\n","      <td>110</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003187450</th>\n","      <td>182</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>Купеческая Москва : образы ушедшей российской ...</td>\n","      <td>Т3(2)5-285.9Москва.я43\\nТ3(2)5-752.39Москва.я4...</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002890388</th>\n","      <td>357</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>Архитектура Москвы, 1920-1960 = Moscow archite...</td>\n","      <td>Н0(2Р=Р)92Москва07я23</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009819689</th>\n","      <td>121</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009817128</th>\n","      <td>178</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01000741775</th>\n","      <td>173</td>\n","      <td>14</td>\n","      <td>Сарабьянов, Дмитрий Владимирович</td>\n","      <td>Модерн : История стиля</td>\n","      <td>Щ103(0)5-239\"модерн\".я6</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009819690</th>\n","      <td>154</td>\n","      <td>14</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009815656</th>\n","      <td>172</td>\n","      <td>14</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002884697</th>\n","      <td>573</td>\n","      <td>14</td>\n","      <td>Азимов, Айзек 1920-1992</td>\n","      <td>Слова на карте = Words on the map : геогр. наз...</td>\n","      <td>Д8я212</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004271448</th>\n","      <td>247</td>\n","      <td>14</td>\n","      <td>Злочевский, Гарольд Давидович</td>\n","      <td>Старинные усадьбы и дачи Подмосковья : библиог...</td>\n","      <td>Т3(2Р-4Мо)я1\\nТ3(2)0-752.29Московская область. я1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009817125</th>\n","      <td>124</td>\n","      <td>13</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01006826042</th>\n","      <td>121</td>\n","      <td>13</td>\n","      <td>none</td>\n","      <td>Москва : Памятники архитектуры XVIII - первой ...</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824080</th>\n","      <td>116</td>\n","      <td>13</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004353863</th>\n","      <td>182</td>\n","      <td>13</td>\n","      <td>Казусь, Игорь Александрович</td>\n","      <td>Советская архитектура 1920-х годов: организаци...</td>\n","      <td>Н0(2)71,0\\nН2л.г(2),0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824102</th>\n","      <td>219</td>\n","      <td>13</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01000669023</th>\n","      <td>161</td>\n","      <td>12</td>\n","      <td>Нащокина, Мария Владимировна</td>\n","      <td>Московский модерн : Проблема западноевроп. вли...</td>\n","      <td>Н0(2Р=Р)92Москва05,0\\nН0(2Р=Р)5-239,0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01000570299</th>\n","      <td>149</td>\n","      <td>12</td>\n","      <td>none</td>\n","      <td>Зодчие Москвы времени эклектики, модерна и нео...</td>\n","      <td>Н0(2Р=Р)92Москва.я21\\nН0(2Р=Р)5-8я21</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2a24808b-24d8-4ae9-ae34-93ae8bff2906')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-2a24808b-24d8-4ae9-ae34-93ae8bff2906 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-2a24808b-24d8-4ae9-ae34-93ae8bff2906');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":120}],"source":["print(\"Самые часто читаемые книги и их популярность в рекомендациях:\")\n","predicted_books.sort_values(by='users_count', ascending=False).head(20)"]},{"cell_type":"markdown","metadata":{"id":"dly4aViHOD6B"},"source":["**Вывод 1**\n","\n","Популярные книги мы рекомендовали как минимум сотне человек\n","\n","-  у них не самый большой рейтинг среди рекомендаций Matrix Factorization:)\n","-  хорошо, ведь иначе абсолютно всем пришлось бы изучать открытки \"С Рождеством Христовым!\"\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":770},"executionInfo":{"elapsed":4,"status":"ok","timestamp":1663067804750,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"zB8q1CtFOgzN","outputId":"dd0697bf-1dae-49d6-d8fa-82d7bb11cf49"},"outputs":[{"output_type":"stream","name":"stdout","text":["Самые популярные книги в рекомендациях и количество их читателей:\n"]},{"output_type":"execute_result","data":{"text/plain":["                recs_pop  users_count                              author  \\\n","sys_numb                                                                    \n","RSL01008586079      1464            8       Карцхия, Александр Амиранович   \n","RSL01005241876       991           10             Карклин, Павел Иванович   \n","RSL01000958257       681           11        Хализев, Валентин Евгеньевич   \n","RSL01003345484       606            5      Рябинин, Юрий Валерьевич 1963-   \n","RSL01002884697       573           14             Азимов, Айзек 1920-1992   \n","RSL01002840579       541           11            Кузнецов, Сергей Львович   \n","RSL01001071360       520            7          Алексеев, Сергей Сергеевич   \n","RSL01004397358       518            9           Ахманова, Ольга Сергеевна   \n","RSL01003332475       491            6          Щукин, Анатолий Николаевич   \n","RSL01001848685       480            8                                none   \n","RSL01000704558       414            5             Шилов, Денис Николаевич   \n","RSL01009824143       387           16                                none   \n","RSL01003817085       383            9                                none   \n","RSL01008141826       381            3        Яковлева, Елена Вячеславовна   \n","RSL01004117881       378            6   Петров, Юрий Александрович, 1955-   \n","RSL01001847464       373            7                                none   \n","RSL01002890388       357           15                                none   \n","RSL01008066836       355            4             Белый, Андрей 1880-1934   \n","RSL01003015223       339            9          Панофский, Эрвин 1892-1968   \n","RSL01008656366       338            8  Выготский, Лев Семенович 1896-1934   \n","\n","                                                            title  \\\n","sys_numb                                                            \n","RSL01008586079  Гражданско-правовая модель регулирования цифро...   \n","RSL01005241876  Военная книга (1934-1939) : библиографический ...   \n","RSL01000958257      Теория литературы : Учеб. для студентов вузов   \n","RSL01003345484                                   Русское юродство   \n","RSL01002884697  Слова на карте = Words on the map : геогр. наз...   \n","RSL01002840579  Гистология, цитология и эмбриология : : учебни...   \n","RSL01001071360  Общая теория права : Курс в 2-х т. / С. С. Але...   \n","RSL01004397358  Словарь лингвистических терминов : лингвостили...   \n","RSL01003332475  Лингводидактический энциклопедический словарь ...   \n","RSL01001848685  Справочник по фондам Архива внешней политики Р...   \n","RSL01000704558  Государственные деятели Российской империи. Гл...   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01003817085  Российское Зарубежье во Франции, 1919-2000 = L...   \n","RSL01008141826  Повышение эффективности очистки производственн...   \n","RSL01004117881  Наемные управляющие в России : опыт бизнес-эли...   \n","RSL01001847464          Лингвистический энциклопедический словарь   \n","RSL01002890388  Архитектура Москвы, 1920-1960 = Moscow archite...   \n","RSL01008066836                                      Стихотворения   \n","RSL01003015223        Ренессанс и \"ренессансы\" в искусстве Запада   \n","RSL01008656366     Мышление и речь : психологические исследования   \n","\n","                                                         bbk  \n","sys_numb                                                      \n","RSL01008586079                                          none  \n","RSL01005241876                                          none  \n","RSL01000958257                                      Ш40я73-1  \n","RSL01003345484                                 Э372.24-438,0  \n","RSL01002884697                                        Д8я212  \n","RSL01002840579            Р716.110я73-1\\nЕ83я73-1\\nЕ866я73-1  \n","RSL01001071360                                          none  \n","RSL01004397358                            Ш1я212\\nШ141.2я212  \n","RSL01003332475                   Ш12-913я212\\nЧ426.11(2)я212  \n","RSL01001848685                                          none  \n","RSL01000704558                                    Т3(2)5-8я1  \n","RSL01009824143                     Щ158.9открытки3(2Р=Р)53-3  \n","RSL01003817085  Т3(2)7-498я2\\nТ3(4Фр9=Рус)6я2\\nТ3(4Фр)6-56я2  \n","RSL01008141826                                          none  \n","RSL01004117881             У03(2)002.9д,0\\nУ03(2)502.121.4,0  \n","RSL01001847464                                          none  \n","RSL01002890388                         Н0(2Р=Р)92Москва07я23  \n","RSL01008066836                               Ш5(2=Р)53-65я44  \n","RSL01003015223                                   Щ103(0)43,0  \n","RSL01008656366                                          none  "],"text/html":["\n","  <div id=\"df-79df127c-8f31-4a52-9921-a36f4de1a941\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>recs_pop</th>\n","      <th>users_count</th>\n","      <th>author</th>\n","      <th>title</th>\n","      <th>bbk</th>\n","    </tr>\n","    <tr>\n","      <th>sys_numb</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>RSL01008586079</th>\n","      <td>1464</td>\n","      <td>8</td>\n","      <td>Карцхия, Александр Амиранович</td>\n","      <td>Гражданско-правовая модель регулирования цифро...</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01005241876</th>\n","      <td>991</td>\n","      <td>10</td>\n","      <td>Карклин, Павел Иванович</td>\n","      <td>Военная книга (1934-1939) : библиографический ...</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01000958257</th>\n","      <td>681</td>\n","      <td>11</td>\n","      <td>Хализев, Валентин Евгеньевич</td>\n","      <td>Теория литературы : Учеб. для студентов вузов</td>\n","      <td>Ш40я73-1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003345484</th>\n","      <td>606</td>\n","      <td>5</td>\n","      <td>Рябинин, Юрий Валерьевич 1963-</td>\n","      <td>Русское юродство</td>\n","      <td>Э372.24-438,0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002884697</th>\n","      <td>573</td>\n","      <td>14</td>\n","      <td>Азимов, Айзек 1920-1992</td>\n","      <td>Слова на карте = Words on the map : геогр. наз...</td>\n","      <td>Д8я212</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002840579</th>\n","      <td>541</td>\n","      <td>11</td>\n","      <td>Кузнецов, Сергей Львович</td>\n","      <td>Гистология, цитология и эмбриология : : учебни...</td>\n","      <td>Р716.110я73-1\\nЕ83я73-1\\nЕ866я73-1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01001071360</th>\n","      <td>520</td>\n","      <td>7</td>\n","      <td>Алексеев, Сергей Сергеевич</td>\n","      <td>Общая теория права : Курс в 2-х т. / С. С. Але...</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004397358</th>\n","      <td>518</td>\n","      <td>9</td>\n","      <td>Ахманова, Ольга Сергеевна</td>\n","      <td>Словарь лингвистических терминов : лингвостили...</td>\n","      <td>Ш1я212\\nШ141.2я212</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003332475</th>\n","      <td>491</td>\n","      <td>6</td>\n","      <td>Щукин, Анатолий Николаевич</td>\n","      <td>Лингводидактический энциклопедический словарь ...</td>\n","      <td>Ш12-913я212\\nЧ426.11(2)я212</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01001848685</th>\n","      <td>480</td>\n","      <td>8</td>\n","      <td>none</td>\n","      <td>Справочник по фондам Архива внешней политики Р...</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01000704558</th>\n","      <td>414</td>\n","      <td>5</td>\n","      <td>Шилов, Денис Николаевич</td>\n","      <td>Государственные деятели Российской империи. Гл...</td>\n","      <td>Т3(2)5-8я1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003817085</th>\n","      <td>383</td>\n","      <td>9</td>\n","      <td>none</td>\n","      <td>Российское Зарубежье во Франции, 1919-2000 = L...</td>\n","      <td>Т3(2)7-498я2\\nТ3(4Фр9=Рус)6я2\\nТ3(4Фр)6-56я2</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01008141826</th>\n","      <td>381</td>\n","      <td>3</td>\n","      <td>Яковлева, Елена Вячеславовна</td>\n","      <td>Повышение эффективности очистки производственн...</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004117881</th>\n","      <td>378</td>\n","      <td>6</td>\n","      <td>Петров, Юрий Александрович, 1955-</td>\n","      <td>Наемные управляющие в России : опыт бизнес-эли...</td>\n","      <td>У03(2)002.9д,0\\nУ03(2)502.121.4,0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01001847464</th>\n","      <td>373</td>\n","      <td>7</td>\n","      <td>none</td>\n","      <td>Лингвистический энциклопедический словарь</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002890388</th>\n","      <td>357</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>Архитектура Москвы, 1920-1960 = Moscow archite...</td>\n","      <td>Н0(2Р=Р)92Москва07я23</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01008066836</th>\n","      <td>355</td>\n","      <td>4</td>\n","      <td>Белый, Андрей 1880-1934</td>\n","      <td>Стихотворения</td>\n","      <td>Ш5(2=Р)53-65я44</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003015223</th>\n","      <td>339</td>\n","      <td>9</td>\n","      <td>Панофский, Эрвин 1892-1968</td>\n","      <td>Ренессанс и \"ренессансы\" в искусстве Запада</td>\n","      <td>Щ103(0)43,0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01008656366</th>\n","      <td>338</td>\n","      <td>8</td>\n","      <td>Выготский, Лев Семенович 1896-1934</td>\n","      <td>Мышление и речь : психологические исследования</td>\n","      <td>none</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-79df127c-8f31-4a52-9921-a36f4de1a941')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-79df127c-8f31-4a52-9921-a36f4de1a941 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-79df127c-8f31-4a52-9921-a36f4de1a941');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":121}],"source":["print(\"Самые популярные книги в рекомендациях и количество их читателей:\")\n","predicted_books.sort_values(by='recs_pop', ascending=False).head(20)"]},{"cell_type":"markdown","metadata":{"id":"OctJt2kCPVtc"},"source":["**Вывод 2**\n","\n","1. Самые популярные книги в нашей системе рекомендуются **1-2 тысячам** пользователей\n","\n","2. У большинства таких книг как минимум `5 читателей` (! возьмем на заметку)\n","\n","3. По описаниям и bbk видим, что у рекомендаций широкий охват жанров)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":281},"executionInfo":{"elapsed":865,"status":"ok","timestamp":1663067807636,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"B0VpYvTPQg3t","outputId":"57b3e7ac-bb46-436b-e1ae-f9f3b7f1c779"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["predicted_books.query('recs_pop > 50').recs_pop.hist();\n","plt.title(\"Книги, которы система порекомендовала > 50 раз:\")\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"uipt1B4WR4D8"},"source":["Возможно, именно здесь находятся рекомендации, которые попали в яблочко)"]},{"cell_type":"markdown","metadata":{"id":"0V0oFUVqdfeK"},"source":["### (4.3) Общий Hit-Rate"]},{"cell_type":"markdown","metadata":{"id":"TET9gOduUs_u"},"source":["- Напишем функцию для попарного сравнения `датафреймов с рекомендациями`\n","\n","- Она пригодится нам для оценки качества предсказаний"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"KJ2Ivqiu9X9c"},"outputs":[],"source":["def compare_two_df(df1, df2):\n","  \"\"\"\n","  Объединяет два датафрейма по парам (читатель, книга)\n","\n","  в результате остается только их пересечение(там, где пары совпали)\n","  \"\"\"\n","\n","  res = pd.merge(df1[['chb', 'sys_numb']],\n","                 df2[['chb', 'sys_numb']],\n","                 on=['chb', 'sys_numb'], how='inner')\n","\n","  return res\n","\n","\n","# Сохраняем датасет с предсказаниями, которые попали в точку\n","compare_results = compare_two_df(all_users_predictions, test_data)"]},{"cell_type":"markdown","metadata":{"id":"oysqmnQbd3qc"},"source":["- Посмотрим, какое количество книг мы в среднем угадали для каждого пользователя.\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":281},"executionInfo":{"elapsed":1884,"status":"ok","timestamp":1663067840532,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"9i_rteRZazDZ","outputId":"f7e0cbf4-8433-4da9-c790-61c0d10a0965"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["true_books = compare_two_df(all_users_predictions, test_data)\n","true_books.groupby('chb').count().hist()\n","plt.title(\"Количество угаданных книг на одного пользователя\")\n","plt.show()"]},{"cell_type":"markdown","metadata":{"id":"pMb8V1BRbRwc"},"source":["**Вывод**\n","\n","Мы \"захватили\" ~500 читателей\n","\n","  - Для  большинства угадали одну книгу\n","\n","  - Есть и те, для которых у нас даже 3 и 4 попадания)\n","\n","Возможно, это рекордсмены, которые заказывают больше 1000 книг? Проверим!"]},{"cell_type":"markdown","metadata":{"id":"32qk-UpSheoL"},"source":["### (4.4) Какие книги мы угадываем?"]},{"cell_type":"markdown","metadata":{"id":"Ef6qm9c3S0G9"},"source":["С реальными читателями мы узнаем это только после запуска системы в продакшен.\n","\n","Попробуем пока разгадать тайну тестовых данных (что работает, а что нет)"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"95DULaxXqJjC"},"outputs":[],"source":["true_books = true_books.set_index('sys_numb').join(predicted_books)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"elapsed":4,"status":"ok","timestamp":1663067954390,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"iSm46JrQr78n","outputId":"9df4c49f-da27-4bd4-b358-c4533c71a4c3"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["                         chb  recs_pop  users_count  \\\n","sys_numb                                              \n","RSL01009824143  200001024503       387           16   \n","RSL01009824143  300000729937       387           16   \n","RSL01009824143  200000944083       387           16   \n","RSL01009824143  200000839923       387           16   \n","RSL01009824143  200000794551       387           16   \n","RSL01009824143  100000954431       387           16   \n","RSL01002890388  200000866607       357           15   \n","RSL01002890388  300000860566       357           15   \n","RSL01003015223  300000973621       339            9   \n","RSL07000372451  300001075805       336           11   \n","RSL01008275241  300001080798       294            9   \n","RSL01003360717  300000923586       246            9   \n","RSL01002834571  100001079500       232            8   \n","RSL01001343016  200000769440       215           10   \n","RSL01009819676  300001017588       192           10   \n","RSL01004353863  300001062342       182           13   \n","RSL01001841694  200000969417       181            9   \n","RSL01001841694  300001046894       181            9   \n","RSL01009817128  300001015464       178           15   \n","RSL01009817128  200001092351       178           15   \n","\n","                                              author  \\\n","sys_numb                                               \n","RSL01009824143                                  none   \n","RSL01009824143                                  none   \n","RSL01009824143                                  none   \n","RSL01009824143                                  none   \n","RSL01009824143                                  none   \n","RSL01009824143                                  none   \n","RSL01002890388                                  none   \n","RSL01002890388                                  none   \n","RSL01003015223            Панофский, Эрвин 1892-1968   \n","RSL07000372451                           отсутствует   \n","RSL01008275241        Кузнецов, Сергей Львович 1951-   \n","RSL01003360717                         Неттер, Фрэнк   \n","RSL01002834571                                  none   \n","RSL01001343016  Бахтин, Михаил Михайлович 1895-1975.   \n","RSL01009819676                                  none   \n","RSL01004353863           Казусь, Игорь Александрович   \n","RSL01001841694                                  none   \n","RSL01001841694                                  none   \n","RSL01009817128                                  none   \n","RSL01009817128                                  none   \n","\n","                                                            title  \\\n","sys_numb                                                            \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01009824143     С Рождеством Христовым! Изоматериал : Открытка   \n","RSL01002890388  Архитектура Москвы, 1920-1960 = Moscow archite...   \n","RSL01002890388  Архитектура Москвы, 1920-1960 = Moscow archite...   \n","RSL01003015223        Ренессанс и \"ренессансы\" в искусстве Запада   \n","RSL07000372451                                        отсутствует   \n","RSL01008275241  Гистология, цитология и эмбриология : учебник ...   \n","RSL01003360717                            Атлас анатомии человека   \n","RSL01002834571  Большая цензура: писатели и журналисты в Стран...   \n","RSL01001343016                     Эстетика словесного творчества   \n","RSL01009819676      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01004353863  Советская архитектура 1920-х годов: организаци...   \n","RSL01001841694                      Народы Дагестана : Монография   \n","RSL01001841694                      Народы Дагестана : Монография   \n","RSL01009817128      С Рождеством Христовым Изоматериал : Открытка   \n","RSL01009817128      С Рождеством Христовым Изоматериал : Открытка   \n","\n","                                                              bbk  \n","sys_numb                                                           \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009824143                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01002890388                              Н0(2Р=Р)92Москва07я23  \n","RSL01002890388                              Н0(2Р=Р)92Москва07я23  \n","RSL01003015223                                        Щ103(0)43,0  \n","RSL07000372451                                        отсутствует  \n","RSL01008275241                      Е05я73-1\\nЕ83я73-1\\nЕ866я73-1  \n","RSL01003360717                                            Е860я61  \n","RSL01002834571  Ш5(2=Р)7-305.2я43\\nЧ610.3(2)7-7я45\\nХ621.152.2я45  \n","RSL01001343016                                               none  \n","RSL01009819676                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01004353863                              Н0(2)71,0\\nН2л.г(2),0  \n","RSL01001841694                                               none  \n","RSL01001841694                                               none  \n","RSL01009817128                          Щ158.9открытки3(2Р=Р)53-3  \n","RSL01009817128                          Щ158.9открытки3(2Р=Р)53-3  "],"text/html":["\n","  <div id=\"df-531d9388-e941-4374-897a-c35cc8a1b7d3\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>chb</th>\n","      <th>recs_pop</th>\n","      <th>users_count</th>\n","      <th>author</th>\n","      <th>title</th>\n","      <th>bbk</th>\n","    </tr>\n","    <tr>\n","      <th>sys_numb</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>200001024503</td>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>300000729937</td>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>200000944083</td>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>200000839923</td>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>200000794551</td>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009824143</th>\n","      <td>100000954431</td>\n","      <td>387</td>\n","      <td>16</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым! Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002890388</th>\n","      <td>200000866607</td>\n","      <td>357</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>Архитектура Москвы, 1920-1960 = Moscow archite...</td>\n","      <td>Н0(2Р=Р)92Москва07я23</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002890388</th>\n","      <td>300000860566</td>\n","      <td>357</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>Архитектура Москвы, 1920-1960 = Moscow archite...</td>\n","      <td>Н0(2Р=Р)92Москва07я23</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003015223</th>\n","      <td>300000973621</td>\n","      <td>339</td>\n","      <td>9</td>\n","      <td>Панофский, Эрвин 1892-1968</td>\n","      <td>Ренессанс и \"ренессансы\" в искусстве Запада</td>\n","      <td>Щ103(0)43,0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL07000372451</th>\n","      <td>300001075805</td>\n","      <td>336</td>\n","      <td>11</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01008275241</th>\n","      <td>300001080798</td>\n","      <td>294</td>\n","      <td>9</td>\n","      <td>Кузнецов, Сергей Львович 1951-</td>\n","      <td>Гистология, цитология и эмбриология : учебник ...</td>\n","      <td>Е05я73-1\\nЕ83я73-1\\nЕ866я73-1</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01003360717</th>\n","      <td>300000923586</td>\n","      <td>246</td>\n","      <td>9</td>\n","      <td>Неттер, Фрэнк</td>\n","      <td>Атлас анатомии человека</td>\n","      <td>Е860я61</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01002834571</th>\n","      <td>100001079500</td>\n","      <td>232</td>\n","      <td>8</td>\n","      <td>none</td>\n","      <td>Большая цензура: писатели и журналисты в Стран...</td>\n","      <td>Ш5(2=Р)7-305.2я43\\nЧ610.3(2)7-7я45\\nХ621.152.2я45</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01001343016</th>\n","      <td>200000769440</td>\n","      <td>215</td>\n","      <td>10</td>\n","      <td>Бахтин, Михаил Михайлович 1895-1975.</td>\n","      <td>Эстетика словесного творчества</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009819676</th>\n","      <td>300001017588</td>\n","      <td>192</td>\n","      <td>10</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01004353863</th>\n","      <td>300001062342</td>\n","      <td>182</td>\n","      <td>13</td>\n","      <td>Казусь, Игорь Александрович</td>\n","      <td>Советская архитектура 1920-х годов: организаци...</td>\n","      <td>Н0(2)71,0\\nН2л.г(2),0</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01001841694</th>\n","      <td>200000969417</td>\n","      <td>181</td>\n","      <td>9</td>\n","      <td>none</td>\n","      <td>Народы Дагестана : Монография</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01001841694</th>\n","      <td>300001046894</td>\n","      <td>181</td>\n","      <td>9</td>\n","      <td>none</td>\n","      <td>Народы Дагестана : Монография</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009817128</th>\n","      <td>300001015464</td>\n","      <td>178</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","    <tr>\n","      <th>RSL01009817128</th>\n","      <td>200001092351</td>\n","      <td>178</td>\n","      <td>15</td>\n","      <td>none</td>\n","      <td>С Рождеством Христовым Изоматериал : Открытка</td>\n","      <td>Щ158.9открытки3(2Р=Р)53-3</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-531d9388-e941-4374-897a-c35cc8a1b7d3')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-531d9388-e941-4374-897a-c35cc8a1b7d3 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-531d9388-e941-4374-897a-c35cc8a1b7d3');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":127}],"source":["true_books.sort_values(by='recs_pop', ascending=False).head(20)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":265},"executionInfo":{"elapsed":294,"status":"ok","timestamp":1663067956068,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"ZXowhoi8spNj","outputId":"4c5350a9-73a9-4d6b-cd22-64bffd70e942"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}],"source":["true_books.recs_pop.hist();"]},{"cell_type":"markdown","metadata":{"id":"MS_XfAy_sfIZ"},"source":["**Вывод 1**\n","\n","- Лишь 4 книги, которыми мы распылялись тысячами, попали в яблочко\n","\n","- В основном выйграли те, которых в рекомендательном датасете было < 200"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":300},"executionInfo":{"elapsed":732,"status":"ok","timestamp":1663067957084,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"tCRbX4pQ6vuk","outputId":"fc9a6f7b-2ebc-49bd-b317-9ac57ca2acc8"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["               test   true_recs\n","count  45610.000000  419.000000\n","mean       1.459855    5.081146\n","std        1.128960    4.089677\n","min        1.000000    2.000000\n","25%        1.000000    2.000000\n","50%        1.000000    3.000000\n","75%        2.000000    6.000000\n","max       21.000000   21.000000"],"text/html":["\n","  <div id=\"df-f463b626-1402-4d67-a434-ad3384097811\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>test</th>\n","      <th>true_recs</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>45610.000000</td>\n","      <td>419.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>1.459855</td>\n","      <td>5.081146</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>1.128960</td>\n","      <td>4.089677</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>1.000000</td>\n","      <td>2.000000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>1.000000</td>\n","      <td>2.000000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>1.000000</td>\n","      <td>3.000000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>2.000000</td>\n","      <td>6.000000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>21.000000</td>\n","      <td>21.000000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f463b626-1402-4d67-a434-ad3384097811')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-f463b626-1402-4d67-a434-ad3384097811 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-f463b626-1402-4d67-a434-ad3384097811');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":129}],"source":["(test_data.set_index('sys_numb')\n",".join(books_popularity)\n",".users_count.describe().to_frame('test')\n",".join(true_books.users_count.describe().to_frame('true_recs')))"]},{"cell_type":"markdown","metadata":{"id":"fVv84JEUAUG5"},"source":["**Вывод 2**\n","\n","- Из 75% книг, которые читали только один раз, мы не угадали ни одной\n","\n","- Это логично, ведь идея Collaborative Filtering - использовать в рекомендациях данные о предпочтениях других читателях\n"]},{"cell_type":"markdown","metadata":{"id":"6ntNZeQ0dd2k"},"source":["###(4.5) Пользователи, которых мы угадываем\n","\n"]},{"cell_type":"markdown","metadata":{"id":"3vTVsEaVeF8r"},"source":["Попробуем выяснить, для каких читателей подходят наши рекомендации на базе Collaborative Filtering.\n","\n","Возможно, мы угадываем книги лишь для того 1% читателей, кто уже прочитал по 100 и 200 книг?"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":473},"executionInfo":{"elapsed":277,"status":"ok","timestamp":1663068259571,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"iVbwNhxJGqix","outputId":"fe104d63-27fa-4fb5-a0e3-9a91e8cb4835"},"outputs":[{"output_type":"stream","name":"stdout","text":["Пользователи, для которых модель смогла предсказать книги:\n"]},{"output_type":"execute_result","data":{"text/plain":["              recs_pop  users_count                                author  \\\n","chb                                                                         \n","200000935252         9            3      Пономаренко, Сергей Владимирович   \n","300001093459        12            5                             По, Айрэн   \n","200001063517        15            3                                  none   \n","200000774458        55            5  Медушевский, Андрей Николаевич 1960-   \n","200001117131         4            2           Папазян, Георгий Самвелович   \n","...                ...          ...                                   ...   \n","300000863416         4            5                           отсутствует   \n","300000863416         4            3                           отсутствует   \n","200001017435         1            2                           отсутствует   \n","100000903206         4            3              Девятко, Инна Феликсовна   \n","200000789223         3            3                             Акс, Джош   \n","\n","                                                          title  \\\n","chb                                                               \n","200000935252  Основные этапы истории развития российской про...   \n","300001093459                                      Родовая Карма   \n","200001063517  Неорганическая химия. Химия элементов : учебни...   \n","200000774458  Демократия и авторитаризм: российский конститу...   \n","200001117131  Социально-педагогические условия реализации во...   \n","...                                                         ...   \n","300000863416                                        отсутствует   \n","300000863416                                        отсутствует   \n","200001017435                                        отсутствует   \n","100000903206  Социологические теории деятельности и практиче...   \n","200000789223  Коллагеновая диета : 28-дневный план для устой...   \n","\n","                                bbk  books_count  \n","chb                                               \n","200000935252                   none            2  \n","300001093459             Э391.219,0            2  \n","200001063517              Г12я731-1            3  \n","200000774458                   none            3  \n","200001117131      Ч481,6(2)385.68,0            3  \n","...                             ...          ...  \n","300000863416            отсутствует          747  \n","300000863416            отсутствует          747  \n","200001017435            отсутствует          805  \n","100000903206                   none         1114  \n","200000789223  Л996.5я27\\nР123.021я9         3914  \n","\n","[419 rows x 6 columns]"],"text/html":["\n","  <div id=\"df-64ebf48a-1bba-490e-aae4-865d52225216\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>recs_pop</th>\n","      <th>users_count</th>\n","      <th>author</th>\n","      <th>title</th>\n","      <th>bbk</th>\n","      <th>books_count</th>\n","    </tr>\n","    <tr>\n","      <th>chb</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>200000935252</th>\n","      <td>9</td>\n","      <td>3</td>\n","      <td>Пономаренко, Сергей Владимирович</td>\n","      <td>Основные этапы истории развития российской про...</td>\n","      <td>none</td>\n","      <td>2</td>\n","    </tr>\n","    <tr>\n","      <th>300001093459</th>\n","      <td>12</td>\n","      <td>5</td>\n","      <td>По, Айрэн</td>\n","      <td>Родовая Карма</td>\n","      <td>Э391.219,0</td>\n","      <td>2</td>\n","    </tr>\n","    <tr>\n","      <th>200001063517</th>\n","      <td>15</td>\n","      <td>3</td>\n","      <td>none</td>\n","      <td>Неорганическая химия. Химия элементов : учебни...</td>\n","      <td>Г12я731-1</td>\n","      <td>3</td>\n","    </tr>\n","    <tr>\n","      <th>200000774458</th>\n","      <td>55</td>\n","      <td>5</td>\n","      <td>Медушевский, Андрей Николаевич 1960-</td>\n","      <td>Демократия и авторитаризм: российский конститу...</td>\n","      <td>none</td>\n","      <td>3</td>\n","    </tr>\n","    <tr>\n","      <th>200001117131</th>\n","      <td>4</td>\n","      <td>2</td>\n","      <td>Папазян, Георгий Самвелович</td>\n","      <td>Социально-педагогические условия реализации во...</td>\n","      <td>Ч481,6(2)385.68,0</td>\n","      <td>3</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>300000863416</th>\n","      <td>4</td>\n","      <td>5</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>747</td>\n","    </tr>\n","    <tr>\n","      <th>300000863416</th>\n","      <td>4</td>\n","      <td>3</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>747</td>\n","    </tr>\n","    <tr>\n","      <th>200001017435</th>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>отсутствует</td>\n","      <td>805</td>\n","    </tr>\n","    <tr>\n","      <th>100000903206</th>\n","      <td>4</td>\n","      <td>3</td>\n","      <td>Девятко, Инна Феликсовна</td>\n","      <td>Социологические теории деятельности и практиче...</td>\n","      <td>none</td>\n","      <td>1114</td>\n","    </tr>\n","    <tr>\n","      <th>200000789223</th>\n","      <td>3</td>\n","      <td>3</td>\n","      <td>Акс, Джош</td>\n","      <td>Коллагеновая диета : 28-дневный план для устой...</td>\n","      <td>Л996.5я27\\nР123.021я9</td>\n","      <td>3914</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>419 rows × 6 columns</p>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-64ebf48a-1bba-490e-aae4-865d52225216')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-64ebf48a-1bba-490e-aae4-865d52225216 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-64ebf48a-1bba-490e-aae4-865d52225216');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":135}],"source":["print(\"Пользователи, для которых модель смогла предсказать книги:\")\n","true_books.set_index('chb').join(user_books_count).sort_values(by='books_count')"]},{"cell_type":"code","source":["print(\"Количество книг в истории пользователей\\n\",\n","      \"для которых мы хорошо предсказываем:\\n\")\n","\n","true_books.set_index('chb').join(user_books_count).books_count.describe()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"NAK-FfP4mLiJ","executionInfo":{"status":"ok","timestamp":1663068415467,"user_tz":-180,"elapsed":247,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"}},"outputId":"c01e96db-1964-435b-ca7a-4820b0d2ab8b"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Количество книг в истории пользователей\n"," для которых мы хорошо предсказываем:\n","\n"]},{"output_type":"execute_result","data":{"text/plain":["count     419.000000\n","mean       53.558473\n","std       215.422847\n","min         2.000000\n","25%         9.000000\n","50%        18.000000\n","75%        36.500000\n","max      3914.000000\n","Name: books_count, dtype: float64"]},"metadata":{},"execution_count":139}]},{"cell_type":"code","source":["print(\"Количество книг в истории читателей\\n\",\n","      \"в общем датасте:\")\n","(transactions\n"," .drop_duplicates()\n"," .groupby('chb').sys_numb.count()\n"," .describe()\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"dJsvoH3071yH","executionInfo":{"status":"ok","timestamp":1663074365910,"user_tz":-180,"elapsed":361,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"}},"outputId":"715e5a70-02a1-416e-947d-0a3f432d1810"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Количество книг в истории читателей\n"," в общем датасте:\n"]},{"output_type":"execute_result","data":{"text/plain":["count    16753.000000\n","mean        15.493703\n","std         75.221343\n","min          2.000000\n","25%          4.000000\n","50%          6.000000\n","75%         14.000000\n","max       6765.000000\n","Name: sys_numb, dtype: float64"]},"metadata":{},"execution_count":11}]},{"cell_type":"markdown","metadata":{"id":"9ItWExZ1fzLC"},"source":["**Вывод**\n","\n","1. 50% пользователей, которым подошли наши рекомендации, прочитали больше 18 книг (вспомним, что по общему датасету этот показатель в 3 раза ниже)\n","\n","1. При этом мы  попадали в яблочко и для пользователей, которые прочитали не больше 3-х книг\n","\n","  - Возможно, они прочитали всего 3, но при этом популярные среди других читателей книги?"]},{"cell_type":"markdown","metadata":{"id":"5Annr7d30Pb7"},"source":["### (4.6) Неявные дубликаты книг \n"]},{"cell_type":"markdown","source":["Появилась идея, что кроме явных дубликатов (которые мы не используем в предсказаниях), в базе данных есть похожие книги с разными идентификаторами:\n","\n","- Книги с разным годом издания\n","\n","- Книги из одного сборника или серии\n","\n","Получается, ALS-модель считает эти книги разными, и в каких-то случаях не может корректно объединить пользователей по интересам.\n","\n","Посмотрим, много ли таких книг?"],"metadata":{"id":"kpBRf0osAqcs"}},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":3,"status":"ok","timestamp":1663076022459,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"So17gVWn0Pb8","outputId":"df76d825-e0bb-4f77-bd5c-325a8ae79d9a"},"outputs":[{"output_type":"stream","name":"stdout","text":["Книг в библиотеке с одинаковыми названиями: 34534\n"]}],"source":["print(\"Книг в библиотеке с одинаковыми названиями:\", items['title'].duplicated().sum())"]},{"cell_type":"markdown","metadata":{"id":"DRlS7VzP0Pb8"},"source":["- 34534!! А ведь это только сами копии, еще есть исходники\n","\n","- Значит, минимум 50 000 одинаковых книг с разными изданиями\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":2419,"status":"ok","timestamp":1663076437358,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"ndWoYNQs0Pb9","outputId":"19745318-f46f-4d7c-c2b5-b0e2de859c9f"},"outputs":[{"output_type":"stream","name":"stdout","text":["=>Примеры неявных дубликатов:<=\n","\n"]},{"output_type":"execute_result","data":{"text/plain":["title                                                                                                                                                                                                                                   author                                 \n","Методические и информационные материалы по правовой пропаганде                                                                                                                                                                          none                                        5\n","Философия : учебник                                                                                                                                                                                                                     none                                        5\n","Веды о силе времени : практические рекомендации для процветания                                                                                                                                                                         Торсунов, Олег Геннадьевич                  5\n","История России в схемах : учебное пособие                                                                                                                                                                                               none                                        6\n","Краткий философский словарь                                                                                                                                                                                                             none                                        6\n","Криминалистика : учебник                                                                                                                                                                                                                none                                        6\n","Основы семиотики заболеваний внутренних органов : учебное пособие для студентов учреждений высшего профессионального образования, обучающихся по специальности 060101.65 \"Лечебное дело\" дисциплины \"Пропедевтика внутренних болезней\"  none                                        6\n","Курс кройки : Новейшее практ. руководство для изучения кройки дам., дет. и верх. платьев без посторон. помощи : Со многими черт. и выкройками : Сост. по фр. покрою Верой Васильевной Шишмаревой                                        Шишмарева, Вера Васильевна                  6\n","Философский словарь                                                                                                                                                                                                                     none                                        7\n","Русское правописание : Руководство, сост. по поручению 2-го Отд-ния акад. наук, акад. Я.К. Гротом                                                                                                                                       Грот, Яков Карлович 1812-1893               7\n","Ведомость справок о судимости / М-во юстиции : Справки о судимости по приговорам общих судебных установлений                                                                                                                            none                                        8\n","История России : учебник                                                                                                                                                                                                                none                                        8\n","Лучшие сказки братьев Гриммов : Пособие для клас. чтения в рус. учеб. заведениях                                                                                                                                                        Гримм, Якоб 1785-1863                       9\n","Русский язык : Опыт практ. учеб. рус. грамматики : Этимология в образцах                                                                                                                                                                Петров, Константин Федорович 1849?-1914    10\n","План города Москвы Карты                                                                                                                                                                                                                Москва, город                              10\n","Name: year_izd, dtype: int64"]},"metadata":{},"execution_count":40}],"source":["print(\"=>Примеры неявных дубликатов:<=\\n\")\n","\n","(items\n"," .groupby(['title', 'author'])\n"," .year_izd.nunique().sort_values()).tail(15)"]},{"cell_type":"markdown","metadata":{"id":"8xraeS5N0Pb9"},"source":["- Часть дубликатов мы еще не видим, т.к. названия могут немного отличаться (особенно, если это книги одной серии или сборника)\n","\n","- Посмотрим пример с братьями Гримм\n","\n","\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":331},"executionInfo":{"elapsed":701,"status":"ok","timestamp":1663076392796,"user":{"displayName":"Елена Хафизова","userId":"10505641158644519701"},"user_tz":-180},"id":"OPAXJtq14R5-","outputId":"5342a4cb-2513-4d88-859f-49459215190b"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["              sys_numb                                              title  \\\n","353346  RSL01003628565  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353349  RSL01003628562  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353351  RSL01003628566  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353352  RSL01003628560  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353354  RSL01003628559  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353356  RSL01003628558  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353357  RSL01003628567  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353359  RSL01003628564  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","353360  RSL01003628563  Лучшие сказки братьев Гриммов : Пособие для кл...   \n","\n","                       author                                            izd  \\\n","353346  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353349  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353351  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353352  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353354  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353356  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353357  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353359  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","353360  Гримм, Якоб 1785-1863  кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"   \n","\n","       year_izd   bbk  \n","353346     1910  none  \n","353349     1905  none  \n","353351     1911  none  \n","353352     1899  none  \n","353354     1893  none  \n","353356     1891  none  \n","353357     1914  none  \n","353359     1909  none  \n","353360     1908  none  "],"text/html":["\n","  <div id=\"df-ca9c014c-32a4-4878-9651-2866310d70b1\">\n","    <div class=\"colab-df-container\">\n","      <div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sys_numb</th>\n","      <th>title</th>\n","      <th>author</th>\n","      <th>izd</th>\n","      <th>year_izd</th>\n","      <th>bbk</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>353346</th>\n","      <td>RSL01003628565</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1910</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353349</th>\n","      <td>RSL01003628562</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1905</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353351</th>\n","      <td>RSL01003628566</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1911</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353352</th>\n","      <td>RSL01003628560</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1899</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353354</th>\n","      <td>RSL01003628559</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1893</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353356</th>\n","      <td>RSL01003628558</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1891</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353357</th>\n","      <td>RSL01003628567</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1914</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353359</th>\n","      <td>RSL01003628564</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1909</td>\n","      <td>none</td>\n","    </tr>\n","    <tr>\n","      <th>353360</th>\n","      <td>RSL01003628563</td>\n","      <td>Лучшие сказки братьев Гриммов : Пособие для кл...</td>\n","      <td>Гримм, Якоб 1785-1863</td>\n","      <td>кн. маг. В. Думнова, п/ф \"Насл. бр. Салаевых\"</td>\n","      <td>1908</td>\n","      <td>none</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>\n","      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ca9c014c-32a4-4878-9651-2866310d70b1')\"\n","              title=\"Convert this dataframe to an interactive table.\"\n","              style=\"display:none;\">\n","        \n","  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n","       width=\"24px\">\n","    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n","    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n","  </svg>\n","      </button>\n","      \n","  <style>\n","    .colab-df-container {\n","      display:flex;\n","      flex-wrap:wrap;\n","      gap: 12px;\n","    }\n","\n","    .colab-df-convert {\n","      background-color: #E8F0FE;\n","      border: none;\n","      border-radius: 50%;\n","      cursor: pointer;\n","      display: none;\n","      fill: #1967D2;\n","      height: 32px;\n","      padding: 0 0 0 0;\n","      width: 32px;\n","    }\n","\n","    .colab-df-convert:hover {\n","      background-color: #E2EBFA;\n","      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n","      fill: #174EA6;\n","    }\n","\n","    [theme=dark] .colab-df-convert {\n","      background-color: #3B4455;\n","      fill: #D2E3FC;\n","    }\n","\n","    [theme=dark] .colab-df-convert:hover {\n","      background-color: #434B5C;\n","      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n","      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n","      fill: #FFFFFF;\n","    }\n","  </style>\n","\n","      <script>\n","        const buttonEl =\n","          document.querySelector('#df-ca9c014c-32a4-4878-9651-2866310d70b1 button.colab-df-convert');\n","        buttonEl.style.display =\n","          google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n","        async function convertToInteractive(key) {\n","          const element = document.querySelector('#df-ca9c014c-32a4-4878-9651-2866310d70b1');\n","          const dataTable =\n","            await google.colab.kernel.invokeFunction('convertToInteractive',\n","                                                     [key], {});\n","          if (!dataTable) return;\n","\n","          const docLinkHtml = 'Like what you see? Visit the ' +\n","            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n","            + ' to learn more about interactive tables.';\n","          element.innerHTML = '';\n","          dataTable['output_type'] = 'display_data';\n","          await google.colab.output.renderOutput(dataTable, element);\n","          const docLink = document.createElement('div');\n","          docLink.innerHTML = docLinkHtml;\n","          element.appendChild(docLink);\n","        }\n","      </script>\n","    </div>\n","  </div>\n","  "]},"metadata":{},"execution_count":37}],"source":["items[items.title.str.contains('Лучшие сказки братьев Гриммов')]"]},{"cell_type":"markdown","source":["**Вывод**\n","\n","На неявных дубликатах модель ALS делает неправильные выводы\n","\n","- Если объединить книги на этапе обучения, то это должно повысить качество предсказаний\n","\n","- Останется вопрос какую из группы похожих книг порекомендовать пользователю) "],"metadata":{"id":"97c0BBXFDtt_"}}]}