code_with_conversion.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['.git', '.ipynb_checkpoints', '1581184759', '1581186354', 'code-Copy1.ipynb', 'code.ipynb', 'converion to tflite.ipynb', 'model.pb', 'models', 'rps.h5', 'sample_submission.csv', 'submit1.csv', 'submit2.csv', 'test.csv', 'train.csv']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import train_test_split\n",
    "import itertools\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from keras.utils.np_utils import to_categorical\n",
    "#from tensorflow.keras.models import Sequential\n",
    "#from tensorflow.keras.layers import Dense,Conv2D,MaxPool2D,Flatten,Dropout\n",
    "from tensorflow.keras.optimizers import RMSprop\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.callbacks import ReduceLROnPlateau\n",
    "\n",
    "sns.set(style='white', context='notebook', palette='deep')\n",
    "\n",
    "\n",
    "\n",
    "import os\n",
    "print(os.listdir(\".\"))\n",
    "\n",
    "# Any results you write to the current directory are saved as output."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.13.1\n"
     ]
    }
   ],
   "source": [
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "outputs": [],
   "source": [
    "train=pd.read_csv(\"train.csv\")\n",
    "test=pd.read_csv(\"test.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_train=train[\"label\"]\n",
    "X_train=train.drop(labels = [\"label\"],axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train=X_train/255\n",
    "test=test/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train.shape\n",
    "X_train=X_train.values.reshape(-1,28,28,1)\n",
    "Y_train = to_categorical(Y_train, num_classes = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train,x_val,y_train,y_val=train_test_split(X_train, Y_train, test_size = 0.3, random_state = 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1c71750f128>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x_train[0][:,:,0],cmap=\"gray_r\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\ops\\resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\keras\\layers\\core.py:143: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 24, 24, 32)        832       \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 12, 12, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 8, 8, 32)          25632     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 4, 4, 32)          0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 4, 4, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 2, 2, 64)          18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 1, 1, 64)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 1, 1, 128)         8320      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 1, 1, 128)         0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1, 1, 128)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 256)               33024     \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                2570      \n",
      "=================================================================\n",
      "Total params: 88,874\n",
      "Trainable params: 88,874\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model=tf.keras.models.Sequential([\n",
    "tf.keras.layers.Conv2D(input_shape=(28,28,1),filters=32,kernel_size=(5,5),activation=\"relu\"),\n",
    "tf.keras.layers.MaxPool2D(pool_size=(2,2)),\n",
    "tf.keras.layers.Conv2D(filters=32,kernel_size=(5,5),activation=\"relu\"),\n",
    "tf.keras.layers.MaxPool2D(pool_size=(2,2)),\n",
    "tf.keras.layers.Dropout(0.25),\n",
    "tf.keras.layers.Conv2D(filters=64,kernel_size=(3,3),activation=\"relu\"),\n",
    "tf.keras.layers.MaxPool2D(pool_size=(2,2), strides=(2,2)),\n",
    "tf.keras.layers.Conv2D(filters=128,kernel_size=(1,1),activation=\"relu\"),\n",
    "tf.keras.layers.MaxPool2D(pool_size=(1,1), strides=(2,2)),\n",
    "tf.keras.layers.Dropout(0.25),\n",
    "tf.keras.layers.Flatten(),\n",
    "tf.keras.layers.Dense(256, activation = \"relu\"),\n",
    "tf.keras.layers.Dropout(0.5),\n",
    "tf.keras.layers.Dense(10, activation = \"softmax\")])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.0)\n",
    "model.compile(optimizer = optimizer,loss = \"categorical_crossentropy\",metrics=[\"accuracy\"])\n",
    "\n",
    "learn_rate= ReduceLROnPlateau(monitor='val_acc',patience=3,verbose=1,factor=0.5,min_lr=0.00001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "datagen = ImageDataGenerator(\n",
    "        featurewise_center=False,  # set input mean to 0 over the dataset\n",
    "        samplewise_center=False,  # set each sample mean to 0\n",
    "        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n",
    "        samplewise_std_normalization=False,  # divide each input by its std\n",
    "        zca_whitening=False,  # apply ZCA whitening\n",
    "        rotation_range=10,  # randomly rotate images in the range (degrees, 0 to 180)\n",
    "        zoom_range = 0.1, # Randomly zoom image \n",
    "        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)\n",
    "        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)\n",
    "        horizontal_flip=False,  # randomly flip images\n",
    "        vertical_flip=False)  # randomly flip images\n",
    "datagen.fit(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\ops\\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Epoch 1/10\n",
      "12600/12600 [==============================] - 4s 291us/sample - loss: 0.1136 - acc: 0.9660s - loss: 0.1128 - a\n",
      "489/489 [==============================] - 67s 136ms/step - loss: 0.6850 - acc: 0.7744 - val_loss: 0.1136 - val_acc: 0.9660\n",
      "Epoch 2/10\n",
      "12600/12600 [==============================] - 3s 264us/sample - loss: 0.0870 - acc: 0.9724\n",
      "489/489 [==============================] - 61s 124ms/step - loss: 0.1806 - acc: 0.9460 - val_loss: 0.0869 - val_acc: 0.9724\n",
      "Epoch 3/10\n",
      "12600/12600 [==============================] - 3s 209us/sample - loss: 0.0483 - acc: 0.9855\n",
      "489/489 [==============================] - 51s 105ms/step - loss: 0.1263 - acc: 0.9635 - val_loss: 0.0482 - val_acc: 0.9855\n",
      "Epoch 4/10\n",
      "12600/12600 [==============================] - 3s 211us/sample - loss: 0.0554 - acc: 0.9845\n",
      "489/489 [==============================] - 51s 105ms/step - loss: 0.1039 - acc: 0.9705 - val_loss: 0.0553 - val_acc: 0.9845\n",
      "Epoch 5/10\n",
      "12600/12600 [==============================] - 3s 216us/sample - loss: 0.0397 - acc: 0.9883\n",
      "489/489 [==============================] - 51s 104ms/step - loss: 0.0947 - acc: 0.9739 - val_loss: 0.0396 - val_acc: 0.9883\n",
      "Epoch 6/10\n",
      "12600/12600 [==============================] - 3s 209us/sample - loss: 0.0415 - acc: 0.9885\n",
      "489/489 [==============================] - 50s 102ms/step - loss: 0.0887 - acc: 0.9762 - val_loss: 0.0414 - val_acc: 0.9885\n",
      "Epoch 7/10\n",
      "12600/12600 [==============================] - 3s 208us/sample - loss: 0.0362 - acc: 0.9892\n",
      "489/489 [==============================] - 52s 106ms/step - loss: 0.0828 - acc: 0.9776 - val_loss: 0.0361 - val_acc: 0.9892\n",
      "Epoch 8/10\n",
      "12600/12600 [==============================] - 3s 206us/sample - loss: 0.0461 - acc: 0.9878\n",
      "489/489 [==============================] - 50s 102ms/step - loss: 0.0801 - acc: 0.9786 - val_loss: 0.0459 - val_acc: 0.9878\n",
      "Epoch 9/10\n",
      "12600/12600 [==============================] - 3s 229us/sample - loss: 0.0292 - acc: 0.9925\n",
      "489/489 [==============================] - 51s 105ms/step - loss: 0.0830 - acc: 0.9783 - val_loss: 0.0291 - val_acc: 0.9925\n",
      "Epoch 10/10\n",
      "12600/12600 [==============================] - 3s 215us/sample - loss: 0.0308 - acc: 0.9920\n",
      "489/489 [==============================] - 52s 107ms/step - loss: 0.0804 - acc: 0.9792 - val_loss: 0.0307 - val_acc: 0.9920\n"
     ]
    }
   ],
   "source": [
    "history = model.fit_generator(datagen.flow(X_train,Y_train, batch_size=86),\n",
    "                              epochs = 10, validation_data = (x_val,y_val),\n",
    "                              verbose = 1, steps_per_epoch=X_train.shape[0] // 86\n",
    "                              , callbacks=[learn_rate])\n",
    "model.save(\"rps.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.legend(['train', 'test'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['acc'])\n",
    "plt.plot(history.history['val_acc'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.legend(['train', 'test'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('model.hdh5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('model.pb')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'D:\\\\datasets\\\\MNIST'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_path=\"D:\\\\datasets\\\\MNIST\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "WARNING: The TensorFlow contrib module will not be included in TensorFlow 2.0.\n",
      "For more information, please see:\n",
      "  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md\n",
      "  * https://github.com/tensorflow/addons\n",
      "If you depend on functionality not listed there, please file an issue.\n",
      "\n",
      "WARNING:tensorflow:This model was compiled with a Keras optimizer (<tensorflow.python.keras.optimizers.RMSprop object at 0x000001C71755FB00>) but is being saved in TensorFlow format with `save_weights`. The model's weights will be saved, but unlike with TensorFlow optimizers in the TensorFlow format the optimizer's state will not be saved.\n",
      "\n",
      "Consider using a TensorFlow optimizer from `tf.train`.\n",
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\keras\\engine\\network.py:1436: update_checkpoint_state (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.train.CheckpointManager to manage checkpoints rather than manually editing the Checkpoint proto.\n",
      "WARNING:tensorflow:Model was compiled with an optimizer, but the optimizer is not from `tf.train` (e.g. `tf.train.AdagradOptimizer`). Only the serving graph was exported. The train and evaluate graphs were not added to the SavedModel.\n",
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\saved_model\\signature_def_utils_impl.py:205: build_tensor_info (from tensorflow.python.saved_model.utils_impl) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This function will only be available through the v1 compatibility library as tf.compat.v1.saved_model.utils.build_tensor_info or tf.compat.v1.saved_model.build_tensor_info.\n",
      "INFO:tensorflow:Signatures INCLUDED in export for Classify: None\n",
      "INFO:tensorflow:Signatures INCLUDED in export for Regress: None\n",
      "INFO:tensorflow:Signatures INCLUDED in export for Predict: ['serving_default']\n",
      "INFO:tensorflow:Signatures INCLUDED in export for Train: None\n",
      "INFO:tensorflow:Signatures INCLUDED in export for Eval: None\n",
      "INFO:tensorflow:No assets to save.\n",
      "INFO:tensorflow:No assets to write.\n",
      "INFO:tensorflow:SavedModel written to: D:\\datasets\\MNIST\\1581186354\\saved_model.pb\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "b'D:\\\\datasets\\\\MNIST\\\\1581186354'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.contrib.saved_model.save_keras_model(model,model_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\lite\\python\\convert_saved_model.py:61: load (from tensorflow.python.saved_model.loader_impl) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This function will only be available through the v1 compatibility library as tf.compat.v1.saved_model.loader.load or tf.compat.v1.saved_model.load. There will be a new function for importing SavedModels in Tensorflow 2.0.\n",
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\training\\saver.py:1266: checkpoint_exists (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use standard file APIs to check for files with this prefix.\n",
      "INFO:tensorflow:Restoring parameters from D:\\datasets\\MNIST\\1581186354\\variables\\variables\n",
      "INFO:tensorflow:The given SavedModel MetaGraphDef contains SignatureDefs with the following keys: {'__saved_model_init_op', 'serving_default'}\n",
      "INFO:tensorflow:input tensors info: \n",
      "INFO:tensorflow:Tensor's key in saved_model's tensor_map: conv2d_input\n",
      "INFO:tensorflow: tensor name: conv2d_input:0, shape: (-1, 28, 28, 1), type: DT_FLOAT\n",
      "INFO:tensorflow:output tensors info: \n",
      "INFO:tensorflow:Tensor's key in saved_model's tensor_map: dense_1\n",
      "INFO:tensorflow: tensor name: dense_1/Softmax:0, shape: (-1, 10), type: DT_FLOAT\n",
      "INFO:tensorflow:Restoring parameters from D:\\datasets\\MNIST\\1581186354\\variables\\variables\n",
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\lite\\python\\convert_saved_model.py:275: convert_variables_to_constants (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.compat.v1.graph_util.convert_variables_to_constants\n",
      "WARNING:tensorflow:From C:\\Users\\DELL\\Miniconda3\\lib\\site-packages\\tensorflow\\python\\framework\\graph_util_impl.py:245: extract_sub_graph (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.compat.v1.graph_util.extract_sub_graph\n",
      "INFO:tensorflow:Froze 12 variables.\n",
      "INFO:tensorflow:Converted 12 variables to const ops.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "358588"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "saved_model_dir = \"D:\\\\datasets\\\\MNIST\\\\1581186354\"\n",
    "converter = tf.lite.TFLiteConverter.from_saved_model(saved_model_dir)\n",
    "tflite_model = converter.convert()\n",
    "open(\"/tmp/model.tflite\", \"wb\").write(tflite_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}