0.2.4

What's Changed

• refactor: optimize batch prediction return value by @ZingLix in #138
• feat: tool conversion from langchain by @Azure99 in #139
• feat: Batch inference and Evaluation framework by @Dobiichi-Origami in #140
• cookbook adjustment by @Dobiichi-Origami in #143
• feat[requestor]: add latency statistic by @danielhjz in #141
• feat: support yi-34b-chat by @danielhjz in #145
• feat: support eb plugin & fix plugin by @danielhjz in #144

Full Changelog: 0.2.3...0.2.4

0.2.3

What's Changed

• chore: support more cookbook by @danielhjz in #125
• chore: updatenoteobooks by @danielhjz in #127
• feat: support chat law by @danielhjz in #128
• Add files via upload by @xwdreamer in #126
• add how_to_use_qianfan_operator.ipynb by @Dobiichi-Origami in #133
• feat: base tool framework by @Azure99 in #134
• feat[trainer]: optimize train_config by @danielhjz in #135

New Contributors

• @Azure99 made their first contribution in #134

Full Changelog: 0.2.2...0.2.3

0.2.2

What's Changed

• feat: Dataset v1.3 by @Dobiichi-Origami in #115
• Chore update cookbook by @Dobiichi-Origami in #114
• feat: add supported by @danielhjz in #116
• add dataset feature introduction in Dataset Readme by @Dobiichi-Origami in #118
• fix: add cookbook qianfan install command by @danielhjz in #119
• feat: evaluation resource api by @Dobiichi-Origami in #117
• Update README.md by @xwdreamer in #121
• feat: add batch_run_on_qianfan on Model by @Dobiichi-Origami in #120
• feat: trainer 0.2.2 by @danielhjz in #122
• Fix api error by @danielhjz in #123

New Contributors

• @xwdreamer made their first contribution in #121

Full Changelog: 0.2.1...0.2.2

0.2.1

What's Changed

• bugfix: log can not get correct lineno by @ZingLix in #92
• introduce langchain_core by @Dobiichi-Origami in #104
• feat: support XuanYuan-70B, QfLlama-13B, SQLCoder-7B, CodeLlama-7B by @ZingLix in #106
• feat: support two different dataset format by @Dobiichi-Origami in #93
• feat: add request_id by @danielhjz in #108
• docs: update pypi doc by @ZingLix in #110
• feat: Dataset readme by @Dobiichi-Origami in #88
• feat: add col_renames method by @Dobiichi-Origami in #111
• feat: experimental prompt hub by @ZingLix in #96
• Trainer ops by @Dobiichi-Origami in #112

Full Changelog: 0.2.0...0.2.1

0.2.0

What's Changed

• fix: cookbook version by @danielhjz in #80
• cookbook: add batch prediction and eb search example by @ZingLix in #84
• update: langchain_agent_with_qianfan_llm.ipynb using newest agent syntax by @Dobiichi-Origami in #83
• feat: add prompt resource by @ZingLix in #86
• update: Qianfan dataset optimization with a few new functions by @Dobiichi-Origami in #85
• feat: support trainr.LLMFinetune by @danielhjz in #75
• feat: version->0.2.0 by @danielhjz in #90

Full Changelog: 0.1.4...0.2.0

0.1.4

What's Changed

• fix: update rtd with new readme by @ZingLix in #66
• typo fixes by @Dobiichi-Origami in #68
• Add cookbook for qianfan with baidu elasticsearch on RAG demo. by @wemysschen in #69
• feat: support batch predict by @ZingLix in #70
• refactor: move rate limiter location by @ZingLix in #73
• cookbook: add wandb cookbook by @ZingLix in #74
• cookbook: add deeplake cookbook by @ZingLix in #76
• chore: add input argument(stop/disable_search) for EB/EB4 model by @ZingLix in #77
• cookbook: ajust dir and file tree by @stonekim in #78
• feat：add Dataset v1 by @Dobiichi-Origami in #71
• feat: add apc aipass model supported by @danielhjz in #79

New Contributors

• @wemysschen made their first contribution in #69
• @stonekim made their first contribution in #78

Full Changelog: 0.1.3...0.1.4

0.1.3

What's Changed

• fix: remove output image by @danielhjz in #55
• feat: support tokenizer api by @ZingLix in #54
• feat: Unify authentication methods by @ZingLix in #58
• Data crud by @Dobiichi-Origami in #60
• chore: update README.md by @danielhjz in #63
• feat: update supported params for default models by @danielhjz in #59
• feat: support auto_concat_truncated messages by @danielhjz in #53
• update SFT cookbook with dataset operations by @danielhjz in #62
• refactor: refactor package import usage by @danielhjz in #64

Full Changelog: 0.1.2...0.1.3

0.1.2

What's Changed

• add function_call_agent.ipynb by @Dobiichi-Origami in #27
• fix: add console deploy -> done status cell by @danielhjz in #41
• Total refactoring of global config by @Dobiichi-Origami in #42
• feat: add model eb8k supported and update version->0.1.2 by @danielhjz in #45
• chore: add notebook for text2image by @danielhjz in #47
• update question_answering.ipynb by @Dobiichi-Origami in #49
• chore: update docs by @danielhjz in #50

Full Changelog: 0.1.1...0.1.2

0.1.1

What's Changed

• add release action by @ZingLix in #15
• fix: issue template by @danielhjz in #16
• support global environment variable setting by @Dobiichi-Origami in #20
• doc: Optimize doc pages by @ZingLix in #23
• feat: Add support for render prompt template api by @ZingLix in #18
• implementation of qianfan langchain agent by @Dobiichi-Origami in #17
• Update completion.py [ change api path] 🐛 by @1296680359 in #28
• doc: add quickstart by @ZingLix in #30
• feat: Add support for tokenizer by @ZingLix in #21
• bugfix: asyncio lock not working when not in main thread by @ZingLix in #24
• revert: remove prompt by @ZingLix in #33
• feat: support text2image by @danielhjz in #26
• fix: update service create api by @ZingLix in #35
• feat: add console cook book by @danielhjz in #32
• feat: add private supported by @danielhjz in #25
• chore: add desc & update version by @danielhjz in #38

New Contributors

• @1296680359 made their first contribution in #28

Full Changelog: 0.1.0...0.1.1

0.1.0

Feature：

• 增加ChatCompletion，Completion对于ERNIE-Bot-4的支持，通过model=ERNIE-Bot-4进行调用
• qianfan SDK正式开源
• 增加对 模型管理 API 支持（qianfan.Model）
• 增加对 模型服务发布 API支持 （qianfan.Service）
• 开放SDK API接口文档

Bug修复：

• 修复同时使用model，endpoint导致的模型使用错误的问题

优化：

• 增加 模型SFT调优 任务控制支持
• 限流接口优化，支持传入QPS限制配置