"<a href=\"https://colab.research.google.com/github/microsoft/autogen/blob/main/notebook/agentchat_qdrant_RetrieveChat.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id=\"toc\"></a>\n",
"# Using RetrieveChat with Qdrant for Retrieve Augmented Code Generation and Question Answering\n",
"\n",
"[Qdrant](https://qdrant.tech/) is a high-performance vector search engine/database.\n",
"\n",
"This notebook demonstrates the usage of `QdrantRetrieveUserProxyAgent` for RAG, based on [agentchat_RetrieveChat.ipynb](https://colab.research.google.com/github/microsoft/autogen/blob/main/notebook/agentchat_RetrieveChat.ipynb).\n",
"\n",
"\n",
"RetrieveChat is a conversational system for retrieve augmented code generation and question answering. In this notebook, we demonstrate how to utilize RetrieveChat to generate code and answer questions based on customized documentations that are not present in the LLM's training dataset. RetrieveChat uses the `RetrieveAssistantAgent` and `QdrantRetrieveUserProxyAgent`, which is similar to the usage of `AssistantAgent` and `UserProxyAgent` in other notebooks (e.g., [Automated Task Solving with Code Generation, Execution & Debugging](https://github.com/microsoft/autogen/blob/main/notebook/agentchat_auto_feedback_from_code_execution.ipynb)).\n",
"\n",
"We'll demonstrate usage of RetrieveChat with Qdrant for code generation and question answering w/ human feedback.\n",
"\n",
"\n",
"## Requirements\n",
"\n",
"AutoGen requires `Python>=3.8`. To run this notebook example, please install the [retrievechat] option.\n",
"The [`config_list_from_json`](https://microsoft.github.io/autogen/docs/reference/oai/openai_utils#config_list_from_json) function loads a list of configurations from an environment variable or a json file.\n"
"print(\"models to use: \", [config_list[i][\"model\"] for i in range(len(config_list))])"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"It first looks for environment variable \"OAI_CONFIG_LIST\" which needs to be a valid json string. If that variable is not found, it then looks for a json file named \"OAI_CONFIG_LIST\". It filters the configs by models (you can filter by other keys as well). Only the gpt-4 and gpt-3.5-turbo models are kept in the list based on the filter condition.\n",
"\n",
"The config list looks like the following:\n",
"```python\n",
"config_list = [\n",
" {\n",
" 'model': 'gpt-4',\n",
" 'api_key': '<your OpenAI API key here>',\n",
" },\n",
" {\n",
" 'model': 'gpt-4',\n",
" 'api_key': '<your Azure OpenAI API key here>',\n",
"We start by initializing the `RetrieveAssistantAgent` and `QdrantRetrieveUserProxyAgent`. The system message needs to be set to \"You are a helpful assistant.\" for RetrieveAssistantAgent. The detailed instructions are given in the user message. Later we will use the `QdrantRetrieveUserProxyAgent.generate_init_prompt` to combine the instructions and a retrieval augmented generation task for an initial prompt to be sent to the LLM assistant.\n",
"<!-- [](https://gitter.im/FLAMLer/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge) -->\n",
":fire: Heads-up: We have migrated [AutoGen](https://microsoft.github.io/autogen/) into a dedicated [github repository](https://github.com/microsoft/autogen). Alongside this move, we have also launched a dedicated [Discord](https://discord.gg/pAbnFJrkgZ) server and a [website](https://microsoft.github.io/autogen/) for comprehensive documentation.\n",
":fire: FLAML is highlighted in OpenAI's [cookbook](https://github.com/openai/openai-cookbook#related-resources-from-around-the-web).\n",
"\n",
":fire: [autogen](https://microsoft.github.io/autogen/) is released with support for ChatGPT and GPT-4, based on [Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference](https://arxiv.org/abs/2303.04673).\n",
"\n",
":fire: FLAML supports Code-First AutoML & Tuning – Private Preview in [Microsoft Fabric Data Science](https://learn.microsoft.com/en-us/fabric/data-science/).\n",
"* FLAML enables building next-gen GPT-X applications based on multi-agent conversations with minimal effort. It simplifies the orchestration, automation and optimization of a complex GPT-X workflow. It maximizes the performance of GPT-X models and augments their weakness.\n",
"* For common machine learning tasks like classification and regression, it quickly finds quality models for user-provided data with low computational resources. It is easy to customize or extend. Users can find their desired customizability from a smooth range.\n",
"* It supports fast and economical automatic tuning (e.g., inference hyperparameters for foundation models, configurations in MLOps/LMOps workflows, pipelines, mathematical/statistical models, algorithms, computing experiments, software configurations), capable of handling large search space with heterogeneous evaluation cost and complex constraints/guidance/early stopping.\n",
"FLAML is powered by a series of [research studies](https://microsoft.github.io/FLAML/docs/Research/) from Microsoft Research and collaborators such as Penn State University, Stevens Institute of Technology, University of Washington, and University of Waterloo.\n",
"Minimal dependencies are installed without extra options. You can install extra options based on the feature you need. For example, use the following to install the dependencies needed by the [`autogen`](https://microsoft.github.io/autogen/) package.\n",
"* (New) The [autogen](https://microsoft.github.io/autogen/) package enables the next-gen GPT-X applications with a generic multi-agent conversation framework.\n",
"It offers customizable and conversable agents which integrate LLMs, tools and human.\n",
"By automating chat among multiple capable agents, one can easily make them collectively perform tasks autonomously or with human feedback, including tasks that require using tools via code. For example,\n",
"Autogen also helps maximize the utility out of the expensive LLMs such as ChatGPT and GPT-4. It offers a drop-in replacement of `openai.Completion` or `openai.ChatCompletion` with powerful functionalites like tuning, caching, templating, filtering. For example, you can optimize generations by LLM with your own tuning data, success metrics and budgets.\n",
"* You can restrict the learners and use FLAML as a fast hyperparameter tuning\n",
"tool for XGBoost, LightGBM, Random Forest etc. or a [customized learner](https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML#estimator-and-search-space).\n",
"* You can also run generic hyperparameter tuning for a [custom function](https://microsoft.github.io/FLAML/docs/Use-Cases/Tune-User-Defined-Function).\n",
"* [Zero-shot AutoML](https://microsoft.github.io/FLAML/docs/Use-Cases/Zero-Shot-AutoML) allows using the existing training API from lightgbm, xgboost etc. while getting the benefit of AutoML in choosing high-performance hyperparameter configurations per task.\n",
"- ML.NET documentation and tutorials for [Model Builder](https://learn.microsoft.com/dotnet/machine-learning/tutorials/predict-prices-with-model-builder), [ML.NET CLI](https://learn.microsoft.com/dotnet/machine-learning/tutorials/sentiment-analysis-cli), and [AutoML API](https://learn.microsoft.com/dotnet/machine-learning/how-to-guides/how-to-use-the-automl-api).\n",
"\n",
"## Contributing\n",
"\n",
"This project welcomes contributions and suggestions. Most contributions require you to agree to a\n",
"Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us\n",
"the rights to use your contribution. For details, visit <https://cla.opensource.microsoft.com>.\n",
"\n",
"If you are new to GitHub [here](https://help.github.com/categories/collaborating-with-issues-and-pull-requests/) is a detailed help source on getting involved with development on GitHub.\n",
"# Research\n",
"\n",
"For technical details, please check our research publications.\n",
"\n",
"* [FLAML: A Fast and Lightweight AutoML Library](https://www.microsoft.com/en-us/research/publication/flaml-a-fast-and-lightweight-automl-library/). Chi Wang, Qingyun Wu, Markus Weimer, Erkang Zhu. MLSys 2021.\n",
"\n",
"```bibtex\n",
"@inproceedings{wang2021flaml,\n",
" title={FLAML: A Fast and Lightweight AutoML Library},\n",
" author={Chi Wang and Qingyun Wu and Markus Weimer and Erkang Zhu},\n",
" year={2021},\n",
" booktitle={MLSys},\n",
"}\n",
"```\n",
"\n",
"* [Frugal Optimization for Cost-related Hyperparameters](https://arxiv.org/abs/2005.01571). Qingyun Wu, Chi Wang, Silu Huang. AAAI 2021.\n",
"\n",
"```bibtex\n",
"@inproceedings{wu2021cfo,\n",
" title={Frugal Optimization for Cost-related Hyperparameters},\n",
" author={Qingyun Wu and Chi Wang and Silu Huang},\n",
"* [An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models](https://aclanthology.org/2021.acl-long.178.pdf). Susan Xueqing Liu, Chi Wang. ACL 2021.\n",
"* [ChaCha for Online AutoML](https://www.microsoft.com/en-us/research/publication/chacha-for-online-automl/). Qingyun Wu, Chi Wang, John Langford, Paul Mineiro and Marco Rossi. ICML 2021.\n",
"* [Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference](https://arxiv.org/abs/2303.04673). Chi Wang, Susan Xueqing Liu, Ahmed H. Awadallah. ArXiv preprint arXiv:2303.04673 (2023).\n",
"\n",
"```bibtex\n",
"@inproceedings{wang2023EcoOptiGen,\n",
" title={Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference},\n",
" author={Chi Wang and Susan Xueqing Liu and Ahmed H. Awadallah},\n",
"* [An Empirical Study on Challenging Math Problem Solving with GPT-4](https://arxiv.org/abs/2306.01337). Yiran Wu, Feiran Jia, Shaokun Zhang, Hangyu Li, Erkang Zhu, Yue Wang, Yin Tat Lee, Richard Peng, Qingyun Wu, Chi Wang. ArXiv preprint arXiv:2306.01337 (2023).\n",
"\n",
"```bibtex\n",
"@inproceedings{wu2023empirical,\n",
" title={An Empirical Study on Challenging Math Problem Solving with GPT-4},\n",
" author={Yiran Wu and Feiran Jia and Shaokun Zhang and Hangyu Li and Erkang Zhu and Yue Wang and Yin Tat Lee and Richard Peng and Qingyun Wu and Chi Wang},\n",
"Based on the context provided, which is about the FLAML library, there is no direct reference to a function specifically called `tune_automl`. However, FLAML does offer functionality for automated machine learning (AutoML) and hyperparameter tuning.\n",
"\n",
"The closest reference to an AutoML tuning operation in the given context is shown in the Quickstart section, which demonstrates how to use FLAML as a scikit-learn style estimator for machine learning tasks like classification and regression. It does talk about automated machine learning and tuning, but doesn't mention a function `tune_automl` by name.\n",
"\n",
"If you are looking for a way to perform tuning with FLAML, the context indicates you can use the `tune` module to run generic hyperparameter tuning for a custom function, as demonstrated in the Quickstart section:\n",
"This is not called `tune_automl` but rather just `tune.run`.\n",
"\n",
"If you need confirmation on whether a function called `tune_automl` specifically exists, the FLAML documentation or its API reference should be checked. If documentation is not enough to confirm and you require to look into the actual code or a structured list of functionalities provided by FLAML, that information isn't available in the given context.\n",
"\n",
"In that case, the instruction should be: `UPDATE CONTEXT`.\n",
"* [FLAML: A Fast and Lightweight AutoML Library](https://www.microsoft.com/en-us/research/publication/flaml-a-fast-and-lightweight-automl-library/). Chi Wang, Qingyun Wu, Markus Weimer, Erkang Zhu. MLSys 2021.\n",
"* [An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models](https://aclanthology.org/2021.acl-long.178.pdf). Susan Xueqing Liu, Chi Wang. ACL 2021.\n",
"* [ChaCha for Online AutoML](https://www.microsoft.com/en-us/research/publication/chacha-for-online-automl/). Qingyun Wu, Chi Wang, John Langford, Paul Mineiro and Marco Rossi. ICML 2021.\n",
"* [Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference](https://arxiv.org/abs/2303.04673). Chi Wang, Susan Xueqing Liu, Ahmed H. Awadallah. ArXiv preprint arXiv:2303.04673 (2023).\n",
"* [An Empirical Study on Challenging Math Problem Solving with GPT-4](https://arxiv.org/abs/2306.01337). Yiran Wu, Feiran Jia, Shaokun Zhang, Hangyu Li, Erkang Zhu, Yue Wang, Yin Tat Lee, Richard Peng, Qingyun Wu, Chi Wang. ArXiv preprint arXiv:2306.01337 (2023).\n",
" title={An Empirical Study on Challenging Math Problem Solving with GPT-4},\n",
" author={Yiran Wu and Feiran Jia and Shaokun Zhang and Hangyu Li and Erkang Zhu and Yue Wang and Yin Tat Lee and Richard Peng and Qingyun Wu and Chi Wang},\n",
"<!-- [](https://gitter.im/FLAMLer/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge) -->\n",
":fire: Heads-up: We have migrated [AutoGen](https://microsoft.github.io/autogen/) into a dedicated [github repository](https://github.com/microsoft/autogen). Alongside this move, we have also launched a dedicated [Discord](https://discord.gg/pAbnFJrkgZ) server and a [website](https://microsoft.github.io/autogen/) for comprehensive documentation.\n",
":fire: [autogen](https://microsoft.github.io/autogen/) is released with support for ChatGPT and GPT-4, based on [Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference](https://arxiv.org/abs/2303.04673).\n",
"* FLAML enables building next-gen GPT-X applications based on multi-agent conversations with minimal effort. It simplifies the orchestration, automation and optimization of a complex GPT-X workflow. It maximizes the performance of GPT-X models and augments their weakness.\n",
"* For common machine learning tasks like classification and regression, it quickly finds quality models for user-provided data with low computational resources. It is easy to customize or extend. Users can find their desired customizability from a smooth range.\n",
"* It supports fast and economical automatic tuning (e.g., inference hyperparameters for foundation models, configurations in MLOps/LMOps workflows, pipelines, mathematical/statistical models, algorithms, computing experiments, software configurations), capable of handling large search space with heterogeneous evaluation cost and complex constraints/guidance/early stopping.\n",
"FLAML is powered by a series of [research studies](https://microsoft.github.io/FLAML/docs/Research/) from Microsoft Research and collaborators such as Penn State University, Stevens Institute of Technology, University of Washington, and University of Waterloo.\n",
"Minimal dependencies are installed without extra options. You can install extra options based on the feature you need. For example, use the following to install the dependencies needed by the [`autogen`](https://microsoft.github.io/autogen/) package.\n",
"* (New) The [autogen](https://microsoft.github.io/autogen/) package enables the next-gen GPT-X applications with a generic multi-agent conversation framework.\n",
"It offers customizable and conversable agents which integrate LLMs, tools and human.\n",
"By automating chat among multiple capable agents, one can easily make them collectively perform tasks autonomously or with human feedback, including tasks that require using tools via code. For example,\n",
"Autogen also helps maximize the utility out of the expensive LLMs such as ChatGPT and GPT-4. It offers a drop-in replacement of `openai.Completion` or `openai.ChatCompletion` with powerful functionalites like tuning, caching, templating, filtering. For example, you can optimize generations by LLM with your own tuning data, success metrics and budgets.\n",
"* You can restrict the learners and use FLAML as a fast hyperparameter tuning\n",
"tool for XGBoost, LightGBM, Random Forest etc. or a [customized learner](https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML#estimator-and-search-space).\n",
"* You can also run generic hyperparameter tuning for a [custom function](https://microsoft.github.io/FLAML/docs/Use-Cases/Tune-User-Defined-Function).\n",
"* [Zero-shot AutoML](https://microsoft.github.io/FLAML/docs/Use-Cases/Zero-Shot-AutoML) allows using the existing training API from lightgbm, xgboost etc. while getting the benefit of AutoML in choosing high-performance hyperparameter configurations per task.\n",
"- ML.NET documentation and tutorials for [Model Builder](https://learn.microsoft.com/dotnet/machine-learning/tutorials/predict-prices-with-model-builder), [ML.NET CLI](https://learn.microsoft.com/dotnet/machine-learning/tutorials/sentiment-analysis-cli), and [AutoML API](https://learn.microsoft.com/dotnet/machine-learning/how-to-guides/how-to-use-the-automl-api).\n",
"If you are new to GitHub [here](https://help.github.com/categories/collaborating-with-issues-and-pull-requests/) is a detailed help source on getting involved with development on GitHub.\n",
"The author of FLAML is Chi Wang, along with other collaborators including Qingyun Wu, Markus Weimer, Erkang Zhu, Silu Huang, Amin Saied, Susan Xueqing Liu, John Langford, Paul Mineiro, Marco Rossi, Moe Kayali, Shaokun Zhang, Feiran Jia, Yiran Wu, Hangyu Li, Yue Wang, Yin Tat Lee, Richard Peng, and Ahmed H. Awadallah, as indicated in the provided references for FLAML's research publications.\n",
"ChatResult(chat_id=None, chat_history=[{'content': 'You\\'re a retrieve augmented coding assistant. You answer user\\'s questions based on your own knowledge and the\\ncontext provided by the user.\\nIf you can\\'t answer the question with or without the current context, you should reply exactly `UPDATE CONTEXT`.\\nFor code generation, you must obey the following rules:\\nRule 1. You MUST NOT install any packages because all the packages needed are already installed.\\nRule 2. You must follow the formats below to write your code:\\n```language\\n# your code\\n```\\n\\nUser\\'s question is: Who is the author of FLAML?\\n\\nContext is: # Research\\n\\nFor technical details, please check our research publications.\\n\\n* [FLAML: A Fast and Lightweight AutoML Library](https://www.microsoft.com/en-us/research/publication/flaml-a-fast-and-lightweight-automl-library/). Chi Wang, Qingyun Wu, Markus Weimer, Erkang Zhu. MLSys 2021.\\n\\n```bibtex\\n@inproceedings{wang2021flaml,\\n title={FLAML: A Fast and Lightweight AutoML Library},\\n author={Chi Wang and Qingyun Wu and Markus Weimer and Erkang Zhu},\\n year={2021},\\n booktitle={MLSys},\\n}\\n```\\n\\n* [Frugal Optimization for Cost-related Hyperparameters](https://arxiv.org/abs/2005.01571). Qingyun Wu, Chi Wang, Silu Huang. AAAI 2021.\\n\\n```bibtex\\n@inproceedings{wu2021cfo,\\n title={Frugal Optimization for Cost-related Hyperparameters},\\n author={Qingyun Wu and Chi Wang and Silu Huang},\\n year={2021},\\n booktitle={AAAI},\\n}\\n```\\n\\n* [Economical Hyperparameter Optimization With Blended Search Strategy](https://www.microsoft.com/en-us/research/publication/economical-hyperparameter-optimization-with-blended-search-strategy/). Chi Wang, Qingyun Wu, Silu Huang, Amin Saied. ICLR 2021.\\n\\n```bibtex\\n@inproceedings{wang2021blendsearch,\\n title={Economical Hyperparameter Optimization With Blended Search Strategy},\\n author={Chi Wang and Qingyun Wu and Silu Huang and Amin Saied},\\n year={2021},\\n booktitle={ICLR},\\n}\\n```\\n\\n* [An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models](https://aclanthology.org/2021.acl-long.178.pdf). Susan Xueqing Liu, Chi Wang. ACL 2021.\\n\\n```bibtex\\n@inproceedings{liuwang2021hpolm,\\n title={An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models},\\n author={Susan Xueqing Liu and Chi Wang},\\n year={2021},\\n booktitle={ACL},\\n}\\n```\\n\\n* [ChaCha for Online AutoML](https://www.microsoft.com/en-us/research/publication/chacha-for-online-automl/). Qingyun Wu, Chi Wang, John Langford, Paul Mineiro and Marco Rossi. ICML 2021.\\n\\n```bibtex\\n@inproceedings{wu2021chacha,\\n title={ChaCha for Online AutoML},\\n author={Qingyun Wu and Chi Wang and John Langford and Paul Mineiro and Marco Rossi},\\n year={2021},\\n booktitle={ICML},\\n}\\n```\\n\\n* [Fair AutoML](https://arxiv.org/abs/2111.06495). Qingyun Wu, Chi Wang. ArXiv preprint arXiv:2111.06495 (2021).\\n\\n```bibtex\\n@inproceedings{wuwang2021fairautoml,\\n title={Fair AutoML},\\n author={Qingyun Wu and Chi Wang},\\n year={2021},\\n booktitle={ArXiv preprint arXiv:2111.06495},\\n}\\n```\\n\\n* [Mining Robust Default Configurations for Resource-constrained AutoML](https://arxiv.org/abs/2202.09927). Moe Kayali, Chi Wang. ArXiv preprint arXiv:2202.09927 (2022).\\n\\n```bibtex\\n@inproceedings{kayaliwang2022default,\\n title={Mining Robust Default Configurations for Resource-constrained AutoML},\\n author={Moe Kayali and Chi Wang},\\n year={2022},\\n booktitle={ArXiv preprint arXiv:2202.09927},\\n}\\n```\\n\\n* [Targeted Hyperparameter Optimization with Lexicographic Preferences Over Multiple Objectives](https://openreview.net/forum?id=0Ij9_q567Ma). Shaokun Zhang, Feiran Jia, Chi Wang, Qingyun Wu. ICLR 2023 (notable-top-5%).\\n\\n```bibtex\\n@inproceedings{zhang2023targeted,\\n title={Targeted Hyperparameter Optimization with Lexicographic Preferences Over Multiple Objectives},\\n author={Shaokun Zhang an
