Merge branch 'main' into add_docker

shroominic · Aug 6, 2023 · 86c866d · 86c866d
2 parents 724817f + 0478187
commit 86c866d
Show file tree

Hide file tree

Showing 36 changed files with 1,133 additions and 682 deletions.
diff --git a/.env.example b/.env.example
@@ -1,10 +1,14 @@
-OPENAI_API_KEY= # your openai api key (required)
-CODEBOX_API_KEY= # your codebox api key (optional, required for production)
-VERBOSE=False # set to True to enable verbose logging
+# (required)
+OPENAI_API_KEY=
+# (optional, required for production)
+# CODEBOX_API_KEY=
+# (set True to enable logging)
+VERBOSE=False
 
-STORAGE_TYPE="redis" # "file" or "redis"
+
+STORAGE_TYPE="file" # "file" or "redis"
 
 RD_HOST=
 RD_PORT=
 
-SESSION_STORAGE_FOLDER="session_storage"
+SESSION_STORAGE_FOLDER="session_storage"
diff --git a/.github/FUNDING.yml b/.github/FUNDING.yml
@@ -1,6 +1,6 @@
 # These are supported funding model platforms
 
-github: # Replace with up to 4 GitHub Sponsors-enabled usernames e.g., [user1, user2]
+github: shroominic
 patreon: # Replace with a single Patreon username
 open_collective: # Replace with a single Open Collective username
 ko_fi: shroominic

diff --git a/.github/workflows/code-check.yml b/.github/workflows/code-check.yml
@@ -0,0 +1,16 @@
+name: code-check
+
+on: [push]
+
+jobs:
+  pre-commit:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2
+    - uses: actions/setup-python@v2
+      with:
+        python-version: 3.9
+    - name: Install pre-commit
+      run: pip install pre-commit
+    - name: Run pre-commit
+      run: pre-commit run --all-files
diff --git a/.github/workflows/python-publish.yml b/.github/workflows/python-publish.yml
@@ -0,0 +1,31 @@
+name: Upload Python Package
+
+on:
+  release:
+    types: [published]
+
+permissions:
+  contents: read
+
+jobs:
+  deploy:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: Set up Python
+      uses: actions/setup-python@v3
+      with:
+        python-version: '3.x'
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install build
+    - name: Build package
+      run: python -m build
+    - name: Publish package
+      uses: pypa/gh-action-pypi-publish@27b31702a0e7fc50959f5ad993c78deac1bdfc29
+      with:
+        user: __token__
+        password: ${{ secrets.PYPI_API_TOKEN }}
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -0,0 +1,27 @@
+repos:
+-   repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v2.3.0
+    hooks:
+    -   id: check-yaml
+    -   id: end-of-file-fixer
+    -   id: trailing-whitespace
+-   repo: https://github.com/psf/black
+    rev: 23.7.0
+    hooks:
+    -   id: black
+-   repo: https://github.com/pre-commit/mirrors-mypy
+    rev: v1.4.1
+    hooks:
+    -   id: mypy
+        args: [--ignore-missing-imports, --follow-imports=skip]
+        additional_dependencies: [types-requests]
+-   repo: https://github.com/pre-commit/mirrors-isort
+    rev: v5.9.3
+    hooks:
+    -   id: isort
+        args: [--profile=black]
+-   repo: https://github.com/PYCQA/flake8
+    rev: 6.1.0
+    hooks:
+    -   id: flake8
+        args: [--max-line-length=88]
diff --git a/README.md b/README.md
@@ -19,9 +19,12 @@ You can run everything local except the LLM using your own OpenAI API Key.
 Get your OpenAI API Key [here](https://platform.openai.com/account/api-keys) and install the package.
 
 ```bash
-pip install codeinterpreterapi
+pip install "codeinterpreterapi[all]"
 ```
 
+Everything for local experiments are installed with the `all` extra.
+For deployments, you can use `pip install codeinterpreterapi` instead which does not install the additional dependencies.
+
 ## Usage
 
 Make sure to set the `OPENAI_API_KEY` environment variable (or use a `.env` file)
@@ -56,7 +59,7 @@ if __name__ == "__main__":
 
 ```
 
-![Bitcoin YTD](https://github.com/shroominic/codeinterpreter-api/blob/main/examples/assets/bitcoin_chart.png?raw=true)  
+![Bitcoin YTD](https://github.com/shroominic/codeinterpreter-api/blob/main/examples/assets/bitcoin_chart.png?raw=true)
 Bitcoin YTD Chart Output
 
 ## Dataset Analysis
@@ -91,7 +94,7 @@ if __name__ == "__main__":
     asyncio.run(main())
 ```
 
-![Iris Dataset Analysis](https://github.com/shroominic/codeinterpreter-api/blob/main/examples/assets/iris_analysis.png?raw=true)  
+![Iris Dataset Analysis](https://github.com/shroominic/codeinterpreter-api/blob/main/examples/assets/iris_analysis.png?raw=true)
 Iris Dataset Analysis Output
 
 ## Production
@@ -127,7 +130,7 @@ streamlit run frontend/app.py
 ## Contact
 
 You can contact me at [contact@shroominic.com](mailto:contact@shroominic.com).
-But I prefer to use [Twitter](https://twitter.com/shroominic) or [Discord](https://gptassistant.app/community) DMs.
+But I prefer to use [Twitter](https://twitter.com/shroominic) or [Discord](https://discord.gg/QYzBtq37) DMs.
 
 ## Support this project
 

diff --git a/codeinterpreterapi/__init__.py b/codeinterpreterapi/__init__.py
@@ -1,2 +1,4 @@
+from codeinterpreterapi.schema import File
 from codeinterpreterapi.session import CodeInterpreterSession
-from codeinterpreterapi.schema import File
+
+__all__ = ["CodeInterpreterSession", "File"]
diff --git a/codeinterpreterapi/agents/__init__.py b/codeinterpreterapi/agents/__init__.py
@@ -0,0 +1,3 @@
+from .functions_agent import OpenAIFunctionsAgent
+
+__all__ = ["OpenAIFunctionsAgent"]
diff --git a/codeinterpreterapi/agents/custom_agent.py b/codeinterpreterapi/agents/custom_agent.py
@@ -0,0 +1,2 @@
+# TODO: override some methods of the ConversationalAgent class
+# to improve the agent's performance
diff --git a/codeinterpreterapi/chains/functions_agent.py → codeinterpreterapi/agents/functions_agent.py b/codeinterpreterapi/chains/functions_agent.py → codeinterpreterapi/agents/functions_agent.py
@@ -1,25 +1,21 @@
 """
 Module implements an agent that uses OpenAI's APIs function enabled API.
 
-This file is a modified version of the original file 
+This file is a modified version of the original file
 from langchain/agents/openai_functions_agent/base.py.
 Credits go to the original authors :)
 """
 
-
 import json
 from dataclasses import dataclass
 from json import JSONDecodeError
 from typing import Any, List, Optional, Sequence, Tuple, Union
 
-from pydantic import root_validator
-
 from langchain.agents import BaseSingleActionAgent
 from langchain.base_language import BaseLanguageModel
 from langchain.callbacks.base import BaseCallbackManager
 from langchain.callbacks.manager import Callbacks
 from langchain.chat_models.openai import ChatOpenAI
-from langchain.schema import BasePromptTemplate
 from langchain.prompts.chat import (
     BaseMessagePromptTemplate,
     ChatPromptTemplate,
@@ -31,13 +27,14 @@
     AgentFinish,
     AIMessage,
     BaseMessage,
+    BasePromptTemplate,
     FunctionMessage,
     OutputParserException,
-    HumanMessage,
     SystemMessage,
 )
 from langchain.tools import BaseTool
 from langchain.tools.convert_to_openai import format_tool_to_openai_function
+from pydantic import root_validator
 
 
 @dataclass

diff --git a/codeinterpreterapi/chains/__init__.py b/codeinterpreterapi/chains/__init__.py
@@ -0,0 +1,9 @@
+from .extract_code import extract_python_code
+from .modifications_check import get_file_modifications
+from .rm_dl_link import remove_download_link
+
+__all__ = [
+    "extract_python_code",
+    "get_file_modifications",
+    "remove_download_link",
+]
diff --git a/codeinterpreterapi/chains/extract_code.py b/codeinterpreterapi/chains/extract_code.py
@@ -0,0 +1,41 @@
+from langchain.base_language import BaseLanguageModel
+from langchain.chat_models.anthropic import ChatAnthropic
+
+
+# TODO: make async
+def extract_python_code(
+    text: str,
+    llm: BaseLanguageModel,
+    retry: int = 2,
+):
+    pass
+
+
+async def test():
+    llm = ChatAnthropic(model="claude-1.3")  # type: ignore
+
+    code = """
+        import matplotlib.pyplot as plt
+
+        x = list(range(1, 11))
+        y = [29, 39, 23, 32, 4, 43, 43, 23, 43, 77]
+
+        plt.plot(x, y, marker='o')
+        plt.xlabel('Index')
+        plt.ylabel('Value')
+        plt.title('Data Plot')
+
+        plt.show()
+        """
+
+    print(extract_python_code(code, llm))
+
+
+if __name__ == "__main__":
+    import asyncio
+
+    import dotenv
+
+    dotenv.load_dotenv()
+
+    asyncio.run(test())
diff --git a/codeinterpreterapi/chains/modifications_check.py b/codeinterpreterapi/chains/modifications_check.py
@@ -2,10 +2,9 @@
 from typing import List, Optional
 
 from langchain.base_language import BaseLanguageModel
-from langchain.chat_models.openai import ChatOpenAI
-from langchain.schema import AIMessage, OutputParserException
+from langchain.chat_models.anthropic import ChatAnthropic
 
-from codeinterpreterapi.prompts import determine_modifications_function, determine_modifications_prompt
+from codeinterpreterapi.prompts import determine_modifications_prompt
 
 
 async def get_file_modifications(
@@ -15,44 +14,45 @@ async def get_file_modifications(
 ) -> Optional[List[str]]:
     if retry < 1:
         return None
-    messages = determine_modifications_prompt.format_prompt(code=code).to_messages()
-    message = await llm.apredict_messages(messages, functions=[determine_modifications_function])
 
-    if not isinstance(message, AIMessage):
-        raise OutputParserException("Expected an AIMessage")
+    prompt = determine_modifications_prompt.format(code=code)
 
-    function_call = message.additional_kwargs.get("function_call", None)
+    result = await llm.apredict(prompt, stop="```")
 
-    if function_call is None:
+    try:
+        result = json.loads(result)
+    except json.JSONDecodeError:
+        result = ""
+    if not result or not isinstance(result, dict) or "modifications" not in result:
         return await get_file_modifications(code, llm, retry=retry - 1)
-    else:
-        function_call = json.loads(function_call["arguments"])
-        return function_call["modifications"]
+    return result["modifications"]
 
 
 async def test():
-    llm = ChatOpenAI(model="gpt-3.5")  # type: ignore
+    llm = ChatAnthropic(model="claude-1.3")  # type: ignore
 
     code = """
-    import matplotlib.pyplot as plt
+        import matplotlib.pyplot as plt
 
-    x = list(range(1, 11))
-    y = [29, 39, 23, 32, 4, 43, 43, 23, 43, 77]
+        x = list(range(1, 11))
+        y = [29, 39, 23, 32, 4, 43, 43, 23, 43, 77]
 
-    plt.plot(x, y, marker='o')
-    plt.xlabel('Index')
-    plt.ylabel('Value')
-    plt.title('Data Plot')
+        plt.plot(x, y, marker='o')
+        plt.xlabel('Index')
+        plt.ylabel('Value')
+        plt.title('Data Plot')
 
-    plt.show()
-    """
+        plt.show()
+        """
 
     print(await get_file_modifications(code, llm))
 
 
 if __name__ == "__main__":
     import asyncio
-    from dotenv import load_dotenv
-    load_dotenv()
+
+    import dotenv
+
+    dotenv.load_dotenv()
 
     asyncio.run(test())
diff --git a/...rpreterapi/chains/remove_download_link.py → codeinterpreterapi/chains/rm_dl_link.py b/...rpreterapi/chains/remove_download_link.py → codeinterpreterapi/chains/rm_dl_link.py
@@ -9,7 +9,9 @@ async def remove_download_link(
     input_response: str,
     llm: BaseLanguageModel,
 ) -> str:
-    messages = remove_dl_link_prompt.format_prompt(input_response=input_response).to_messages()
+    messages = remove_dl_link_prompt.format_prompt(
+        input_response=input_response
+    ).to_messages()
     message = await llm.apredict_messages(messages)
 
     if not isinstance(message, AIMessage):
@@ -21,14 +23,18 @@ async def remove_download_link(
 async def test():
     llm = ChatOpenAI(model="gpt-3.5-turbo-0613")  # type: ignore
 
-    example = "I have created the plot to your dataset.\n\nLink to the file [here](sandbox:/plot.png)."
-
+    example = (
+        "I have created the plot to your dataset.\n\n"
+        "Link to the file [here](sandbox:/plot.png)."
+    )
     print(await remove_download_link(example, llm))
 
 
 if __name__ == "__main__":
     import asyncio
+
     from dotenv import load_dotenv
+
     load_dotenv()
 
     asyncio.run(test())
diff --git a/codeinterpreterapi/config.py b/codeinterpreterapi/config.py
@@ -1,7 +1,8 @@
-from pydantic import BaseSettings
-from dotenv import load_dotenv
 from typing import Optional
 
+from dotenv import load_dotenv
+from pydantic import BaseSettings
+
 # .env file
 load_dotenv(dotenv_path="./.env")
 

diff --git a/codeinterpreterapi/prompts/__init__.py b/codeinterpreterapi/prompts/__init__.py
@@ -1,3 +1,9 @@
+from .modifications_check import determine_modifications_prompt
+from .remove_dl_link import remove_dl_link_prompt
 from .system_message import system_message as code_interpreter_system_message
-from .modifications_check import determine_modifications_function, determine_modifications_prompt
-from .remove_dl_link import remove_dl_link_prompt
+
+__all__ = [
+    "determine_modifications_prompt",
+    "remove_dl_link_prompt",
+    "code_interpreter_system_message",
+]
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,3 @@
		from .functions_agent import OpenAIFunctionsAgent

		__all__ = ["OpenAIFunctionsAgent"]
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		# TODO: override some methods of the ConversationalAgent class
		# to improve the agent's performance