not-again-ai 0.3.1__tar.gz → 0.4.1__tar.gz

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2022-2023 DaveCoDev
+Copyright (c) 2022-2024 DaveCoDev
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/PKG-INFO

@@ -1,12 +1,12 @@
 Metadata-Version: 2.1
 Name: not-again-ai
-Version: 0.3.1
+Version: 0.4.1
 Summary: Designed to once and for all collect all the little things that come up over and over again in AI projects and put them in one place.
 Home-page: https://github.com/DaveCoDev/not-again-ai
 License: MIT
 Author: DaveCoDev
 Author-email: dave.co.dev@gmail.com
-Requires-Python: >=3.10,<3.13
+Requires-Python: >=3.11,<3.13
 Classifier: Development Status :: 3 - Alpha
 Classifier: Intended Audience :: Developers
 Classifier: Intended Audience :: Science/Research
@@ -14,7 +14,6 @@ Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3 :: Only
@@ -22,13 +21,13 @@ Classifier: Typing :: Typed
 Provides-Extra: llm
 Provides-Extra: statistics
 Provides-Extra: viz
-Requires-Dist: jinja2 (>=3.1.2,<4.0.0) ; extra == "llm"
-Requires-Dist: numpy (>=1.26.2,<2.0.0) ; extra == "statistics" or extra == "viz"
-Requires-Dist: openai (>=1.6.1,<2.0.0) ; extra == "llm"
-Requires-Dist: pandas (>=2.1.4,<3.0.0) ; extra == "viz"
-Requires-Dist: scikit-learn (>=1.3.2,<2.0.0) ; extra == "statistics"
-Requires-Dist: scipy (>=1.11.4,<2.0.0) ; extra == "statistics"
-Requires-Dist: seaborn (>=0.13.0,<0.14.0) ; extra == "viz"
+Requires-Dist: numpy (>=1.26.3,<2.0.0) ; extra == "statistics" or extra == "viz"
+Requires-Dist: openai (>=1.10.0,<2.0.0) ; extra == "llm"
+Requires-Dist: pandas (>=2.2.0,<3.0.0) ; extra == "viz"
+Requires-Dist: python-liquid (>=1.10.2,<2.0.0) ; extra == "llm"
+Requires-Dist: scikit-learn (>=1.4.0,<2.0.0) ; extra == "statistics"
+Requires-Dist: scipy (>=1.12.0,<2.0.0) ; extra == "statistics"
+Requires-Dist: seaborn (>=0.13.2,<0.14.0) ; extra == "viz"
 Requires-Dist: tiktoken (>=0.5.2,<0.6.0) ; extra == "llm"
 Project-URL: Documentation, https://github.com/DaveCoDev/not-again-ai
 Project-URL: Repository, https://github.com/DaveCoDev/not-again-ai
@@ -39,7 +38,6 @@ Description-Content-Type: text/markdown
 [![GitHub Actions][github-actions-badge]](https://github.com/johnthagen/python-blueprint/actions)
 [![Packaged with Poetry][poetry-badge]](https://python-poetry.org/)
 [![Nox][nox-badge]](https://github.com/wntrblm/nox)
-[![Code style: Black][black-badge]](https://github.com/psf/black)
 [![Ruff][ruff-badge]](https://github.com/astral-sh/ruff)
 [![Type checked with mypy][mypy-badge]](https://mypy-lang.org/)
 
@@ -56,7 +54,7 @@ Description-Content-Type: text/markdown
 
 # Installation
 
-Requires: Python 3.10, 3.11, or 3.12
+Requires: Python 3.11, or 3.12
 
 Install the entire package from [PyPI](https://pypi.org/project/not-again-ai/) with: 
 
@@ -81,6 +79,15 @@ The base package includes only functions that have minimal external dependencies
 ## LLM (Large Language Model)
 [README](https://github.com/DaveCoDev/not-again-ai/blob/main/readmes/llm.md)
 
+Supports OpenAI chat completions and text embeddings. Includes functions for creating chat completion prompts, token management, and context management.
+
+One example: 
+```python
+client = openai_client()
+messages = [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Hello!"}]
+response = chat_completion(messages=messages, model="gpt-3.5-turbo", max_tokens=100, client=client)["message"]
+>>> "Hello! How can I help you today?"
+```
 
 ## Statistics
 [README](https://github.com/DaveCoDev/not-again-ai/blob/main/readmes/statistics.md)
@@ -266,9 +273,11 @@ To pass arguments to `pytest` through `nox`:
 
 ## Code Style Checking
 
-[PEP 8](https://peps.python.org/pep-0008/) is the universally accepted style guide for
-Python code. PEP 8 code compliance is verified using [Ruff](https://github.com/astral-sh/ruff).
-Ruff is configured in the `[tool.ruff]` section of `pyproject.toml`.
+[PEP 8](https://peps.python.org/pep-0008/) is the universally accepted style guide for Python
+code. PEP 8 code compliance is verified using [Ruff][Ruff]. Ruff is configured in the
+`[tool.ruff]` section of [`pyproject.toml`](./pyproject.toml).
+
+[Ruff]: https://github.com/astral-sh/ruff
 
 To lint code, run:
 
@@ -284,12 +293,7 @@ To automatically fix fixable lint errors, run:
 
 ## Automated Code Formatting
 
-Code is automatically formatted using [black](https://github.com/psf/black). Imports are
-automatically sorted and grouped using [Ruff](https://github.com/astral-sh/ruff).
-
-These tools are configured by:
-
-- [`pyproject.toml`](./pyproject.toml)
+[Ruff][Ruff] is used to automatically format code and group and sort imports.
 
 To automatically format code, run:
 

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/README.md

@@ -3,7 +3,6 @@
 [![GitHub Actions][github-actions-badge]](https://github.com/johnthagen/python-blueprint/actions)
 [![Packaged with Poetry][poetry-badge]](https://python-poetry.org/)
 [![Nox][nox-badge]](https://github.com/wntrblm/nox)
-[![Code style: Black][black-badge]](https://github.com/psf/black)
 [![Ruff][ruff-badge]](https://github.com/astral-sh/ruff)
 [![Type checked with mypy][mypy-badge]](https://mypy-lang.org/)
 
@@ -20,7 +19,7 @@
 
 # Installation
 
-Requires: Python 3.10, 3.11, or 3.12
+Requires: Python 3.11, or 3.12
 
 Install the entire package from [PyPI](https://pypi.org/project/not-again-ai/) with: 
 
@@ -45,6 +44,15 @@ The base package includes only functions that have minimal external dependencies
 ## LLM (Large Language Model)
 [README](https://github.com/DaveCoDev/not-again-ai/blob/main/readmes/llm.md)
 
+Supports OpenAI chat completions and text embeddings. Includes functions for creating chat completion prompts, token management, and context management.
+
+One example: 
+```python
+client = openai_client()
+messages = [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Hello!"}]
+response = chat_completion(messages=messages, model="gpt-3.5-turbo", max_tokens=100, client=client)["message"]
+>>> "Hello! How can I help you today?"
+```
 
 ## Statistics
 [README](https://github.com/DaveCoDev/not-again-ai/blob/main/readmes/statistics.md)
@@ -230,9 +238,11 @@ To pass arguments to `pytest` through `nox`:
 
 ## Code Style Checking
 
-[PEP 8](https://peps.python.org/pep-0008/) is the universally accepted style guide for
-Python code. PEP 8 code compliance is verified using [Ruff](https://github.com/astral-sh/ruff).
-Ruff is configured in the `[tool.ruff]` section of `pyproject.toml`.
+[PEP 8](https://peps.python.org/pep-0008/) is the universally accepted style guide for Python
+code. PEP 8 code compliance is verified using [Ruff][Ruff]. Ruff is configured in the
+`[tool.ruff]` section of [`pyproject.toml`](./pyproject.toml).
+
+[Ruff]: https://github.com/astral-sh/ruff
 
 To lint code, run:
 
@@ -248,12 +258,7 @@ To automatically fix fixable lint errors, run:
 
 ## Automated Code Formatting
 
-Code is automatically formatted using [black](https://github.com/psf/black). Imports are
-automatically sorted and grouped using [Ruff](https://github.com/astral-sh/ruff).
-
-These tools are configured by:
-
-- [`pyproject.toml`](./pyproject.toml)
+[Ruff][Ruff] is used to automatically format code and group and sort imports.
 
 To automatically format code, run:
 

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "not-again-ai"
-version = "0.3.1"
+version = "0.4.1"
 description = "Designed to once and for all collect all the little things that come up over and over again in AI projects and put them in one place."
 authors = ["DaveCoDev <dave.co.dev@gmail.com>"]
 license = "MIT"
@@ -16,7 +16,6 @@ classifiers = [
     "Programming Language :: Python",
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3 :: Only",
-    "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
     "Typing :: Typed",
@@ -26,28 +25,25 @@ classifiers = [
 # Some packages, such as scipy, constrain their upper bound of Python versions they support.
 # Without also constraining the upper bound here, Poetry will not select those versions and will
 # result in an old version being resolved/locked.
-python = "^3.10, <3.13"
+python = "^3.11, <3.13"
 
 # Optional dependencies are defined here, and groupings are defined below.
-jinja2 = { version = "^3.1.2", optional = true }
-numpy = { version = "^1.26.2", optional = true }
-openai = { version = "^1.6.1", optional = true }
-pandas = { version = "^2.1.4", optional = true }
-scipy = { version = "^1.11.4", optional = true }
-scikit-learn = { version = "^1.3.2", optional = true }
-seaborn = { version = "^0.13.0", optional = true }
+numpy = { version = "^1.26.3", optional = true }
+openai = { version = "^1.10.0", optional = true }
+pandas = { version = "^2.2.0", optional = true }
+python-liquid = { version = "^1.10.2", optional = true }
+scipy = { version = "^1.12.0", optional = true }
+scikit-learn = { version = "^1.4.0", optional = true }
+seaborn = { version = "^0.13.2", optional = true }
 tiktoken = { version = "^0.5.2", optional = true }
 
 [tool.poetry.extras]
-llm = ["jinja2", "openai", "tiktoken"]
+llm = ["openai", "python-liquid", "tiktoken"]
 statistics = ["numpy", "scikit-learn", "scipy"]
 viz = ["numpy", "pandas", "seaborn"]
 
 [tool.poetry.group.nox.dependencies]
 nox-poetry = "*"
-# TODO: Remove this after virtualenv supports platformdirs 4.
-#   https://github.com/pypa/virtualenv/issues/2666
-platformdirs = "<4"
 
 [tool.poetry.group.test.dependencies]
 pytest = "*"
@@ -65,9 +61,6 @@ mypy = "*"
 [tool.poetry.group.lint.dependencies]
 ruff = "*"
 
-[tool.poetry.group.fmt.dependencies]
-black = "*"
-
 [tool.poetry.group.docs.dependencies]
 mkdocs-material = "*"
 mkdocs-htmlproofer-plugin = "*"
@@ -111,8 +104,6 @@ unfixable = ["F401"]
 [tool.ruff.isort]
 force-sort-within-sections = true
 split-on-trailing-comma = false
-# For non-src directory projects, explicitly set top level package names:
-# known-first-party = ["my-app"]
 
 [tool.ruff.flake8-tidy-imports]
 ban-relative-imports = "all"
@@ -120,13 +111,6 @@ ban-relative-imports = "all"
 [tool.ruff.flake8-bugbear]
 extend-immutable-calls = ["typer.Argument"]
 
-[tool.black]
-line-length = 120
-target-version = ["py39", "py310", "py311", "py312"]
-# black will automatically exclude all files listed in .gitignore
-# If you need to exclude additional folders, consider using extend-exclude to avoid disabling the
-# default .gitignore behaviour.
-
 [tool.pytest.ini_options]
 addopts = [
     "--strict-config",

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/src/not_again_ai/llm/__init__.py

@@ -1,5 +1,5 @@
 try:
-    import jinja2  # noqa
+    import liquid  # noqa
     import openai  # noqa
     import tiktoken  # noqa
 except ImportError:

not_again_ai-0.4.1/src/not_again_ai/llm/chat_completion.py

@@ -0,0 +1,102 @@
+import contextlib
+import json
+from typing import Any
+
+from openai import OpenAI
+
+
+def chat_completion(
+    messages: list[dict[str, str]],
+    model: str,
+    client: OpenAI,
+    tools: list[dict[str, Any]] | None = None,
+    tool_choice: str = "auto",
+    max_tokens: int | None = None,
+    temperature: float = 0.7,
+    json_mode: bool = False,
+    **kwargs: Any,
+) -> dict[str, Any]:
+    """Get an OpenAI chat completion response: https://platform.openai.com/docs/api-reference/chat/create
+
+    Args:
+        messages (list): A list of messages comprising the conversation so far.
+        model (str): ID of the model to use. See the model endpoint compatibility table:
+            https://platform.openai.com/docs/models/model-endpoint-compatibility
+            for details on which models work with the Chat API.
+        client (OpenAI): An instance of the OpenAI client.
+        tools (list[dict[str, Any]], optional): A list of tools the model may generate JSON inputs for.
+            Defaults to None.
+        tool_choice (str, optional): The tool choice to use. Can be "auto", "none", or a specific function name.
+            Defaults to "auto".
+        max_tokens (int, optional): The maximum number of tokens to generate in the chat completion.
+            Defaults to None, which automatically limits to the model's maximum context length.
+        temperature (float, optional): What sampling temperature to use, between 0 and 2.
+            Higher values like 0.8 will make the output more random,
+            while lower values like 0.2 will make it more focused and deterministic. Defaults to 0.7.
+        json_mode (bool, optional): When JSON mode is enabled, the model is constrained to only
+            generate strings that parse into valid JSON object and will return a dictionary.
+            See https://platform.openai.com/docs/guides/text-generation/json-mode
+        **kwargs: Additional keyword arguments to pass to the OpenAI client chat completion.
+
+    Returns:
+        dict: A dictionary containing the following keys:
+            - "finish_reason" (str): The reason the model stopped generating further tokens.
+                Can be "stop", "length", or "tool_calls".
+            - "tool_names" (list[str], optional): The names of the tools called by the model.
+            - "tool_args_list" (list[dict], optional): The arguments of the tools called by the model.
+            - "message" (str | dict): The content of the generated assistant message.
+                If json_mode is True, this will be a dictionary.
+            - "completion_tokens" (int): The number of tokens used by the model to generate the completion.
+            - "prompt_tokens" (int): The number of tokens in the generated response.
+    """
+    response_format = {"type": "json_object"} if json_mode else None
+
+    kwargs.update(
+        {
+            "messages": messages,
+            "model": model,
+            "tools": tools,
+            "max_tokens": max_tokens,
+            "temperature": temperature,
+            "response_format": response_format,
+            "n": 1,
+        }
+    )
+
+    if tools is not None:
+        if tool_choice not in ["none", "auto"]:
+            kwargs["tool_choice"] = {"type": "function", "function": {"name": tool_choice}}
+        else:
+            kwargs["tool_choice"] = tool_choice
+
+    # Call the function with the set parameters
+    response = client.chat.completions.create(**kwargs)
+
+    response_data = {}
+    finish_reason = response.choices[0].finish_reason
+    response_data["finish_reason"] = finish_reason
+
+    # Not checking finish_reason=="tool_calls" here because when a user providea function name as tool_choice,
+    # the finish reason is "stop", not "tool_calls"
+    tool_calls = response.choices[0].message.tool_calls
+    if tool_calls:
+        tool_names = []
+        tool_args_list = []
+        for tool_call in tool_calls:
+            tool_names.append(tool_call.function.name)
+            tool_args_list.append(json.loads(tool_call.function.arguments))
+        response_data["tool_names"] = tool_names
+        response_data["tool_args_list"] = tool_args_list
+    elif finish_reason == "stop" or finish_reason == "length":
+        message = response.choices[0].message.content
+        if json_mode:
+            with contextlib.suppress(json.JSONDecodeError):
+                message = json.loads(message)
+        response_data["message"] = message
+
+    usage = response.usage
+    if usage is not None:
+        response_data["completion_tokens"] = usage.completion_tokens
+        response_data["prompt_tokens"] = usage.prompt_tokens
+
+    return response_data

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/src/not_again_ai/llm/context_management.py

@@ -18,7 +18,7 @@ def priority_truncation(
     variables: dict[str, str],
     priority: list[str],
     token_limit: int,
-    model: str = "gpt-3.5-turbo-0613",
+    model: str = "gpt-3.5-turbo-1106",
 ) -> list[dict[str, str]]:
     """Formats messages_unformatted and injects variables into the messages in the order of priority, truncating the messages to fit the token limit.
 
@@ -37,7 +37,7 @@ def priority_truncation(
         variables: A dictionary where each key-value pair represents a variable name and its value to inject.
         priority: A list of variable names in their order of priority.
         token_limit: The maximum number of tokens allowed in the messages.
-        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-0613".
+        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-1106".
     """
 
     # Check if all variables in the priority list are in the variables dict.

not_again_ai-0.4.1/src/not_again_ai/llm/embeddings.py

@@ -0,0 +1,62 @@
+from typing import Any
+
+from openai import OpenAI
+
+
+def embed_text(
+    text: str | list[str],
+    client: OpenAI,
+    model: str = "text-embedding-3-large",
+    dimensions: int | None = None,
+    encoding_format: str = "float",
+    **kwargs: Any,
+) -> list[float] | str | list[list[float]] | list[str]:
+    """Generates an embedding vector for a given text using OpenAI's API.
+
+    Args:
+        text (str | list[str]): The input text to be embedded. Each text should not exceed 8191 tokens, which is the max for V2 and V3 models
+        client (OpenAI): The OpenAI client used to interact with the API.
+        model (str, optional): The ID of the model to use for embedding.
+            Defaults to "text-embedding-3-large".
+            Choose from text-embedding-3-small, text-embedding-3-large, text-embedding-ada-002.
+            See https://platform.openai.com/docs/models/embeddings for more details.
+        dimensions (int | None, optional): The number of dimensions for the output embeddings.
+            This is only supported in "text-embedding-3" and later models. Defaults to None.
+        encoding_format (str, optional): The format for the returned embeddings. Can be either "float" or "base64".
+            Defaults to "float".
+
+    Returns:
+        list[float] | str | list[list[float]] | list[str]: The embedding vector represented as a list of floats or base64 encoded string.
+            If multiple text inputs are provided, a list of embedding vectors is returned.
+            The length and format of the vector depend on the model, encoding_format, and dimensions.
+
+    Raises:
+        ValueError: If 'text-embedding-ada-002' model is used and dimensions are specified,
+            as this model does not support specifying dimensions.
+
+    Example:
+        client = OpenAI()
+        embedding = embed_text("Example text", client, model="text-embedding-ada-002")
+    """
+    if model == "text-embedding-ada-002" and dimensions:
+        # text-embedding-ada-002 does not support dimensions
+        raise ValueError("text-embedding-ada-002 does not support dimensions")
+
+    kwargs = {
+        "model": model,
+        "input": text,
+        "encoding_format": encoding_format,
+    }
+    if dimensions:
+        kwargs["dimensions"] = dimensions
+
+    response = client.embeddings.create(**kwargs)
+
+    responses = []
+    for embedding in response.data:
+        responses.append(embedding.embedding)
+
+    if len(responses) == 1:
+        return responses[0]
+
+    return responses

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/src/not_again_ai/llm/openai_client.py

@@ -21,10 +21,12 @@ def openai_client(
             Defaults to 'openai'.
         api_key (str, optional): The API key to authenticate the client. If not provided,
             OpenAI automatically uses `OPENAI_API_KEY` from the environment.
-        organization (str, optional): The ID of the organization (for enterprise users). If not provided,
+        organization (str, optional): The ID of the organization. If not provided,
             OpenAI automotically uses `OPENAI_ORG_ID` from the environment.
-        timeout (float, optional): TBD
-        max_retries (int, optional): TBD
+        timeout (float, optional): By default requests time out after 10 minutes.
+        max_retries (int, optional): Certain errors are automatically retried 2 times by default,
+            with a short exponential backoff. Connection errors (for example, due to a network connectivity problem),
+            408 Request Timeout, 409 Conflict, 429 Rate Limit, and >=500 Internal errors are all retried by default.
 
     Returns:
         OpenAI: An instance of the OpenAI client.
@@ -34,7 +36,7 @@ def openai_client(
         NotImplementedError: If the specified API type is recognized but not yet supported (e.g., 'azure_openai').
 
     Examples:
-        >>> client = oai_client(api_type="openai", api_key="YOUR_API_KEY")
+        >>> client = openai_client(api_type="openai", api_key="YOUR_API_KEY")
     """
     if api_type not in ["openai", "azure_openai"]:
         raise InvalidOAIAPITypeError(f"Invalid OAIAPIType: {api_type}. Must be 'openai' or 'azure_openai'.")

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/src/not_again_ai/llm/prompts.py

@@ -1,4 +1,4 @@
-import jinja2
+from liquid import Template
 
 
 def _validate_message(message: dict[str, str]) -> bool:
@@ -19,15 +19,13 @@ def _validate_message(message: dict[str, str]) -> bool:
 
 
 def chat_prompt(messages_unformatted: list[dict[str, str]], variables: dict[str, str]) -> list[dict[str, str]]:
-    """Formats a list of messages for OpenAI's chat completion API using Jinja2 templating.
-
-    The content of each message is treated as a Jinja2 template that is rendered
-    with the provided variables.
+    """
+    Formats a list of messages for OpenAI's chat completion API using Liquid templating.
 
     Args:
         messages_unformatted: A list of dictionaries where each dictionary
             represents a message. Each message must have 'role' and 'content'
-            keys with string values, where content is a Jinja2 template.
+            keys with string values, where content is a Liquid template.
         variables: A dictionary where each key-value pair represents a variable
             name and its value for template rendering.
 
@@ -47,10 +45,13 @@ def chat_prompt(messages_unformatted: list[dict[str, str]], variables: dict[str,
             {"role": "user", "content": "Help me write Python code for the fibonnaci sequence"}
         ]
     """
+
     messages_formatted = messages_unformatted.copy()
     for message in messages_formatted:
-        # Validate each message and return a ValueError if any message is invalid
         if not _validate_message(message):
             raise ValueError(f"Invalid message: {message}")
-        message["content"] = jinja2.Template(message["content"]).render(**variables)
+
+        liquid_template = Template(message["content"])
+        message["content"] = liquid_template.render(**variables)
+
     return messages_formatted

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/src/not_again_ai/llm/tokens.py

@@ -1,15 +1,14 @@
 import tiktoken
 
 
-def truncate_str(text: str, max_len: int, model: str = "gpt-3.5-turbo-0613") -> str:
+def truncate_str(text: str, max_len: int, model: str = "gpt-3.5-turbo-1106") -> str:
     """Truncates a string to a maximum token length.
 
     Args:
         text: The string to truncate.
         max_len: The maximum number of tokens to keep.
-        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-0613".
-            See https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb
-            for a list of OpenAI models.
+        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-1106".
+            See https://platform.openai.com/docs/models for a list of OpenAI models.
 
     Returns:
         The truncated string.
@@ -30,12 +29,13 @@ def truncate_str(text: str, max_len: int, model: str = "gpt-3.5-turbo-0613") ->
         return text
 
 
-def num_tokens_in_string(text: str, model: str = "gpt-3.5-turbo-0613") -> int:
+def num_tokens_in_string(text: str, model: str = "gpt-3.5-turbo-1106") -> int:
     """Return the number of tokens in a string.
 
     Args:
         text: The string to count the tokens.
-        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-0613".
+        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-1106".
+            See https://platform.openai.com/docs/models for a list of OpenAI models.
 
     Returns:
         The number of tokens in the string.
@@ -48,17 +48,17 @@ def num_tokens_in_string(text: str, model: str = "gpt-3.5-turbo-0613") -> int:
     return len(encoding.encode(text))
 
 
-def num_tokens_from_messages(messages: list[dict[str, str]], model: str = "gpt-3.5-turbo-0613") -> int:
+def num_tokens_from_messages(messages: list[dict[str, str]], model: str = "gpt-3.5-turbo-1106") -> int:
     """Return the number of tokens used by a list of messages.
     NOTE: Does not support counting tokens used by function calling.
     Reference: # https://github.com/openai/openai-cookbook/blob/main/examples/How_to_format_inputs_to_ChatGPT_models.ipynb
+    and https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb
 
     Args:
         messages: A list of messages to count the tokens
             should ideally be the result after calling llm.prompts.chat_prompt.
-        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-0613".
-            See https://github.com/openai/openai-cookbook/blob/main/examples/How_to_count_tokens_with_tiktoken.ipynb
-            for a list of OpenAI models.
+        model: The model to use for tokenization. Defaults to "gpt-3.5-turbo-1106".
+            See https://platform.openai.com/docs/models for a list of OpenAI models.
 
     Returns:
         The number of tokens used by the messages.
@@ -71,16 +71,21 @@ def num_tokens_from_messages(messages: list[dict[str, str]], model: str = "gpt-3
     if model in {
         "gpt-3.5-turbo-0613",
         "gpt-3.5-turbo-16k-0613",
+        "gpt-3.5-turbo-1106",
         "gpt-4-0314",
         "gpt-4-32k-0314",
         "gpt-4-0613",
         "gpt-4-32k-0613",
+        "gpt-4-1106-preview",
+        "gpt-4-turbo-preview",
+        "gpt-4-0125-preview",
     }:
         tokens_per_message = 3  # every message follows <|start|>{role/name}\n{content}<|end|>\n
         tokens_per_name = 1  # if there's a name, the role is omitted
     elif model == "gpt-3.5-turbo-0301":
         tokens_per_message = 4
         tokens_per_name = -1
+    # Approximate catch-all. Assumes future versions of 3.5 and 4 will have the same token counts as the 0613 versions.
     elif "gpt-3.5-turbo" in model:
         return num_tokens_from_messages(messages, model="gpt-3.5-turbo-0613")
     elif "gpt-4" in model:

{not_again_ai-0.3.1 → not_again_ai-0.4.1}/src/not_again_ai/viz/time_series.py

@@ -13,9 +13,11 @@ from not_again_ai.viz.utils import reset_plot_libs
 def ts_lineplot(
     ts_data: list[float] | (npt.NDArray[np.float64] | npt.NDArray[np.int64]),
     save_pathname: str,
-    ts_x: list[float]
-    | (npt.NDArray[np.float64] | (npt.NDArray[np.datetime64] | (npt.NDArray[np.int64] | pd.Series)))
-    | None = None,
+    ts_x: (
+        list[float]
+        | (npt.NDArray[np.float64] | (npt.NDArray[np.datetime64] | (npt.NDArray[np.int64] | pd.Series)))
+        | None
+    ) = None,
     ts_names: list[str] | None = None,
     title: str | None = None,
     xlabel: str | None = "Time",

not_again_ai-0.3.1/src/not_again_ai/llm/chat_completion.py

@@ -1,77 +0,0 @@
-import json
-from typing import Any
-
-from openai import OpenAI
-
-
-def chat_completion(
-    messages: list[dict[str, str]],
-    model: str,
-    client: OpenAI,
-    functions: list[dict[str, Any]] | None = None,
-    max_tokens: int | None = None,
-    temperature: float = 0.7,
-    **kwargs: Any,
-) -> dict[str, Any]:
-    """Get an OpenAI chat completion response: https://platform.openai.com/docs/api-reference/chat/create
-
-    Args:
-        messages (list): A list of messages comprising the conversation so far.
-        model (str): ID of the model to use. See the model endpoint compatibility table:
-            https://platform.openai.com/docs/models/model-endpoint-compatibility
-            for details on which models work with the Chat API.
-        client (OpenAI): An instance of the OpenAI client.
-        functions (list, optional): A list of functions the model may generate JSON inputs for. Defaults to None.
-        max_tokens (int, optional): The maximum number of tokens to generate in the chat completion.
-            Defaults to limited to the model's context length.
-        temperature (float, optional): What sampling temperature to use, between 0 and 2.
-            Higher values like 0.8 will make the output more random,
-            while lower values like 0.2 will make it more focused and deterministic. Defaults to 0.7.
-        **kwargs: Additional keyword arguments to pass to the OpenAI client chat completion.
-
-    Returns:
-        dict: A dictionary containing the following keys:
-            - "finish_reason" (str): The reason the model stopped generating further tokens. Can be "stop" or "function_call".
-            - "function_name" (str, optional): The name of the function called by the model, present only if "finish_reason" is "function_call".
-            - "function_args" (dict, optional): The arguments of the function called by the model, present only if "finish_reason" is "function_call".
-            - "message" (str, optional): The content of the generated assistant message, present only if "finish_reason" is "stop".
-            - "completion_tokens" (int): The number of tokens used by the model to generate the completion.
-            - "prompt_tokens" (int): The number of tokens in the generated response.
-    """
-    if functions is None:
-        response = client.chat.completions.create(
-            messages=messages,  # type: ignore
-            model=model,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            n=1,
-            **kwargs,
-        )
-    else:
-        response = client.chat.completions.create(  # type: ignore
-            messages=messages,
-            model=model,
-            functions=functions,
-            function_call="auto",
-            max_tokens=max_tokens,
-            temperature=temperature,
-            n=1,
-            **kwargs,
-        )
-
-    response_data = {}
-    finish_reason = response.choices[0].finish_reason
-    response_data["finish_reason"] = finish_reason
-    if finish_reason == "function_call":
-        function_call = response.choices[0].message.function_call
-        if function_call is not None:
-            response_data["function_name"] = function_call.name  # type: ignore
-            response_data["function_args"] = json.loads(function_call.arguments)
-    elif finish_reason == "stop" or finish_reason == "length":
-        message = response.choices[0].message
-        response_data["message"] = message.content  # type: ignore
-    usage = response.usage
-    if usage is not None:
-        response_data["completion_tokens"] = usage.completion_tokens  # type: ignore
-        response_data["prompt_tokens"] = usage.prompt_tokens  # type: ignore
-    return response_data