sinapsis-chatbots-base 0.1.10

Sinapsis base templates for LLM text completion.

Sinapsis Chatbots Base

Package with base support for chat completion tasks

🐍 Installation • 🚀 Features • 📚 Usage example • 📙 Documentation • 🔍 License

The sinapsis-chatbots-base module provides core functionality for llm chat completion tasks

🐍 Installation

Install using your package manager of choice. We encourage the use of uv

Example with uv:

  uv pip install sinapsis-chatbots-base --extra-index-url https://pypi.sinapsis.tech

or with raw pip:

  pip install sinapsis-chatbots-base --extra-index-url https://pypi.sinapsis.tech

[!IMPORTANT] Templates may require extra dependencies. For development, we recommend installing the package with all the optional dependencies:

with uv:

  uv pip install sinapsis-chatbots-base[all] --extra-index-url https://pypi.sinapsis.tech

or with raw pip:

  pip install sinapsis-chatbots-base[all] --extra-index-url https://pypi.sinapsis.tech

🚀 Features

LLMTextCompletionBase

Base class for all templates intended to perform chat (text) completion tasks.

These attributes apply to LLMTextCompletionBase:

• init_args(LLMInitArgs, required): Base model arguments, including the 'llm_model_name'.
• completion_args(LLMCompletionArgs, optional): Base generation arguments, including 'max_tokens', 'temperature', 'top_p', and 'top_k'.
• chat_history_key(str, optional): Key in the packet's generic_data to find the conversation history.
• rag_context_key(str, optional): Key in the packet's generic_data to find RAG context to inject.
• system_prompt(str | Path, optional): The system prompt (or path to one) to instruct the model.
• pattern(dict, optional): A regex pattern used to post-process the model's response.
• keep_before(bool, optional): If True, keeps text before the 'pattern' match; otherwise, keeps text after.

QueryContextualize

A base class for contextualizing queries based on certain keywords.

These attributes apply to QueryContextualize:

• keywords(list[str], required): A list of keywords to be used for retrieving context.

QueryContextualizeFromFile

A subclass of QueryContextualize that retrieves context from files loaded into the generic_data.

These attributes apply to QueryContextualizeFromFile:

• keywords(list[str], required): A list of keywords to be used for retrieving context.
• generic_keys(list[str], required): A list of keywords that can be used to retrieve specific context data from the generic_data of the container.

[!TIP] Use CLI command sinapsis info --all-template-names to show a list with all the available Template names installed with Sinapsis Data Tools.

[!TIP] Use CLI command sinapsis info --example-template-config TEMPLATE_NAME to produce an example Agent config for the Template specified in TEMPLATE_NAME.

For example, for QueryContextualizeFromFile use sinapsis info --example-template-config QueryContextualizeFromFile to produce the following example config:

agent:
  name: my_test_agent
templates:
- template_name: InputTemplate
  class_name: InputTemplate
  attributes: {}
- template_name: QueryContextualizeFromFile
  class_name: QueryContextualizeFromFile
  template_input: InputTemplate
  attributes:
    keywords: '`replace_me:list[str]`'
    generic_keys: '`replace_me:list[str]`'

📚 Usage example

The following agent passes a text message through a TextPacket and checks if there is context with the chosen keyword

Config

agent:
  name: chat_completion
  description: Agent with a chatbot that makes a call to the LLM model using a context uploaded from a file

templates:
- template_name: InputTemplate
  class_name: InputTemplate
  attributes: {}

- template_name: PyPDFLoaderWrapper
  class_name: PyPDFLoaderWrapper
  template_input: InputTemplate
  attributes:
    add_document_as_text_packet: false
    pypdfloader_init:
      file_path: '/path/to/a/file.pdf'

- template_name: TextInput
  class_name: TextInput
  template_input: PyPDFLoaderWrapper
  attributes:
    text: what is AI?

- template_name: QueryContextualizeFromFile
  class_name: QueryContextualizeFromFile
  template_input: TextInput
  attributes:
    keywords: 'Artificial Intelligence'
    generic_keys: 'PyPDFLoaderWrapper'

📙 Documentation

Documentation for this and other sinapsis packages is available on the sinapsis website

Tutorials for different projects within sinapsis are available at sinapsis tutorials page

🔍 License

This project is licensed under the AGPLv3 license, which encourages open collaboration and sharing. For more details, please refer to the LICENSE file.

For commercial use, please refer to our official Sinapsis website for information on obtaining a commercial license.