@@ -1 +1 @@
-gzip compressed data, was "clonellm-0.0.2.tar", max compression
+gzip compressed data, was "clonellm-0.0.3.tar", max compression@@ -1,12 +1,12 @@
--rw-r--r-- 0 0 0 1070 2024-05-24 08:18:44.196305 clonellm-0.0.2/LICENSE
--rw-r--r-- 0 0 0 9668 2024-05-24 08:18:44.200305 clonellm-0.0.2/README.md
--rw-r--r-- 0 0 0 1274 2024-05-24 08:18:44.200305 clonellm-0.0.2/pyproject.toml
--rw-r--r-- 0 0 0 209 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/__init__.py
--rw-r--r-- 0 0 0 745 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/_base.py
--rw-r--r-- 0 0 0 2038 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/_prompt.py
--rw-r--r-- 0 0 0 1119 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/_typing.py
--rw-r--r-- 0 0 0 10078 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/core.py
--rw-r--r-- 0 0 0 3190 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/embed.py
--rw-r--r-- 0 0 0 1205 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/memory.py
--rw-r--r-- 0 0 0 0 2024-05-24 08:18:44.200305 clonellm-0.0.2/src/clonellm/py.typed
--rw-r--r-- 0 0 0 10510 1970-01-01 00:00:00.000000 clonellm-0.0.2/PKG-INFO
+-rw-r--r-- 0 0 0 1070 2024-06-01 01:45:57.209604 clonellm-0.0.3/LICENSE
+-rw-r--r-- 0 0 0 9851 2024-06-01 01:45:57.209604 clonellm-0.0.3/README.md
+-rw-r--r-- 0 0 0 1326 2024-06-01 01:45:57.213604 clonellm-0.0.3/pyproject.toml
+-rw-r--r-- 0 0 0 209 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/__init__.py
+-rw-r--r-- 0 0 0 745 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/_base.py
+-rw-r--r-- 0 0 0 1525 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/_prompt.py
+-rw-r--r-- 0 0 0 1217 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/_typing.py
+-rw-r--r-- 0 0 0 9553 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/core.py
+-rw-r--r-- 0 0 0 3220 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/embed.py
+-rw-r--r-- 0 0 0 1205 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/memory.py
+-rw-r--r-- 0 0 0 0 2024-06-01 01:45:57.213604 clonellm-0.0.3/src/clonellm/py.typed
+-rw-r--r-- 0 0 0 10693 1970-01-01 00:00:00.000000 clonellm-0.0.3/PKG-INFO- Files identical despite different names
- Files 2% similar despite different names
@@ -6,29 +6,29 @@
</h1>
<p align="center">
<p align="center">Create an AI clone of yourself using LLMs.</p>
</p>
<h4 align="center">
<a href="https://pypi.org/project/clonellm/" target="_blank">
- <img src="https://img.shields.io/badge/release-v0.0.2-green" alt="Latest Release">
+ <img src="https://img.shields.io/badge/release-v0.0.3-green" alt="Latest Release">
</a>
<a href="https://pypi.org/project/clonellm/" target="_blank">
<img src="https://img.shields.io/pypi/v/clonellm.svg" alt="PyPI Version">
</a>
<a target="_blank">
<img src="https://img.shields.io/badge/python-3.9%20%7C%203.10%20%7C%203.11%20%7C%203.12-blue" alt="Python Versions">
</a>
<a target="_blank">
<img src="https://img.shields.io/pypi/l/clonellm" alt="PyPI License">
</a>
</h4>
## Introduction
-A minimal Python package that enables you to create an AI clone of yourself using LLMs. Built on top of LiteLLM and Langchain, CloneLLM utilizes the Retrieval-Augmented Generation (RAG) to tailor AI responses as if you are answering the questions.
+A minimal Python package that enables you to create an AI clone of yourself using LLMs. Built on top of LiteLLM and LangChain, CloneLLM utilizes the Retrieval-Augmented Generation (RAG) to tailor AI responses as if you are answering the questions.
You can input texts and documents about yourself — including personal information, professional experience, educational background, etc. — which are then embedded into a vector space for dynamic retrieval. This AI clone can act as a virtual assistant or digital representation, capable of handling queries and tasks in a manner that reflects the your own knowledge, tone, style and mannerisms.
## Installation
### Prerequisites
Before installing CloneLLM, make sure you have Python 3.9 or newer installed on your machine.
@@ -65,15 +65,15 @@
documents = [
Document(page_content="My name is Mehdi Samsami."),
open("cv.txt", "r").read(),
]-Step 2. Initialize an embedding model using CloneLLM's LiteLLMEmbeddings or Langchain's embeddings. Then, initialize a clone with your documents, embedding model, and your referred LLM.
+Step 2. Initialize an embedding model using CloneLLM's LiteLLMEmbeddings or LangChain's embeddings. Then, initialize a clone with your documents, embedding model, and your referred LLM.
from clonellm import CloneLLM, LiteLLMEmbeddings
embedding = LiteLLMEmbeddings(model="text-embedding-ada-002")
clone = CloneLLM(model="gpt-4-turbo", documents=documents, embedding=embedding)@@ -94,15 +94,15 @@
### Models
At its core, CloneLLM utilizes LiteLLM for interactions with various LLMs. This is why you can choose from many different providers (100+ LLMs) supported by LiteLLM, including Bedrock, Azure, OpenAI, Cohere, Anthropic, Ollama, Sagemaker, HuggingFace, Replicate, etc.
### Document loaders
-You can use Langchain's document loaders to seamlessly import data from various sources into `Document` format. Take, for example, text and HTML loaders:
+You can use LangChain's document loaders to seamlessly import data from various sources into `Document` format. Take, for example, text and HTML loaders:
```python
# !pip install unstructured
from langchain_community.document_loaders import TextLoader, UnstructuredHTMLLoader
documents = TextLoader("cv.txt").load() + UnstructuredHTMLLoader("linkedin.html").load()
@@ -115,15 +115,15 @@ file_path='chat.json', jq_schema='.messages[].content', text_content=False ).load()
### Embeddings
-With `LiteLLMEmbeddings`, CloneLLM allows you to utilize embedding models from a variety of providers supported by LiteLLM. Additionally, you can select any preferred embedding model from Langchain's extensive range. Take, for example, the Hugging Face embedding:
+With `LiteLLMEmbeddings`, CloneLLM allows you to utilize embedding models from a variety of providers supported by LiteLLM. Additionally, you can select any preferred embedding model from LangChain's extensive range. Take, for example, the Hugging Face embedding:
```python
# !pip install --upgrade --quiet sentence_transformers
from langchain_community.embeddings import HuggingFaceEmbeddings
from clonellm import CloneLLM
import os
os.environ["COHERE_API_KEY"] = "cohere-api-key"
@@ -276,14 +276,17 @@
- [x] Add support for conversation history
- [ ] Add support for RAG with no embedding (ingest the entire context into the prompt)
- [x] Add support for string documents
- [x] Fix mypy errors
- [x] Rename `completion` methods to `invoke`
- [x] Add support for streaming completion
- [ ] Add support for custom system prompts
-- [ ] Add an attribute to return supported models
+- [x] Make `LiteLLMEmbeddings.all_embedding_models` a property
+- [ ] Add an attribute to `CloneLLM` to return supported models
- [x] Add initial version of README
+- [ ] Describe `CloneLLM.clear_memory` method in README
- [ ] Add documents
-- [ ] Add usage examples
-- [ ] Add initial unit tests
+- [x] Add usage examples
+- [x] Add unit tests for non-core modules
+- [ ] Add unit tests for core module
- [x] Add GitHub workflow to run tests on PR
- [x] Add GitHub workflow to publish to PyPI on release
@@ -1,13 +1,13 @@
[Logo]
*�**�**�**�**�**�* C�Cl�lo�on�ne�eL�LL�LM�M *�**�**�**�**�**�*
Create an AI clone of yourself using LLMs.
*�**�**�* _�[�[_�L�L_�a�a_�t�t_�e�e_�s�s_�t�t_� _�R�R_�e�e_�l�l_�e�e_�a�a_�s�s_�e�e_�]�]_�[�[_�P�P_�y�y_�P�P_�I�I_� _�V�V_�e�e_�r�r_�s�s_�i�i_�o�o_�n�n_�]�][�[P�Py�yt�th�ho�on�n V�Ve�er�rs�si�io�on�ns�s]�][�[P�Py�yP�PI�I L�Li�ic�ce�en�ns�se�e]�] *�**�**�*
## Introduction A minimal Python package that enables you to create an AI clone
-of yourself using LLMs. Built on top of LiteLLM and Langchain, CloneLLM
+of yourself using LLMs. Built on top of LiteLLM and LangChain, CloneLLM
utilizes the Retrieval-Augmented Generation (RAG) to tailor AI responses as if
you are answering the questions. You can input texts and documents about
yourself � including personal information, professional experience,
educational background, etc. � which are then embedded into a vector space
for dynamic retrieval. This AI clone can act as a virtual assistant or digital
representation, capable of handling queries and tasks in a manner that reflects
the your own knowledge, tone, style and mannerisms. ## Installation ###
@@ -18,38 +18,38 @@
the project directory cd clonellm # Install the package pip install . ``` ##
Usage ### Getting started **Step 1**. Gather documents that contain relavant
information about you. These documents form the base from which your AI clone
will learn to mimic your tone, style, and expertise. ```python from
langchain_core.documents import Document documents = [ Document
(page_content="My name is Mehdi Samsami."), open("cv.txt", "r").read(), ] ```
**Step 2**. Initialize an embedding model using CloneLLM's `LiteLLMEmbeddings`
-or Langchain's embeddings. Then, initialize a clone with your documents,
+or LangChain's embeddings. Then, initialize a clone with your documents,
embedding model, and your referred LLM. ```python from clonellm import
CloneLLM, LiteLLMEmbeddings embedding = LiteLLMEmbeddings(model="text-
embedding-ada-002") clone = CloneLLM(model="gpt-4-turbo", documents=documents,
embedding=embedding) ``` **Step 3**. Configure environment variables to store
API keys for embedding and LLM models. ```bash export OPENAI_API_KEY=sk-... ```
**Step 4**. Fit the clone to the data (documents). ```python clone.fit() ```
**Step 5**. Invoke the clone to ask questions. ```python clone.invoke("What's
your name?") # Response: My name is Mehdi Samsami. How can I help you? ``` ###
Models At its core, CloneLLM utilizes LiteLLM for interactions with various
LLMs. This is why you can choose from many different providers (100+ LLMs)
supported by LiteLLM, including Bedrock, Azure, OpenAI, Cohere, Anthropic,
Ollama, Sagemaker, HuggingFace, Replicate, etc. ### Document loaders You can
-use Langchain's document loaders to seamlessly import data from various sources
+use LangChain's document loaders to seamlessly import data from various sources
into `Document` format. Take, for example, text and HTML loaders: ```python #
!pip install unstructured from langchain_community.document_loaders import
TextLoader, UnstructuredHTMLLoader documents = TextLoader("cv.txt").load() +
UnstructuredHTMLLoader("linkedin.html").load() ``` Or JSON loader: ```python #
!pip install jq from langchain_community.document_loaders import JSONLoader
documents = JSONLoader( file_path='chat.json', jq_schema='.messages[].content',
text_content=False ).load() ``` ### Embeddings With `LiteLLMEmbeddings`,
CloneLLM allows you to utilize embedding models from a variety of providers
supported by LiteLLM. Additionally, you can select any preferred embedding
-model from Langchain's extensive range. Take, for example, the Hugging Face
+model from LangChain's extensive range. Take, for example, the Hugging Face
embedding: ```python # !pip install --upgrade --quiet sentence_transformers
from langchain_community.embeddings import HuggingFaceEmbeddings from clonellm
import CloneLLM import os os.environ["COHERE_API_KEY"] = "cohere-api-key"
embedding = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-
base-v2") clone = CloneLLM(model="command-xlarge-beta", documents=documents,
embedding=embedding) ``` Or, the Llama-cpp embedding: ```python # !pip install
--upgrade --quiet llama-cpp-python from langchain_community.embeddings import
@@ -107,12 +107,14 @@
bugs, or improving documentation. - � **Share:** Post about CloneLLM on
LinkedIn or other social platforms. Thank you for your interest in CloneLLM. We
look forward to seeing what you'll create with your AI clone! ## TODO - [x] Add
pre commit configuration file - [x] Add setup.py script - [x] Add support for
conversation history - [ ] Add support for RAG with no embedding (ingest the
entire context into the prompt) - [x] Add support for string documents - [x]
Fix mypy errors - [x] Rename `completion` methods to `invoke` - [x] Add support
-for streaming completion - [ ] Add support for custom system prompts - [ ] Add
-an attribute to return supported models - [x] Add initial version of README -
-[ ] Add documents - [ ] Add usage examples - [ ] Add initial unit tests - [x]
-Add GitHub workflow to run tests on PR - [x] Add GitHub workflow to publish to
-PyPI on release
+for streaming completion - [ ] Add support for custom system prompts - [x] Make
+`LiteLLMEmbeddings.all_embedding_models` a property - [ ] Add an attribute to
+`CloneLLM` to return supported models - [x] Add initial version of README - [ ]
+Describe `CloneLLM.clear_memory` method in README - [ ] Add documents - [x] Add
+usage examples - [x] Add unit tests for non-core modules - [ ] Add unit tests
+for core module - [x] Add GitHub workflow to run tests on PR - [x] Add GitHub
+workflow to publish to PyPI on release- Files 4% similar despite different names
@@ -1,14 +1,14 @@
[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"
[tool.poetry]
name = "clonellm"
-version = "0.0.2"
+version = "0.0.3"
description = "Python package to create an AI clone of yourself using LLMs."
packages = [{ from = "src", include = "clonellm" }]
license = "MIT"
authors = ["Mehdi Samsami <[email protected]>"]
readme = "README.md"
keywords = ["llm", "language models", "nlp", "rag", "ai", "ai clone"]
repository = "https://github.com/msamsami/clonellm"
@@ -23,14 +23,15 @@
pytest = "*"
mypy = "*"
pytest-asyncio = "*"
ruff = "^0.4"
types-setuptools = "^69.0.0.20240106"
[tool.mypy]
+exclude=["tests", "examples"]
namespace_packages = false
files = ["src/clonellm/**/*.py"]
check_untyped_defs = true
disable_error_code = ["empty-body"]
disallow_untyped_defs = false
strict = true
@@ -38,14 +39,15 @@
module = [
"litellm",
"openai.lib.streaming",
]
ignore_missing_imports = true
[tool.pytest.ini_options]
+testpaths = ["tests"]
filterwarnings = ["ignore"]
[tool.ruff]
include = ["*.py", "*.pyi", "**/pyproject.toml", "*.ipynb"]
line-length = 130
# Use `ruff linter` to list available linters
# https://beta.ruff.rs/docs/rules/- Files identical despite different names
- Files 11% similar despite different names
@@ -1,10 +1,10 @@
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
-__all__ = ("context_prompt", "user_profile_prompt", "contextualize_question_prompt", "history_prompt", "question_prompt")
+__all__ = ("context_prompt", "user_profile_prompt", "history_prompt", "question_prompt")
context_prompt = ChatPromptTemplate.from_messages(
[
("system", "You are my human clone. Act as I would. Do not ask questions about me. If anyone asks, you are me."),
("system", "You are having a conversation with someone who wants to learn more about you."),
("system", "You MUST speak in first person. If you do not, you will be terminated."),
@@ -23,27 +23,14 @@
user_profile_prompt = ChatPromptTemplate.from_messages(
[
("system", "Here is all the known information about me: {user_profile}"),
]
)
-contextualize_question_prompt = ChatPromptTemplate.from_messages(
- [
- (
- "system",
- (
- "Given a chat history and the latest user question hich might reference context in the chat history, "
- "formulate a standalone question which can be understood without the chat history. "
- "Do NOT answer the question, just reformulate it if needed and otherwise return it as is."
- ),
- ),
- ]
-)
-
history_prompt = ChatPromptTemplate.from_messages(
[
MessagesPlaceholder(variable_name="chat_history"),
]
)
question_prompt = ChatPromptTemplate.from_messages(- Files 11% similar despite different names
@@ -1,7 +1,8 @@
+from __future__ import annotations
import datetime
from typing import Any, Optional
from pydantic import BaseModel
__all__ = ("UserProfile",)
@@ -15,17 +16,17 @@
birth_date: Optional[datetime.date | str] = None
gender: Optional[str] = None
city: Optional[str] = None
state: Optional[str] = None
country: Optional[str] = None
phone_number: Optional[str] = None
email: Optional[str] = None
- education_experience: Optional[dict[str, Any]] = None
- work_experience: Optional[dict[str, Any]] = None
- expertise: Optional[list[str]] = None
+ education_experience: Optional[dict[str, Any] | list[dict[str, Any]]] = None
+ work_experience: Optional[dict[str, Any] | list[dict[str, Any]]] = None
+ expertise: Optional[list[str] | dict[str, Any]] = None
home_page: Optional[str] = None
github_page: Optional[str] = None
linkedin_page: Optional[str] = None
@property
def full_name(self) -> str:
return " ".join([self.first_name, self.middle_name or "", self.last_name])- Files 4% similar despite different names
@@ -1,29 +1,27 @@
from __future__ import annotations
import json
import logging
-from typing import Any, AsyncIterator, Iterator, Optional
+from operator import itemgetter
+from typing import Any, AsyncIterator, cast, Iterator, Optional
from typing_extensions import Self
import uuid
-from langchain.chains.combine_documents import create_stuff_documents_chain
-from langchain.chains.history_aware_retriever import create_history_aware_retriever
-from langchain.chains.retrieval import create_retrieval_chain
from langchain.text_splitter import CharacterTextSplitter, TextSplitter
from langchain_core.documents import Document
from langchain_core.embeddings import Embeddings
from langchain_core.output_parsers import StrOutputParser
from langchain_core.runnables import RunnablePassthrough, RunnableSerializable
from langchain_core.runnables.history import RunnableWithMessageHistory
from langchain_core.vectorstores import VectorStoreRetriever
from langchain_community.chat_models import ChatLiteLLM
from langchain_community.vectorstores import Chroma
from ._base import LiteLLMMixin
-from ._prompt import context_prompt, user_profile_prompt, history_prompt, contextualize_question_prompt, question_prompt
+from ._prompt import context_prompt, user_profile_prompt, history_prompt, question_prompt
from ._typing import UserProfile
from .embed import LiteLLMEmbeddings
from .memory import get_session_history, clear_session_history
logging.getLogger("langchain_core").setLevel(logging.ERROR)
__all__ = ("CloneLLM",)
@@ -154,79 +152,77 @@
prompt = context_prompt.copy()
if self.user_profile:
prompt += user_profile_prompt.format_messages(user_profile=self._user_profile)
prompt += question_prompt
return {"context": self._get_retriever(), "input": RunnablePassthrough()} | prompt | self._llm | StrOutputParser()
def _get_rag_chain_with_history(self) -> RunnableWithMessageHistory:
- contextualize_system_prompt = contextualize_question_prompt + history_prompt + question_prompt
- history_aware_retriever = create_history_aware_retriever(self._llm, self._get_retriever(), contextualize_system_prompt)
-
prompt = context_prompt
if self.user_profile:
prompt += user_profile_prompt.format_messages(user_profile=self._user_profile)
prompt += history_prompt
prompt += question_prompt
- question_answer_chain = create_stuff_documents_chain(self._llm, prompt)
- rag_chain = create_retrieval_chain(history_aware_retriever, question_answer_chain)
+ context = itemgetter("input") | self._get_retriever()
+ first_step = RunnablePassthrough.assign(context=context)
+ rag_chain = first_step | prompt | self._llm | StrOutputParser()
+
return RunnableWithMessageHistory(
- rag_chain,
+ rag_chain, # type: ignore[arg-type]
get_session_history,
input_messages_key="input",
history_messages_key="chat_history",
- output_messages_key="answer",
output_parser=StrOutputParser(),
)
def invoke(self, prompt: str) -> str:
self._check_is_fitted()
if self.memory:
rag_chain_with_history = self._get_rag_chain_with_history()
response = rag_chain_with_history.invoke({"input": prompt}, config={"configurable": {"session_id": self._session_id}})
- return response["answer"] # type: ignore[no-any-return]
+ return cast(str, response.content)
rag_chain = self._get_rag_chain()
return rag_chain.invoke(prompt)
async def ainvoke(self, prompt: str) -> str:
self._check_is_fitted()
if self.memory:
rag_chain_with_history = self._get_rag_chain_with_history()
response = await rag_chain_with_history.ainvoke(
{"input": prompt}, config={"configurable": {"session_id": self._session_id}}
)
- return response["answer"] # type: ignore[no-any-return]
+ return cast(str, response.content)
rag_chain = self._get_rag_chain()
return await rag_chain.ainvoke(prompt)
def stream(self, prompt: str) -> Iterator[str]:
self._check_is_fitted()
if self.memory:
rag_chain_with_history = self._get_rag_chain_with_history()
- iterator = rag_chain_with_history.stream({"input": prompt}, config={"configurable": {"session_id": self._session_id}})
- for chunk in iterator:
- if "answer" in chunk:
- yield chunk["answer"]
- else:
- yield ""
- rag_chain = self._get_rag_chain()
- for chunk in rag_chain.stream(prompt):
- yield chunk
+ for chunk in rag_chain_with_history.stream(
+ {"input": prompt}, config={"configurable": {"session_id": self._session_id}}
+ ):
+ yield chunk.content
+ else:
+ rag_chain = self._get_rag_chain()
+ for chunk in rag_chain.stream(prompt):
+ yield chunk
async def astream(self, prompt: str) -> AsyncIterator[str]:
self._check_is_fitted()
if self.memory:
rag_chain_with_history = self._get_rag_chain_with_history()
async for chunk in rag_chain_with_history.astream(
{"input": prompt}, config={"configurable": {"session_id": self._session_id}}
):
- yield chunk["answer"]
- rag_chain = self._get_rag_chain()
- async for chunk in rag_chain.astream(prompt):
- yield chunk
+ yield chunk.content
+ else:
+ rag_chain = self._get_rag_chain()
+ async for chunk in rag_chain.astream(prompt):
+ yield chunk
def clear_memory(self) -> None:
clear_session_history(self._session_id)
self._session_id = str(uuid.uuid4())
def __repr__(self) -> str:
return f"CloneLLM<(model='{self.model}', memory={self.memory})>"- Files 3% similar despite different names
@@ -20,29 +20,29 @@
"""
def __init__(self, model: str, api_key: Optional[str] = None, dimensions: Optional[int] = None, **kwargs: Any) -> None:
super().__init__(model, api_key, **kwargs)
self.dimensions = dimensions
def embed_documents(self, texts: list[str]) -> list[list[float]]:
- """Call out to LLM's embedding endpoint for embedding search docs.
+ """Call out to LLM's embedding endpoint for embedding a list of documents.
Args:
texts (list[str]): The list of texts to embed.
Returns:
list[list[float]]: List of embeddings, one for each text.
"""
response = embedding(
model=self.model, input=texts, api_key=self.api_key, dimensions=self.dimensions, **self._litellm_kwargs
)
return [r["embedding"] for r in response.data]
async def aembed_documents(self, texts: list[str]) -> list[list[float]]:
- """Call out to LLM's embedding endpoint async for embedding search docs.
+ """Call out to LLM's embedding endpoint async for embedding a list of documents.
Args:
texts (list[str]): The list of texts to embed.
Returns:
list[list[float]]: List of embeddings, one for each text.
"""
@@ -70,14 +70,15 @@
Returns:
list[float]: Embedding for the text.
"""
embeddings = await self.aembed_documents([text])
return embeddings[0]
+ @property
def all_embedding_models(self) -> list[str]:
"""
Returns the names of supported embedding models.
"""
return cast(list[str], all_embedding_models)
def __repr__(self) -> str:- Files identical despite different names
- Files 6% similar despite different names
@@ -1,10 +1,10 @@
Metadata-Version: 2.1
Name: clonellm
-Version: 0.0.2
+Version: 0.0.3
Summary: Python package to create an AI clone of yourself using LLMs.
Home-page: https://github.com/msamsami/clonellm
License: MIT
Keywords: llm,language models,nlp,rag,ai,ai clone
Author: Mehdi Samsami
Author-email: [email protected]
Requires-Python: >=3.9,<3.13
@@ -28,29 +28,29 @@
</h1>
<p align="center">
<p align="center">Create an AI clone of yourself using LLMs.</p>
</p>
<h4 align="center">
<a href="https://pypi.org/project/clonellm/" target="_blank">
- <img src="https://img.shields.io/badge/release-v0.0.2-green" alt="Latest Release">
+ <img src="https://img.shields.io/badge/release-v0.0.3-green" alt="Latest Release">
</a>
<a href="https://pypi.org/project/clonellm/" target="_blank">
<img src="https://img.shields.io/pypi/v/clonellm.svg" alt="PyPI Version">
</a>
<a target="_blank">
<img src="https://img.shields.io/badge/python-3.9%20%7C%203.10%20%7C%203.11%20%7C%203.12-blue" alt="Python Versions">
</a>
<a target="_blank">
<img src="https://img.shields.io/pypi/l/clonellm" alt="PyPI License">
</a>
</h4>
## Introduction
-A minimal Python package that enables you to create an AI clone of yourself using LLMs. Built on top of LiteLLM and Langchain, CloneLLM utilizes the Retrieval-Augmented Generation (RAG) to tailor AI responses as if you are answering the questions.
+A minimal Python package that enables you to create an AI clone of yourself using LLMs. Built on top of LiteLLM and LangChain, CloneLLM utilizes the Retrieval-Augmented Generation (RAG) to tailor AI responses as if you are answering the questions.
You can input texts and documents about yourself — including personal information, professional experience, educational background, etc. — which are then embedded into a vector space for dynamic retrieval. This AI clone can act as a virtual assistant or digital representation, capable of handling queries and tasks in a manner that reflects the your own knowledge, tone, style and mannerisms.
## Installation
### Prerequisites
Before installing CloneLLM, make sure you have Python 3.9 or newer installed on your machine.
@@ -87,15 +87,15 @@
documents = [
Document(page_content="My name is Mehdi Samsami."),
open("cv.txt", "r").read(),
]-Step 2. Initialize an embedding model using CloneLLM's LiteLLMEmbeddings or Langchain's embeddings. Then, initialize a clone with your documents, embedding model, and your referred LLM.
+Step 2. Initialize an embedding model using CloneLLM's LiteLLMEmbeddings or LangChain's embeddings. Then, initialize a clone with your documents, embedding model, and your referred LLM.
from clonellm import CloneLLM, LiteLLMEmbeddings
embedding = LiteLLMEmbeddings(model="text-embedding-ada-002")
clone = CloneLLM(model="gpt-4-turbo", documents=documents, embedding=embedding)@@ -116,15 +116,15 @@
### Models
At its core, CloneLLM utilizes LiteLLM for interactions with various LLMs. This is why you can choose from many different providers (100+ LLMs) supported by LiteLLM, including Bedrock, Azure, OpenAI, Cohere, Anthropic, Ollama, Sagemaker, HuggingFace, Replicate, etc.
### Document loaders
-You can use Langchain's document loaders to seamlessly import data from various sources into `Document` format. Take, for example, text and HTML loaders:
+You can use LangChain's document loaders to seamlessly import data from various sources into `Document` format. Take, for example, text and HTML loaders:
```python
# !pip install unstructured
from langchain_community.document_loaders import TextLoader, UnstructuredHTMLLoader
documents = TextLoader("cv.txt").load() + UnstructuredHTMLLoader("linkedin.html").load()
@@ -137,15 +137,15 @@ file_path='chat.json', jq_schema='.messages[].content', text_content=False ).load()
### Embeddings
-With `LiteLLMEmbeddings`, CloneLLM allows you to utilize embedding models from a variety of providers supported by LiteLLM. Additionally, you can select any preferred embedding model from Langchain's extensive range. Take, for example, the Hugging Face embedding:
+With `LiteLLMEmbeddings`, CloneLLM allows you to utilize embedding models from a variety of providers supported by LiteLLM. Additionally, you can select any preferred embedding model from LangChain's extensive range. Take, for example, the Hugging Face embedding:
```python
# !pip install --upgrade --quiet sentence_transformers
from langchain_community.embeddings import HuggingFaceEmbeddings
from clonellm import CloneLLM
import os
os.environ["COHERE_API_KEY"] = "cohere-api-key"
@@ -298,15 +298,18 @@
- [x] Add support for conversation history
- [ ] Add support for RAG with no embedding (ingest the entire context into the prompt)
- [x] Add support for string documents
- [x] Fix mypy errors
- [x] Rename `completion` methods to `invoke`
- [x] Add support for streaming completion
- [ ] Add support for custom system prompts
-- [ ] Add an attribute to return supported models
+- [x] Make `LiteLLMEmbeddings.all_embedding_models` a property
+- [ ] Add an attribute to `CloneLLM` to return supported models
- [x] Add initial version of README
+- [ ] Describe `CloneLLM.clear_memory` method in README
- [ ] Add documents
-- [ ] Add usage examples
-- [ ] Add initial unit tests
+- [x] Add usage examples
+- [x] Add unit tests for non-core modules
+- [ ] Add unit tests for core module
- [x] Add GitHub workflow to run tests on PR
- [x] Add GitHub workflow to publish to PyPI on release
@@ -1,8 +1,8 @@
-Metadata-Version: 2.1 Name: clonellm Version: 0.0.2 Summary: Python package to
+Metadata-Version: 2.1 Name: clonellm Version: 0.0.3 Summary: Python package to
create an AI clone of yourself using LLMs. Home-page: https://github.com/
msamsami/clonellm License: MIT Keywords: llm,language models,nlp,rag,ai,ai
clone Author: Mehdi Samsami Author-email: [email protected] Requires-
Python: >=3.9,<3.13 Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3 Classifier: Programming
Language :: Python :: 3.9 Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11 Classifier: Programming
@@ -11,15 +11,15 @@
Repository, https://github.com/msamsami/clonellm Description-Content-Type:
text/markdown
[Logo]
*�**�**�**�**�**�* C�Cl�lo�on�ne�eL�LL�LM�M *�**�**�**�**�**�*
Create an AI clone of yourself using LLMs.
*�**�**�* _�[�[_�L�L_�a�a_�t�t_�e�e_�s�s_�t�t_� _�R�R_�e�e_�l�l_�e�e_�a�a_�s�s_�e�e_�]�]_�[�[_�P�P_�y�y_�P�P_�I�I_� _�V�V_�e�e_�r�r_�s�s_�i�i_�o�o_�n�n_�]�][�[P�Py�yt�th�ho�on�n V�Ve�er�rs�si�io�on�ns�s]�][�[P�Py�yP�PI�I L�Li�ic�ce�en�ns�se�e]�] *�**�**�*
## Introduction A minimal Python package that enables you to create an AI clone
-of yourself using LLMs. Built on top of LiteLLM and Langchain, CloneLLM
+of yourself using LLMs. Built on top of LiteLLM and LangChain, CloneLLM
utilizes the Retrieval-Augmented Generation (RAG) to tailor AI responses as if
you are answering the questions. You can input texts and documents about
yourself � including personal information, professional experience,
educational background, etc. � which are then embedded into a vector space
for dynamic retrieval. This AI clone can act as a virtual assistant or digital
representation, capable of handling queries and tasks in a manner that reflects
the your own knowledge, tone, style and mannerisms. ## Installation ###
@@ -30,38 +30,38 @@
the project directory cd clonellm # Install the package pip install . ``` ##
Usage ### Getting started **Step 1**. Gather documents that contain relavant
information about you. These documents form the base from which your AI clone
will learn to mimic your tone, style, and expertise. ```python from
langchain_core.documents import Document documents = [ Document
(page_content="My name is Mehdi Samsami."), open("cv.txt", "r").read(), ] ```
**Step 2**. Initialize an embedding model using CloneLLM's `LiteLLMEmbeddings`
-or Langchain's embeddings. Then, initialize a clone with your documents,
+or LangChain's embeddings. Then, initialize a clone with your documents,
embedding model, and your referred LLM. ```python from clonellm import
CloneLLM, LiteLLMEmbeddings embedding = LiteLLMEmbeddings(model="text-
embedding-ada-002") clone = CloneLLM(model="gpt-4-turbo", documents=documents,
embedding=embedding) ``` **Step 3**. Configure environment variables to store
API keys for embedding and LLM models. ```bash export OPENAI_API_KEY=sk-... ```
**Step 4**. Fit the clone to the data (documents). ```python clone.fit() ```
**Step 5**. Invoke the clone to ask questions. ```python clone.invoke("What's
your name?") # Response: My name is Mehdi Samsami. How can I help you? ``` ###
Models At its core, CloneLLM utilizes LiteLLM for interactions with various
LLMs. This is why you can choose from many different providers (100+ LLMs)
supported by LiteLLM, including Bedrock, Azure, OpenAI, Cohere, Anthropic,
Ollama, Sagemaker, HuggingFace, Replicate, etc. ### Document loaders You can
-use Langchain's document loaders to seamlessly import data from various sources
+use LangChain's document loaders to seamlessly import data from various sources
into `Document` format. Take, for example, text and HTML loaders: ```python #
!pip install unstructured from langchain_community.document_loaders import
TextLoader, UnstructuredHTMLLoader documents = TextLoader("cv.txt").load() +
UnstructuredHTMLLoader("linkedin.html").load() ``` Or JSON loader: ```python #
!pip install jq from langchain_community.document_loaders import JSONLoader
documents = JSONLoader( file_path='chat.json', jq_schema='.messages[].content',
text_content=False ).load() ``` ### Embeddings With `LiteLLMEmbeddings`,
CloneLLM allows you to utilize embedding models from a variety of providers
supported by LiteLLM. Additionally, you can select any preferred embedding
-model from Langchain's extensive range. Take, for example, the Hugging Face
+model from LangChain's extensive range. Take, for example, the Hugging Face
embedding: ```python # !pip install --upgrade --quiet sentence_transformers
from langchain_community.embeddings import HuggingFaceEmbeddings from clonellm
import CloneLLM import os os.environ["COHERE_API_KEY"] = "cohere-api-key"
embedding = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-
base-v2") clone = CloneLLM(model="command-xlarge-beta", documents=documents,
embedding=embedding) ``` Or, the Llama-cpp embedding: ```python # !pip install
--upgrade --quiet llama-cpp-python from langchain_community.embeddings import
@@ -119,12 +119,14 @@
bugs, or improving documentation. - � **Share:** Post about CloneLLM on
LinkedIn or other social platforms. Thank you for your interest in CloneLLM. We
look forward to seeing what you'll create with your AI clone! ## TODO - [x] Add
pre commit configuration file - [x] Add setup.py script - [x] Add support for
conversation history - [ ] Add support for RAG with no embedding (ingest the
entire context into the prompt) - [x] Add support for string documents - [x]
Fix mypy errors - [x] Rename `completion` methods to `invoke` - [x] Add support
-for streaming completion - [ ] Add support for custom system prompts - [ ] Add
-an attribute to return supported models - [x] Add initial version of README -
-[ ] Add documents - [ ] Add usage examples - [ ] Add initial unit tests - [x]
-Add GitHub workflow to run tests on PR - [x] Add GitHub workflow to publish to
-PyPI on release
+for streaming completion - [ ] Add support for custom system prompts - [x] Make
+`LiteLLMEmbeddings.all_embedding_models` a property - [ ] Add an attribute to
+`CloneLLM` to return supported models - [x] Add initial version of README - [ ]
+Describe `CloneLLM.clear_memory` method in README - [ ] Add documents - [x] Add
+usage examples - [x] Add unit tests for non-core modules - [ ] Add unit tests
+for core module - [x] Add GitHub workflow to run tests on PR - [x] Add GitHub
+workflow to publish to PyPI on release