ModelScope

ModelScope (Home | GitHub) is built upon the notion of “Model-as-a-Service” (MaaS). It seeks to bring together most advanced machine learning models from the AI community, and streamlines the process of leveraging AI models in real-world applications. The core ModelScope library open-sourced in this repository provides the interfaces and implementations that allow developers to perform model inference, training and evaluation.

This will help you get started with ModelScope embedding models using LangChain.

Overview

Integration details

Provider	Package
ModelScope	langchain-modelscope-integration

Setup

To access ModelScope embedding models you’ll need to create a/an ModelScope account, get an API key, and install the langchain-modelscope-integration integration package.

Credentials

Head to ModelScope to sign up to ModelScope.

import getpass
import os

if not os.getenv("MODELSCOPE_SDK_TOKEN"):
    os.environ["MODELSCOPE_SDK_TOKEN"] = getpass.getpass(
        "Enter your ModelScope SDK token: "
    )

Installation

The LangChain ModelScope integration lives in the langchain-modelscope-integration package:

%pip install -qU langchain-modelscope-integration

Instantiation

Now we can instantiate our model object:

from langchain_modelscope import ModelScopeEmbeddings

embeddings = ModelScopeEmbeddings(
    model_id="damo/nlp_corom_sentence-embedding_english-base",
)

Downloading Model to directory: /root/.cache/modelscope/hub/damo/nlp_corom_sentence-embedding_english-base

2024-12-27 16:15:11,175 - modelscope - WARNING - Model revision not specified, use revision: v1.0.0
2024-12-27 16:15:11,443 - modelscope - INFO - initiate model from /root/.cache/modelscope/hub/damo/nlp_corom_sentence-embedding_english-base
2024-12-27 16:15:11,444 - modelscope - INFO - initiate model from location /root/.cache/modelscope/hub/damo/nlp_corom_sentence-embedding_english-base.
2024-12-27 16:15:11,445 - modelscope - INFO - initialize model from /root/.cache/modelscope/hub/damo/nlp_corom_sentence-embedding_english-base
2024-12-27 16:15:12,115 - modelscope - WARNING - No preprocessor field found in cfg.
2024-12-27 16:15:12,116 - modelscope - WARNING - No val key and type key found in preprocessor domain of configuration.json file.
2024-12-27 16:15:12,116 - modelscope - WARNING - Cannot find available config to build preprocessor at mode inference, current config: {'model_dir': '/root/.cache/modelscope/hub/damo/nlp_corom_sentence-embedding_english-base'}. trying to build by task and model information.
2024-12-27 16:15:12,318 - modelscope - WARNING - No preprocessor field found in cfg.
2024-12-27 16:15:12,319 - modelscope - WARNING - No val key and type key found in preprocessor domain of configuration.json file.
2024-12-27 16:15:12,319 - modelscope - WARNING - Cannot find available config to build preprocessor at mode inference, current config: {'model_dir': '/root/.cache/modelscope/hub/damo/nlp_corom_sentence-embedding_english-base', 'sequence_length': 128}. trying to build by task and model information.

Indexing and Retrieval

Embedding models are often used in retrieval-augmented generation (RAG) flows, both as part of indexing data as well as later retrieving it. For more detailed instructions, please see our RAG tutorials.

Below, see how to index and retrieve data using the embeddings object we initialized above. In this example, we will index and retrieve a sample document in the InMemoryVectorStore.

# Create a vector store with a sample text
from langchain_core.vectorstores import InMemoryVectorStore

text = "LangChain is the framework for building context-aware reasoning applications"

vectorstore = InMemoryVectorStore.from_texts(
    [text],
        embedding=embeddings,
)

# Use the vectorstore as a retriever
retriever = vectorstore.as_retriever()

# Retrieve the most similar text
retrieved_documents = retriever.invoke("What is LangChain?")

# show the retrieved document's content
retrieved_documents[0].page_content

/root/miniconda3/envs/langchain/lib/python3.10/site-packages/transformers/modeling_utils.py:1113: FutureWarning: The `device` argument is deprecated and will be removed in v5 of Transformers.
  warnings.warn(
/root/miniconda3/envs/langchain/lib/python3.10/site-packages/transformers/modeling_utils.py:1113: FutureWarning: The `device` argument is deprecated and will be removed in v5 of Transformers.
  warnings.warn(

'LangChain is the framework for building context-aware reasoning applications'

Direct Usage

Under the hood, the vectorstore and retriever implementations are calling embeddings.embed_documents(...) and embeddings.embed_query(...) to create embeddings for the text(s) used in from_texts and retrieval invoke operations, respectively.

You can directly call these methods to get embeddings for your own use cases.

Embed single texts

You can embed single texts or documents with embed_query:

single_vector = embeddings.embed_query(text)
print(str(single_vector)[:100])  # Show the first 100 characters of the vector

[-0.6046376824378967, -0.3595953583717346, 0.11333226412534714, -0.030444221571087837, 0.23397332429

Embed multiple texts

You can embed multiple texts with embed_documents:

text2 = (
    "LangGraph is a library for building stateful, multi-actor applications with LLMs"
)
two_vectors = embeddings.embed_documents([text, text2])
for vector in two_vectors:
    print(str(vector)[:100])  # Show the first 100 characters of the vector

[-0.6046381592750549, -0.3595949709415436, 0.11333223432302475, -0.030444379895925522, 0.23397321999
[-0.36103254556655884, -0.7602502107620239, 0.6505364775657654, 0.000658963865134865, 1.185304522514

API Reference

For detailed documentation on ModelScopeEmbeddings features and configuration options, please refer to the API reference.