> ## Documentation Index
> Fetch the complete documentation index at: https://langchain-zh.cn/llms.txt
> Use this file to discover all available pages before exploring further.

# SageMaker 集成

> 使用 LangChain Python 与 SageMaker 嵌入模型集成。

让我们加载 `SageMaker Endpoints Embeddings` 类。如果您在 SageMaker 上托管例如您自己的 Hugging Face 模型，可以使用该类。

有关如何执行此操作的说明，请参见 [SageMaker 上使用 Hugging Face 进行自定义推理](https://www.philschmid.de/custom-inference-huggingface-sagemaker)。

**注意**：为了处理批量请求，您需要调整自定义 `inference.py` 脚本中 `predict_fn()` 函数内的返回行：

从

`return {"vectors": sentence_embeddings[0].tolist()}`

改为：

`return {"vectors": sentence_embeddings.tolist()}`。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
!pip3 install langchain boto3
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
import json
from typing import Dict, List

from langchain_community.embeddings import SagemakerEndpointEmbeddings
from langchain_community.embeddings.sagemaker_endpoint import EmbeddingsContentHandler


class ContentHandler(EmbeddingsContentHandler):
    content_type = "application/json"
    accepts = "application/json"

    def transform_input(self, inputs: list[str], model_kwargs: Dict) -> bytes:
        """
        Transforms the input into bytes that can be consumed by SageMaker endpoint.
        Args:
            inputs: List of input strings.
            model_kwargs: Additional keyword arguments to be passed to the endpoint.
        Returns:
            The transformed bytes input.
        """
        # Example: inference.py expects a JSON string with a "inputs" key:
        input_str = json.dumps({"inputs": inputs, **model_kwargs})
        return input_str.encode("utf-8")

    def transform_output(self, output: bytes) -> List[List[float]]:
        """
        Transforms the bytes output from the endpoint into a list of embeddings.
        Args:
            output: The bytes output from SageMaker endpoint.
        Returns:
            The transformed output - list of embeddings
        Note:
            The length of the outer list is the number of input strings.
            The length of the inner lists is the embedding dimension.
        """
        # Example: inference.py returns a JSON string with the list of
        # embeddings in a "vectors" key:
        response_json = json.loads(output.read().decode("utf-8"))
        return response_json["vectors"]


content_handler = ContentHandler()


embeddings = SagemakerEndpointEmbeddings(
    # credentials_profile_name="credentials-profile-name",
    endpoint_name="huggingface-pytorch-inference-2023-03-21-16-14-03-834",
    region_name="us-east-1",
    content_handler=content_handler,
)


# client = boto3.client(
#     "sagemaker-runtime",
#     region_name="us-west-2"
# )
# embeddings = SagemakerEndpointEmbeddings(
#     endpoint_name="huggingface-pytorch-inference-2023-03-21-16-14-03-834",
#     client=client
#     content_handler=content_handler,
# )
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
query_result = embeddings.embed_query("foo")
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
doc_results = embeddings.embed_documents(["foo"])
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
doc_results
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
```

***

<div className="source-links">
  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/i18n\zh-CN\oss\python\integrations\embeddings\sagemaker-endpoint.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>

  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>
</div>