使用 Clarifai 进行嵌入¶

LlamaIndex 支持 Clarifai 的嵌入模型。

您必须拥有 Clarifai 账户和个人访问令牌（PAT）。
点击此处获取或创建 PAT。

请将 CLARIFAI_PAT 设置为环境变量，或者您可以将 PAT 作为参数传递给 ClarifaiEmbedding 类

In [ ]:

%pip install llama-index-embeddings-clarifai

%pip install llama-index-embeddings-clarifai

In [ ]:

!export CLARIFAI_PAT=YOUR_KEY

!export CLARIFAI_PAT=YOUR_KEY

如果您在 Colab 上打开此 Notebook，可能需要安装 LlamaIndex 🦙。

In [ ]:

!pip install llama-index

!pip install llama-index

模型可以通过完整 URL 或通过 model_name、用户 ID 和应用 ID 的组合进行引用。

In [ ]:

from llama_index.embeddings.clarifai import ClarifaiEmbedding

# Create a clarifai embedding class just with model_url, assuming that CLARIFAI_PAT is set as an environment variable
embed_model = ClarifaiEmbedding(
    model_url="https://clarifai.com/clarifai/main/models/BAAI-bge-base-en"
)

# Alternatively you can initialize the class with model_name, user_id, app_id and pat as well.
embed_model = ClarifaiEmbedding(
    model_name="BAAI-bge-base-en",
    user_id="clarifai",
    app_id="main",
    pat=CLARIFAI_PAT,
)

from llama_index.embeddings.clarifai import ClarifaiEmbedding # Create a clarifai embedding class just with model_url, assuming that CLARIFAI_PAT is set as an environment variable embed_model = ClarifaiEmbedding( model_url="https://clarifai.com/clarifai/main/models/BAAI-bge-base-en" ) # Alternatively you can initialize the class with model_name, user_id, app_id and pat as well. embed_model = ClarifaiEmbedding( model_name="BAAI-bge-base-en", user_id="clarifai", app_id="main", pat=CLARIFAI_PAT, )

In [ ]:

embeddings = embed_model.get_text_embedding("Hello World!")
print(len(embeddings))
print(embeddings[:5])

embeddings = embed_model.get_text_embedding("Hello World!") print(len(embeddings)) print(embeddings[:5])

嵌入文本列表

In [ ]:

text = "roses are red violets are blue."
text2 = "Make hay while the sun shines."

text = "roses are red violets are blue." text2 = "Make hay while the sun shines."

In [ ]:

embeddings = embed_model._get_text_embeddings([text2, text])
print(len(embeddings))
print(embeddings[0][:5])
print(embeddings[1][:5])

embeddings = embed_model._get_text_embeddings([text2, text]) print(len(embeddings)) print(embeddings[0][:5]) print(embeddings[1][:5])