langchain/docs/extras/integrations/text_embedding/openai.ipynb

OpenAI

Let's load the OpenAI Embedding class.

In [6]:

from langchain.embeddings import OpenAIEmbeddings

In [29]:

embeddings = OpenAIEmbeddings()

In [30]:

text = "This is a test document."

In [31]:

query_result = embeddings.embed_query(text)

In [32]:

query_result[:5]

Out [32]:

[-0.003186025367556387,
 0.011071979803637493,
 -0.004020420763285827,
 -0.011658221276953042,
 -0.0010534035786864363]

In [33]:

doc_result = embeddings.embed_documents([text])

In [34]:

doc_result[0][:5]

Out [34]:

[-0.003186025367556387,
 0.011071979803637493,
 -0.004020420763285827,
 -0.011658221276953042,
 -0.0010534035786864363]

Let's load the OpenAI Embedding class with first generation models (e.g. text-search-ada-doc-001/text-search-ada-query-001). Note: These are not recommended models - see here

In [1]:

from langchain.embeddings.openai import OpenAIEmbeddings

In [23]:

embeddings = OpenAIEmbeddings(model="text-search-ada-doc-001")

In [24]:

text = "This is a test document."

In [25]:

query_result = embeddings.embed_query(text)

In [26]:

query_result[:5]

Out [26]:

[0.004452846988523035,
 0.034550655976098514,
 -0.015029939040690051,
 0.03827273883655212,
 0.005785414075152477]

In [27]:

doc_result = embeddings.embed_documents([text])

In [28]:

doc_result[0][:5]

Out [28]:

[0.004452846988523035,
 0.034550655976098514,
 -0.015029939040690051,
 0.03827273883655212,
 0.005785414075152477]

In [ ]:

# if you are behind an explicit proxy, you can use the OPENAI_PROXY environment variable to pass through
os.environ["OPENAI_PROXY"] = "http://proxy.yourcompany.com:8080"