Problem in db

[LakshyaTambi]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

[Yesterday 18:40] Harshika Sharma (Guest)
import getpass
import os
from langchain_cohere import CohereEmbeddings
 
os.environ["COHERE_API_KEY"] = "u4amCdXLmpaG8GEGJcjHAILcFb5YneUdMQWpe6sf"
 
embeddings = CohereEmbeddings(cohere_api_key=os.environ["COHERE_API_KEY"],
    model="embed-english-light-v3.0"
)  # It is mandatory to pass a model parameter to initialize the CohereEmbeddings object
 
text = "This is a test document."
query_result = embeddings.embed_query(text)
print(query_result)

Description

Traceback (most recent call last):
File "c:\Users\Dell\try1.py", line 10, in
query_result = cohere_embeddings.embed_query(text)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Dell\AppData\Roaming\Python\Python312\site-packages\langchain_cohere\embeddings.py", line 193, in embed_query
return self.embed([text], input_type="search_query")[0]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Dell\AppData\Roaming\Python\Python312\site-packages\langchain_cohere\embeddings.py", line 143, in embed
return [list(map(float, e)) for e in embeddings]
^^^^^^^^^^^^^^^^^^^
ValueError: could not convert string to float: 'float_'

System Info

all ok

[dosubot[bot]]

The error you're encountering, ValueError: could not convert string to float: 'float_', suggests that the embeddings returned by the embed_query method contain a string 'float_' instead of numerical values.

Here are a few steps to troubleshoot and resolve this issue:

1. Check API Key and Model: Ensure that the API key and model name are correct and valid. Sometimes, an incorrect API key or model name can lead to unexpected responses.

2. Update Dependencies: Make sure you have the latest version of the langchain_cohere package and its dependencies. You can update the package using:

   pip install --upgrade langchain_cohere

3. Verify Environment Configuration: Ensure that the COHERE_API_KEY environment variable is correctly set and accessible in your script.

4. Inspect the Response: Modify the embed_query method to print the raw response from the Cohere API. This can help you understand the structure of the response and identify any anomalies.

   def embed_query(self, text):
       response = self.embed([text], input_type="search_query")
       print(response)  # Add this line to inspect the response
       return response[0]

5. Example Code: Here is a corrected version of your example code with the necessary adjustments:

   import os
   from langchain_cohere import CohereEmbeddings
   
   os.environ["COHERE_API_KEY"] = "u4amCdXLmpaG8GEGJcjHAILcFb5YneUdMQWpe6sf"
   
   embeddings = CohereEmbeddings(
       apiKey=os.environ["COHERE_API_KEY"],
       model="embed-english-light-v3.0"
   )
   
   text = "This is a test document."
   query_result = embeddings.embedQuery(text)
   print(query_result)

If the issue persists, please provide the raw response from the Cohere API for further analysis. This will help in identifying the root cause of the problem.

To continue talking to Dosu, mention @dosu.

[maximega]

Seeing the same issue in my code - downgrading to langchain_cohere==0.1.9 seemed to fix it for me.