Help with Python (word embeddings) in Azure ML Studio
Budget: $30 – $250 AUD
I am looking for a Python expert who can help me troubleshoot and resolve an issue with word embeddings in Azure ML Studio.
Specifically, I need assistance with troubleshooting word embeddings in Azure ML Studio.
I have created some code as an adaptation from this tutorial: https://learn.microsoft.com/en-us/azure/cognitive-services/openai/tutorials/embeddings?tabs=command-line
I'm trying to run word embeddings on job descriptions and match them to user queries. The issue I have is that the ada_v2 embeddings are in a series format and the user embeddings in a list and I can't figure out how to solve it.
Specifically, I get the following error.
The function search_docs returns the following error
UFuncTypeError: ufunc 'multiply' did not contain a loop with signature matching types (dtype('<U34392'), dtype('<U34392')) -> dtype('<U34392')
When processing the ada_v2 embeddings I also get the following warning: When running the function to get the embeddings on the text I also get this warning: /tmp/ipykernel_26241/167280897.py:1: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
Attached a sample file from the data I processed
The desired outcome of this project is to successfully troubleshoot and resolve the issue with word embeddings in Azure ML Studio.
Specifically, I need assistance with troubleshooting word embeddings in Azure ML Studio.
I have created some code as an adaptation from this tutorial: https://learn.microsoft.com/en-us/azure/cognitive-services/openai/tutorials/embeddings?tabs=command-line
I'm trying to run word embeddings on job descriptions and match them to user queries. The issue I have is that the ada_v2 embeddings are in a series format and the user embeddings in a list and I can't figure out how to solve it.
Specifically, I get the following error.
The function search_docs returns the following error
UFuncTypeError: ufunc 'multiply' did not contain a loop with signature matching types (dtype('<U34392'), dtype('<U34392')) -> dtype('<U34392')
When processing the ada_v2 embeddings I also get the following warning: When running the function to get the embeddings on the text I also get this warning: /tmp/ipykernel_26241/167280897.py:1: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
Attached a sample file from the data I processed
The desired outcome of this project is to successfully troubleshoot and resolve the issue with word embeddings in Azure ML Studio.