Question Answering project based on using pinecone and CSVReader.

[JYPark-Code]

In the Question Answering task, my input is an Excel (.csv) file. After loading the data using CSVReader, how can I embed the Excel file? In my last project, I completed this task using PDFReader and splitting the contents with the "text_splitter" function. However, for a .csv file, I'm not sure which type of data splitting is needed for embedding in Pinecone. I've checked the official documentation and Chase's YouTube videos several times, but I'm still unsure how to solve this problem.

[caribbeanchill]

I'm trying to solve the same. Did you find a solution?

[JYPark-Code]

I've been working on a different project and feature, and I'm experiencing a delay in implementing an Excel or CSV file based on the Langchain project. From what I understand, using CSVReader from Langchain imports all the data from the Excel sheet without indexing. After doing some research on YouTube, I found out that we need to use the 'Agent' feature, and ChatGPT itself has to use numpy or dataframe code to query the index. Before this 'agent' process, we need to input the index that includes tabular data, but I'm not sure about how to create that pipeline.

[mcohoon04]

any additional progress with this yet?

[chirag-periwal]

@JYPark-Code did you find a solution around it?