Research on local documents doesn't work

[yuxmi]

I copied the code from the "Research on Local Documents" section, and all my local files in the DOC_PATH directory are plaintxt files. I keep getting the same error:

Traceback (most recent call last): File "/content/script.py", line 15, in report = asyncio.run(get_report(query=query, report_type=report_type, report_source=report_source)) File "/usr/lib/python3.10/asyncio/runners.py", line 44, in run return loop.run_until_complete(main) File "/usr/lib/python3.10/asyncio/base_events.py", line 649, in run_until_complete return future.result() File "/content/script.py", line 5, in get_report researcher = GPTResearcher(query=query, report_type=report_type, report_source=report_source) TypeError: GPTResearcher.init() got an unexpected keyword argument 'report_source'

[assafelovic]

Hey did you update to latest version pip install -U gpt-researcher

[arsaboo]

It is not working for me either. I am using docker and have added the full path to the .env.

@ElishaKay Do we need to mount the path in the container?

[ElishaKay]

@yuxmi, you might have an easier path via Docker Desktop:

Step 1: Install Docker Desktop

Step 2: Walk through this Tutorial

@arsaboo another requirement I should have mentioned: the path to your files should be within the code of this project. That will enable you to skip the complexity of mounting stuff into Docker

Are you working off the latest version of master? Please share the error you're getting

[arsaboo]

@ElishaKay I am not seeing any error, but it is not reading any local files. Here are the contents of my gpt-researcher folder (which has the my-docs folder) Here's how I have added it to my .env:

DOC_PATH=./my-docs

[ElishaKay]

@arsaboo what type of files are you trying to read? It looks like there was a bug with reading CSV's (required the pandas dependency)

I opened the PR for that here

[arsaboo]

I don't think I saw an error. It's just that the local files were not used. I used a pdf file

[ElishaKay]

@arsaboo ask chat.openai to help you debug the PDF Loader in document.py

Write in the prompt that you want GPT to generate a python file to debug the PDF Loader library in the above documents.py file

Then place the gpt-generated debugger python file within the gpt-researcher repo & run it from the command line. That should hopefully help you get closer to the root cause

[amscosta]

Hi It is working for me using my local DOC_PATH="/home/c0274/docs-to-gpt-researcher" and using the Quick Start instructions(steps 1,2,3).

However is not working using the following suggested snippet : from gpt_researcher import GPTResearcher import asyncio

async def get_report(query: str, report_type: str, report_source: str) -> str: researcher = GPTResearcher(query, report_type,report_source) research_result = await researcher.conduct_research() report = await researcher.write_report() return report

if name == "main": query = "what is the effect of the speed of lvad support?" report_type = "research_report" report_source = "local" report = asyncio.run(get_report(query, report_type,report_source)) print(report)

It keeps searching the web only. I tried : report_source = "local" report_source = "documents" report_source = "path/to/my/pdf" report_source = "path/to/my/pdf/my.pdf"

No success.

[assafelovic]

@amscosta did you try upgrading gpt-researcher to latest? pip install gpt-researcher -U

Im closing for now since this is stale but please feel free to open if the issue persists. Thanks!