how to train with new data

[sarthaknagarro]

Please provide us with the following information:

This issue is for a: (mark with an x)

- [ ] bug report -> please search issues before submitting
- [ X ] feature request
- [ X ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

I have deployed the setup successfully. Now I want to use my own pdf to train the model. I have added them in the storage account and them indexed the documents on the indexes "gptkbindex". It is able to do so but the UI fails after the indexer runs.

Any log messages given by the failure

Expected/desired behavior

It should work

OS and Version?

Windows 7, 8 or 10. Linux (which distribution). macOS (Yosemite? El Capitan? Sierra?) linux ubuntu 20.04 LTS

azd version?

run azd version and copy paste here.

Versions

azd version 1.0.1 (commit e0cd1aca716fa5d08704beade7dcc734fe68f5f1)

Mention any other details that might be useful

---

Thanks! We'll be in touch soon.

[KonTheCat]

I have been able to successfully train with my own data by doing the following, starting from a deployment with sample data:

1. delete the blob storage container.
2. delete the index in the search service.
3. delete the two models in the OpenAI service.
4. replace sample data with your own data.
5. Run azd up - this redeploys the resources deleted above and re-runs the process for adding new data.

[github-actions[bot]]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.

[pamelafox]

To upload new data, you can pass --removeall into prepdocs.py to remove the current docs, then replace data folder with your own documents and rerun prepdocs.sh As for the error you see, you'll need to check the full traceback in the App Service default_docker logs to find the traceback.

[github-actions[bot]]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.