i have rasa local (server 1) for coding and testing.
i have another server (server 2) for rasa deployment.
i commit the changes from local (server 1) to server (server 2) using GIT, so when i uploaded the cache folder into Rasa folder on the host (not pods) which is used by docker hub to create an image, i tried both options:
1)- I added the .cache folder to GIT and uploaded to server.
2)- i uploaded the cache folder manually on server 2.
You said i need to copy the cache folder to the container: which pod (rasa app, rasa production…) and how?
Could you set logging to info ? You should be able to see messages from transformers about whether the cached files were found (either loading file {} from cache at... or loading file {} from ...
Has this issues been resolved by now?
I’m also having memory issues on my rasa server deployment, so first I’m trying to start the rasa server locally without downloading any files.
I pre-downloaded all files according to the sample script above, placed them in the cache folder and referred to this folder in the config file using cache_dir. When running command rasa run --cors "*" --debug, I expect the local rasa server to initialize using the files in the folder, however, it still connects to the huggingface API and downloads the files.
Any help is appreciated!