I am using Rasa Nlu server and aws bucket to host the trained models. I am trying to access the models from another rasa server. Basically I have 2 servers. one for training and another for parsing.
What is the best way to load the models to parse from another server? I am not able to figure this out
I setup an endpoint config and mentioned in doc and when i run this command
python -m rasa_nlu.server --config config.yml --path https://s3.region.amazonaws.com/bucket-name/my-models --endpoints /path_to_file/model-config.yml
It throws an error
File "/anaconda3/lib/python3.6/zipfile.py", line 1175, in _RealGetContents raise BadZipFile("File is not a zip file") zipfile.BadZipFile: File is not a zip file Exception ignored in: <bound method DataRouter.__del__ of <rasa_nlu.data_router.DataRouter object at 0x10f18d278>> Traceback (most recent call last): File "/anaconda3/lib/python3.6/site-packages/rasa_nlu/data_router.py", line 122, in __del__ AttributeError: 'DataRouter' object has no attribute 'pool'
My model server config file has model: url: “https://s3.region.amazonaws.com/buket-name/models-folder/”