Rasa X training failed with Spacy pipeline

[Deprecated] Rasa X Community Edition
1

Hi, I deployed Rasa X using the default server quick-install method. I was able to connect my git repository. But when I train a model, the training fails. After checking the logs for rasa-worker pod, I found out that it’s failing to load the Spacy model.

rasa.nlu.model.InvalidModelError: Please confirm that en_core_web_sm is an available spaCy model. You need to download one upfront. For example:
python -m spacy download en_core_web_md
More informaton can be found on https://rasa.com/docs/rasa/components#spacynlp

2021-05-31 07:06:45 ERROR    rasa.server  - An unexpected error occurred during training. Error: Please confirm that en_core_web_sm is an available spaCy model. You need to download one upfront. For example:
python -m spacy download en_core_web_md
More informaton can be found on https://rasa.com/docs/rasa/components#spacynlp

I think I need to run python -m spacy download en_core_web_md and link the mode. But how can I do that when deploying using Rasa X?

2

Use en_core_web_md instead in the configuration.

If you want to use sm or lg models, you’d either need to execute into the containers and install them, or create a new image and deploy it in the cluster.

3

I have the same use case but I want to install a german spacy model. How would I go about installing that? I tried installing it inside the docker container, but I can’t seem to install anything inside the rasa containers:

ERROR: Could not install packages due to an OSError: [Errno 13] Permission denied: ‘/opt/venv/lib/python3.8/site-packages/wheel’ Check the permissions.

WARNING: You are using pip version 21.1.3; however, version 21.2.4 is available. You should consider upgrading via the ‘/opt/venv/bin/python -m pip install --upgrade pip’ command. rasa@a98e74d54097:~$ pip install -U pip setuptools wheel --user ERROR: Can not perform a ‘–user’ install. User site-packages are not visible in this virtualenv. WARNING: You are using pip version 21.1.3; however, version 21.2.4 is available. You should consider upgrading via the ‘/opt/venv/bin/python -m pip install --upgrade pip’ command.

And I also don’t know which container needs it:image