Rasa X - spacy - docker images

I tried to deploy a Rasa Open Source project on Google Cloud VM with Rasa X server through the docker-compose method.

However, I have the following errors with rasa-production and rasa-worker docker images:

rasa-worker_1 | rasa.nlu.model.InvalidModelError: Model 'fr' is not a linked spaCy model. Please download and/or link a spaCy model, e.g. by running: rasa-worker_1 | python -m spacy download en_core_web_md rasa-worker_1 | python -m spacy link en_core_web_md en

rasa-production_1 | rasa.nlu.model.InvalidModelError: Model 'fr' is not a linked spaCy model. Please download and/or link a spaCy model, e.g. by running: rasa-production_1 | python -m spacy download en_core_web_md rasa-production_1 | python -m spacy link en_core_web_md en

By the way, my chatbot is in french and locally I had to download french spacy corpus and link it to the shortcut fr that I did with :

python -m spacy download fr_core_news_md

python -m spacy link fr_core_news_md fr

I’m not sure but I think that this step is not included in the default rasa docker images.

My question: would it be possible to modify the docker images so that this steps are included ?

If not, how can I add these steps directly from the Google Cloud VM ?