We have the following config.yml file:
language: "../../language_models/nb_core_news_lg"
recipe: default.v1
# Read the following docs on how to fine-tune a model using config.
# https://rasa.com/docs/rasa/tuning-your-model#component-lifecycle
pipeline:
## Required pipeline components ##
- name: SpacyNLP
# note: this path is relative to ./live-bots/bot_name, where it's imported.
model: "../../language_models/nb_core_news_lg"
case_sensitive: False
On running the model (using rasa shell for example), then the following output is seen:
2022-07-27 10:43:46 INFO root - Starting Rasa server on http://0.0.0.0:5005
2022-07-27 10:43:48 INFO rasa.core.processor - Loading model models/20220725-101216-quick-game.tar.gz...
2022-07-27 10:43:49 INFO rasa.nlu.utils.spacy_utils - Trying to load SpaCy model with name '../../language_models/nb_core_news_lg'.
2022-07-27 10:43:54 INFO rasa.nlu.utils.spacy_utils - Trying to load SpaCy model with name '../../language_models/nb_core_news_lg'.
This process however requires about 2Gb of RAM. This is strange as the Rasa model 20220725-101216-quick-game.tar.gz only takes up about 50Mb of data, and the language model nb_core_news_lg takes up about 600Mb of data. Where is the extra ~1.3Gb of data coming from? Am I loading in the model twice in the config.yml file?