NBoost is a scalable, search-api-boosting platform for deploying transformer models to improve the relevance of search results on different platforms (i.e. Elasticsearch)
Apache License 2.0
675
stars
69
forks
source link
When using QA: "....0-py3.7.egg/nboost/plugins/models/__init__.pyline 24, in resolve_model" #46
I have the distilbert-base-uncased-distilled-squad model downloaded into the .cache directory and have the following configuration. All values are at default except these.
The distilber model does not seem to be resolvable as it isn't in the CLASS_MAP.
I had the qa.model_dir in the above script but that also does not resolve the issue.
Question: should I rebuild nboost after adding the model to the CLAAS_MAP?
File "/home/steph/anaconda3/envs/bert/lib/python3.7/site-packages/nboost-0.3.0-py3.7.egg/nboost/plugins/models/__init__.py", line 24, in resolve_model
raise ImportError('Class "%s" not in %s.' % CLASS_MAP.keys())
TypeError: not enough arguments for format string
I have the distilbert-base-uncased-distilled-squad model downloaded into the .cache directory and have the following configuration. All values are at default except these. The distilber model does not seem to be resolvable as it isn't in the CLASS_MAP.
I had the qa.model_dir in the above script but that also does not resolve the issue. Question: should I rebuild nboost after adding the model to the CLAAS_MAP?