paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-04-29 11:09:27 -05:00
Code Issues Packages Projects Releases Wiki Activity

Add some logging to the classifier

Browse Source
This commit is contained in:
shamoon 2025-01-18 07:37:51 -08:00
parent c8ec70c05f
commit c099a96c06
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
src/documents/classifier.py
View File

@ -239,6 +239,15 @@ class DocumentClassifier:
cache.set(CLASSIFIER_HASH_KEY, hasher.hexdigest(), CACHE_50_MINUTES)
cache.set(CLASSIFIER_VERSION_KEY, self.FORMAT_VERSION, CACHE_50_MINUTES)
return False
logger.debug(
f"""Found:
Last training: {self.last_doc_change_time}
Latest doc change: {latest_doc_change}
Last auto type hash: {self.last_auto_type_hash.hex() if self.last_auto_type_hash else None}
Current hash: {hasher.digest().hex()}
Result: Retraining required
""",
)
# subtract 1 since -1 (null) is also part of the classes.
@ -251,7 +260,7 @@ class DocumentClassifier:
logger.debug(
f"{docs_queryset.count()} documents, {num_tags} tag(s), {num_correspondents} correspondent(s), "
f"{num_document_types} document type(s). {num_storage_paths} storage path(es)",
f"{num_document_types} document type(s). {num_storage_paths} storage path(s)",
)
from sklearn.feature_extraction.text import CountVectorizer