paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-09-03 01:56:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

Cache classifier loading with lru_cache

Browse Source
This commit is contained in:
shamoon
2025-08-30 19:21:43 -07:00
parent 875dc6602b
commit fc4cb08bda
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/documents/classifier.py
View File

@@ -4,6 +4,7 @@ import logging
import pickle import pickle
import re import re
import warnings import warnings
from functools import lru_cache
from hashlib import sha256 from hashlib import sha256
from pathlib import Path from pathlib import Path
from typing import TYPE_CHECKING from typing import TYPE_CHECKING
@@ -50,6 +51,7 @@ class ClassifierModelCorruptError(Exception):
pass pass
@lru_cache(maxsize=1)
def load_classifier(*, raise_exception: bool = False) -> DocumentClassifier | None: def load_classifier(*, raise_exception: bool = False) -> DocumentClassifier | None:
if not settings.MODEL_FILE.is_file(): if not settings.MODEL_FILE.is_file():
logger.debug( logger.debug(
@@ -61,6 +63,11 @@ def load_classifier(*, raise_exception: bool = False) -> DocumentClassifier | No
classifier = DocumentClassifier() classifier = DocumentClassifier()
try: try:
classifier.load() classifier.load()
logger.debug("classifier_id=%s", id(classifier))
logger.debug(
"classifier_data_vectorizer_hash=%s",
classifier.data_vectorizer_hash,
)
except IncompatibleClassifierVersionError as e: except IncompatibleClassifierVersionError as e:
logger.info(f"Classifier version incompatible: {e.message}, will re-train") logger.info(f"Classifier version incompatible: {e.message}, will re-train")