qurator-spk/sbb_binarization
1
0
Fork 0
mirror of https://github.com/qurator-spk/sbb_binarization.git synced 2025-06-09 04:10:00 +02:00
Code Issues Releases Wiki Activity

comment out explicit GC calls / session management

Browse source
This commit is contained in:
Robert Sachunsky 2023-02-14 23:34:27 +00:00
parent 4086c6956a
commit 6dc83dd1a5
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
sbb_binarize/sbb_binarize.py
View file

@ -87,8 +87,8 @@ class SbbBinarizer:
predicted_patches = model.predict(image_patches, verbose=0) predicted_patches = model.predict(image_patches, verbose=0)
# We have to manually call garbage collection and clear_session here to avoid memory leaks. # We have to manually call garbage collection and clear_session here to avoid memory leaks.
# Taken from https://medium.com/dive-into-ml-ai/dealing-with-memory-leak-issue-in-keras-model-training-e703907a6501 # Taken from https://medium.com/dive-into-ml-ai/dealing-with-memory-leak-issue-in-keras-model-training-e703907a6501
gc.collect() #gc.collect()
tf.keras.backend.clear_session() #tf.keras.backend.clear_session()
# The result is a white-on-black image that needs to be inverted to be displayed as black-on-white image # The result is a white-on-black image that needs to be inverted to be displayed as black-on-white image
# We do this by converting the binary values to a boolean numpy-array and then inverting the values # We do this by converting the binary values to a boolean numpy-array and then inverting the values