- 21 9月, 2018 7 次提交
-
-
由 Max Frei 提交于
Unified regex expression for both, Linux and Windows.
-
由 Adam Kelly 提交于
- Current implementation creates self.log_dir in set_log_dir() function, which creates an empty log directory if none exists. This causes find_last() to fail after creating a model because it finds this new empty directory instead of the previous training directory. - New implementation moves log_dir creation to the train() function to ensure it is only created when it will be used.
-
由 keineahnung2345 提交于
-
由 Waleed 提交于
-
由 Shivam K. Thakkar 提交于
VIA has changed JSON formatting in later versions. Now instead of a dictionary, "regions" has a list, see the issue #928
-
由 keineahnung2345 提交于
-
由 keineahnung2345 提交于
-
- 13 7月, 2018 2 次提交
- 12 7月, 2018 12 次提交
-
-
由 Corey Hu 提交于
-
由 Corey Hu 提交于
-
由 Corey Hu 提交于
-
由 Corey Hu 提交于
-
由 Corey Hu 提交于
-
由 Veemon 提交于
-
由 Waleed Abdulla 提交于
-
由 Nick 提交于
Add the possibility to exclude some sources from augmentation by passing a list of sources. This is useful when you want to retrain a model having few images.
-
由 Nick 提交于
Add an optional parameter for calling a list of keras.callbacks to be add to the original list.
-
由 rob 提交于
-
由 Waleed Abdulla 提交于
Improve previous commit to not break on older versions of Keras.
-
由 gtgalone 提交于
-
- 06 6月, 2018 1 次提交
-
-
由 Waleed Abdulla 提交于
Raise clear error if last training weights are not foundIf using the --weights=last (or --model=last) to resume trainingbut the weights are not found now it raises a clear error message.
-
- 05 6月, 2018 10 次提交
-
-
由 Austin Egri 提交于
modified comment for image augmentation line import to include correct 'pip3 install imgaug' instructions
-
由 Julien Rebetez 提交于
This allows one to set a callable in Config.BACKBONE to use a custom backbone model.
-
由 Julien Rebetez 提交于
-
由 Julien Rebetez 提交于
Fix matterport/Mask_RCNN#170
-
由 Julien Rebetez 提交于
Print a message when re-starting from saved epoch
-
由 ziyi 提交于
-
由 Ji Hou 提交于
fix typo
-
由 Waleed Abdulla 提交于
Reported here: https://github.com/matterport/Mask_RCNN/pull/500
-
由 Waleed Abdulla 提交于
-
由 yezqNLP 提交于
-
- 04 6月, 2018 1 次提交
-
-
由 Waleed Abdulla 提交于
-
- 22 4月, 2018 3 次提交
-
-
由 Ryan 提交于
Fixed the typo in the headline of the README.md file. "Spash" should be "Splash"
-
由 Waleed Abdulla 提交于
TF replaced keep_dims with keepdims a while ago and now shows a warning when using the old name.
-
由 Waleed Abdulla 提交于
A previous fix to let validation run across more than one batch caused an issue with multi-GPU training. The issue seems to be in how Keras averages loss and metric values, where it expects them to be scalars rather than arrays. This fix causes scalar outputs from a model to remain scalar in multi-GPU training.
-
- 16 4月, 2018 1 次提交
-
-
由 Kolesnikov Sergey 提交于
-
- 14 4月, 2018 1 次提交
-
-
由 Steven Hickson 提交于
-
- 13 4月, 2018 2 次提交
-
-
由 Waleed Abdulla 提交于
-
由 concerttttt 提交于
-