GitLab

Set default (directories) lmdb datasets to 'train_lmdb' and 'test_lmdb'. Added check for lmdb datasets.

Bugfix: corrected relation between epochs and iterations used in Caffe2

See merge request !14

Fix hardcoded test looping to make it to process the size of the test dataset instead.

See merge request !13

# Conflicts:
#   pom.xml

Fix harcoded testing loop

See merge request !11

Release version

See merge request !10

Adapt cnnpredictor template

See merge request !9

Update README.md: Checkout tag v0.4.0 since other releases/commits are not supported. Added note for windows 10.

Removed template and generation for CNNBufferFile since it is not needed to read the Caffe2 protobuf files.

Corrected target code since training parameter "learning_rate_decay" is unsupported for the backend Caffe2.

Fix convolution layer bug

See merge request !7

Modified CNNTrainer.ftl: improved way to enter the optimizer parameters in template since the support checker for training parameters is already implemented.
Corrected  order of training parameters in target files based on the modification.

# Conflicts:
#	src/test/resources/target_code/CNNCreator_Alexnet.py
#	src/test/resources/target_code/CNNCreator_CifarClassifierNetwork.py
#	src/test/resources/target_code/CNNCreator_VGG16.py

CNNCreator.ftl: Modified path constants to make their value dynamic instead of hard coded during code generation. Modified methods save_net and load_net.

Adapt cnntrainer template

See merge request !6

Added cnntrainer which calls the "train" function from cnncreator with the specified training parameters.

Adapt cnncreator template

See merge request !5

Added indented block in all layer templates due to changes made in CNNCreator.ftl in order to avoid a python error