Changelog

Introduce the new Dl4jCNNExplorer class (usage examples), which allows you to use zoo models or custom trained models to perform prediction on an individual image, without loading a .arff dataset.
Implement the ScoreCAM saliency map generation method and integrate it with the Dl4jCNNExplorer (usage examples).
Implement a progress bar (for command line and GUI) to aid usability with long-running tasks.

wekaDeeplearning4j-1.7.0.zip
0d03b8702268fa7c6a904d587bf3f4356d0ad5edca8f4a778bfb313863e6b319
wekaDeeplearning4j-cuda-10.0-1.7.0-linux-x86_64.zip
e06b70490b5caeab8ae1f11ac9afcd49ca8e671401183cbe562cc42858be63a6
wekaDeeplearning4j-cuda-10.0-1.7.0-windows-x86_64.zip
8e1076c8e9a534b96f4314da0d8dcfbf6982b1e0206770ea51e8c994f669a15b
wekaDeeplearning4j-cuda-10.1-1.7.0-linux-x86_64.zip
153fce7c627fccad6ddf13d0b1edf7ffb7d8d3b6e17db6947fb1da6bcc86e6ee
wekaDeeplearning4j-cuda-10.1-1.7.0-windows-x86_64.zip
97b3140f4641288d4ebd56b9c398afc2cb9fccc11036ab57b9db6808f0fef9f0
wekaDeeplearning4j-cuda-10.2-1.7.0-linux-x86_64.zip
a013a616b52bfda3541e10ecf720a9ee6e98842f599972e10634883ddcd697b3
wekaDeeplearning4j-cuda-10.2-1.7.0-windows-x86_64.zip
f94418525efde2488d9483abc8685bf6051061a8d72a65279bbbc0d29bf5b9c1

Provide feedback