Closed lc82111 closed 7 years ago
Thank you for your keras implementation which is very helpfull. But, I found several inconsistencies. For example, the first conv layer should do stride (2, 2) conv. For example, the residual block should have THREE rather than TWO conv layer.
Fixed. Its not an exact match to the original resnet paper, because it is based on the Keras port of the ResNet50 model.
Thank you for your keras implementation which is very helpfull. But, I found several inconsistencies. For example, the first conv layer should do stride (2, 2) conv. For example, the residual block should have THREE rather than TWO conv layer.