segmentation_models icon indicating copy to clipboard operation
segmentation_models copied to clipboard

Error when trying to remove dropout layers

Open MyJumperBroke23 opened this issue 5 years ago • 1 comments

Hi, because coreml does not support "class 'efficientnet.model.get_dropout..FixedDropout", I am trying to remove the dropout layers.

I've tried

layers = [l for l in model.layers]
x = layers[0](layers[0].input)
for i in range(1, 68):
    layers[i].trainable = False
    x = layers[i](x)
for i in range(69, len(layers)):
    layers[i].trainable = False
    x = layers[i](x)

but this results in

ValueError: A merge layer should be called on a list of inputs.

How do I remove intermediate dropout layers?

MyJumperBroke23 avatar Apr 18 '20 16:04 MyJumperBroke23

hey, I know it's an old threat, but I have a similar problem now.. Did you find a solution to that?

elliestath avatar May 19 '22 12:05 elliestath