pytorch2keras
pytorch2keras copied to clipboard
gmalivenko
PyTorch to Keras model convertor
**Describe the bug** KeyError: '31_np'. **To Reproduce** `class PCN2(nn.Module): def __init__(self): super().__init__() self.conv1 = nn.Conv2d(3, 20, kernel_size=3, stride=1) self.conv2 = nn.Conv2d(20, 40, kernel_size=3, stride=1) self.conv3 = nn.Conv2d(40, 70, kernel_size=2, stride=1)...
**Describe the bug** A get the error : Layer weight shape (4, 19, 4, 300) not compatible with provided weight shape (4, 19, 1, 300) **To Reproduce** ``` class ConvNet(nn.Module):...
Thanks for this nice project. I think `tensorflow` should not be in the `requirements.txt` file because if `tensorflow-gpu` is also installed, it's overwritten and this messes things up.
Hello, I'm trying to convert a version of MobileNetv2 to keras, and the conversion almost finishes but fails in convert_gather. I'm trying to figure out how the code gets the...
**Describe the bug** I am trying to convert a pytorch model that has a PixelShuffle layer. The layer is converted as a Lambda layer, but unfortunately, it doesn't work properly....
`graph node: CRNN type: onnx::Transpose inputs: ['68'] outputs: ['CRNN'] name in state_dict: attrs: {'perm': [2, 0, 1]} is_terminal: False Converting transpose ... !!! Cannot permute batch dimension. Result may be...
It seems that there is an error in convert.convert_constant function, where it is defined as ``` def convert_constant(params, w_name, scope_name, inputs, layers, weights): def target_layer(params=params): return keras.backend.constant(np.float32(params['value'])) lambda_layer = keras.layers.Lambda(target_layer)...
Add support of recurrent layers to the converter: - [ ] GRU - [ ] LSTM
Hi, once again thanks for awesome work. It really helps with shortening pytorch to production path (pytorch=>keras=>tf). I was running my pytorch model based on inception4 architecture. I encountered this...
