search

microsoft / MMdnn

MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.
MIT License
5.8k stars 965 forks source link

when will caffe_emitter support output of layer with more than 1 blob? #844

Open hnuhchen opened 4 years ago

hnuhchen commented 4 years ago

Platform (like ubuntu 16.04/win10): ubuntu 16.04 Python version: python 3.5.2 Source framework with version (like Tensorflow 1.4.1 with GPU): pytorch or tensorflow Destination framework with version (like CNTK 2.3 with GPU): caffe Pre-trained model path (webpath or webdisk path):

Running scripts:

hello: In caffe_emitter, i found the code:

def emit_Slice(self, IR_node):
        pass

So, if i want to convert layers with more than 1 output blob(such as torch.split->slice(caffe)), the function emit_Slice() is passed, and the conversion is obviously failed.

linmajia commented 4 years ago

@hnuhchen , thank you very much for the feedback. We are looking into this issue.