How can we convert a t7 model to keras' h5 ? I am trying to do so for c3d-sports1m-kinetics.t7 that you can find here https://github.com/kenshohara/3D-ResNets/releases
The least I can ask for is a way to load the t7 model to python (pytorch) and then extract its weights, but I couldn't do it with the load_lua() function...
I get an error while trying to do it using this function https://github.com/pytorch/pytorch/blob/c6529f4851bb8ac95f05d3f17dea178a0367aaee/torch/utils/serialization/read_lua_file.py
The error that I get is the following :
Traceback (most recent call last):
File "convert_t7_to_hdf5.py", line 574, in <module>
a = load_lua("model.t7")
File "convert_t7_to_hdf5.py", line 571, in load_lua
return reader.read()
File "convert_t7_to_hdf5.py", line 542, in read
typeidx = self.read_int()
File "convert_t7_to_hdf5.py", line 440, in read_int
return self._read('i')
File "convert_t7_to_hdf5.py", line 431, in _read
result = struct.unpack(fmt, self.f.read(sz))
ValueError: read of closed file
As mentioned in this link, https://github.com/pytorch/pytorch/issues/15307#issuecomment-448086741
with torchfile package, the load was successful. You can dump the contents of model to a file and then understand the contents. Each layer information is stored as a dictionary. Knowing the model architecture would make it easier to parse the contents.
>>> import torchfile
>>> model = torchfile.load('c3d-sports1m-kinetics.t7')
>>> module = model.modules[0].modules[0]
>>> module.name
b'conv1a'
>>> module['weight'].shape
(64, 3, 3, 3, 3)
>>> module['bias'].shape
(64,)