I attached NetVlad layer to my model and I get an error when I try to save it. My guess is it's because the model is partially in Keras and partially in Tensorflow. Wonder if you had any issues with that?
Traceback (most recent call last):
File "train_model_i3dnetvladclr.py", line 994, in <module>
train_model()
File "train_model_i3dnetvladclr.py", line 735, in train_model
i3d_rgb_full.save('1.hdf5')
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/keras/engine/network.py", line 1152, in save
save_model(self, filepath, overwrite, include_optimizer)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/keras/engine/saving.py", line 449, in save_wrapper
save_function(obj, filepath, overwrite, *args, **kwargs)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/keras/engine/saving.py", line 541, in save_model
_serialize_model(model, h5dict, include_optimizer)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/keras/engine/saving.py", line 130, in _serialize_model
model_config = json.dumps(model_config, default=get_json_type)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/json/__init__.py", line 238, in dumps
**kw).encode(obj)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/json/encoder.py", line 199, in encode
chunks = self.iterencode(o, _one_shot=True)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/json/encoder.py", line 257, in iterencode
return _iterencode(o, 0)
File "/home/ubuntu/anaconda3/envs/tensorflow_p36/lib/python3.6/site-packages/keras/engine/saving.py", line 120, in get_json_type
raise TypeError('Not JSON Serializable: %s' % (obj,))
TypeError: Not JSON Serializable: ?
terminate called without an active exception
Aborted (core dumped)
I attached NetVlad layer to my model and I get an error when I try to save it. My guess is it's because the model is partially in Keras and partially in Tensorflow. Wonder if you had any issues with that?