wavenet$ KERAS_BACKEND=theano python wavenet.py predict with models/run_20160914_113209/config.json predict_seconds=1 sample_temperature=0.001
WARNING (theano.configdefaults): Only clang++ is supported. With g++, we end up with strange g++/OSX bugs.
Using Theano backend.
WARNING - root - Changed type of config entry "run_dir" from NoneType to unicode
WARNING - root - Changed type of config entry "optimizer.epsilon" from NoneType to float
INFO - wavenet - Running command 'predict'
WARNING - wavenet - No observers have been added to this run
INFO - wavenet - Started
INFO - predict - Using checkpoint from epoch: 37
INFO - predict - Saving to "models/run_20160914_113209/samples/sample_epoch-00037_01s__sample-temp-0.001_seed-920639969.wav"
INFO - build_model - Receptive Field: 4095 (928ms)
ERROR - wavenet - Failed after 0:00:03!
Traceback (most recent calls WITHOUT Sacred internals):
File "wavenet.py", line 316, in predict
model.load_weights(os.path.join(checkpoint_dir, last_checkpoint))
File "/Users/m/virtualenvs/wavenet/lib/python2.7/site-packages/keras/engine/topology.py", line 2500, in load_weights
self.load_weights_from_hdf5_group(f)
File "/Users/m/virtualenvs/wavenet/lib/python2.7/site-packages/keras/engine/topology.py", line 2552, in load_weights_from_hdf5_group
str(len(flattened_layers)) + ' layers.')
Exception: You are trying to load a weight file containing 69 layers into a model with 90 layers.
not sure what the problem is, but i think it might have to do with the model checkpoint in use?
Here's the issue I encounter:
wavenet$ KERAS_BACKEND=theano python wavenet.py predict with models/run_20160914_113209/config.json predict_seconds=1 sample_temperature=0.001 WARNING (theano.configdefaults): Only clang++ is supported. With g++, we end up with strange g++/OSX bugs. Using Theano backend. WARNING - root - Changed type of config entry "run_dir" from NoneType to unicode WARNING - root - Changed type of config entry "optimizer.epsilon" from NoneType to float INFO - wavenet - Running command 'predict' WARNING - wavenet - No observers have been added to this run INFO - wavenet - Started INFO - predict - Using checkpoint from epoch: 37 INFO - predict - Saving to "models/run_20160914_113209/samples/sample_epoch-00037_01s__sample-temp-0.001_seed-920639969.wav" INFO - build_model - Receptive Field: 4095 (928ms) ERROR - wavenet - Failed after 0:00:03! Traceback (most recent calls WITHOUT Sacred internals): File "wavenet.py", line 316, in predict model.load_weights(os.path.join(checkpoint_dir, last_checkpoint)) File "/Users/m/virtualenvs/wavenet/lib/python2.7/site-packages/keras/engine/topology.py", line 2500, in load_weights self.load_weights_from_hdf5_group(f) File "/Users/m/virtualenvs/wavenet/lib/python2.7/site-packages/keras/engine/topology.py", line 2552, in load_weights_from_hdf5_group str(len(flattened_layers)) + ' layers.') Exception: You are trying to load a weight file containing 69 layers into a model with 90 layers.
not sure what the problem is, but i think it might have to do with the model checkpoint in use?