Hello,
I've managed to do datasets that look closely like yours but instead of having 15 sid I have only 2. So, I modified everything that was "15" to "2" in two_tier_spkdird.py. It bugs though.
Traceback (most recent call last):
File "two_tier_spkdird.py", line 621, in <module>
cost, h0 = train_fn(seqs, sids, h0, reset, mask)
File "/home/vincent/.local/lib/python2.7/site-packages/theano/compile/function_module.py", line 917, in __call__
storage_map=getattr(self.fn, 'storage_map', None))
File "/home/vincent/.local/lib/python2.7/site-packages/theano/gof/link.py", line 325, in raise_with_op
reraise(exc_type, exc_value, exc_trace)
File "/home/vincent/.local/lib/python2.7/site-packages/theano/compile/function_module.py", line 903, in __call__
self.fn() if output_subset is None else\
IndexError: Index out of bounds.
Apply node that caused the error: GpuAdvancedSubtensor1(SampleLevel.Embedding, GpuContiguous.0)
Toposort index: 298
Inputs types: [GpuArrayType<None>(float32, matrix), GpuArrayType<None>(int64, vector)]
Inputs shapes: [(256, 256), (2097152,)]
Inputs strides: [(1024, 4), (8,)]
Inputs values: ['not shown', 'not shown']
Inputs type_num: [11, 7]
Outputs clients: [[GpuReshape{2}(GpuAdvancedSubtensor1.0, TensorConstant{[ -1 4096]})]]
Backtrace when the node is created(use Theano flag traceback.limit=N to make it longer):
File "two_tier_spkdird.py", line 384, in <module>
prev_samples,spkids,
File "two_tier_spkdird.py", line 306, in sample_level_predictor
prev_samples)
File "/home/vincent/sampleRNN_acoustic_scene_generation/lib/ops.py", line 268, in Embedding
return vectors[indices.flatten()].reshape(output_shape)
File "/home/vincent/.local/lib/python2.7/site-packages/theano/gpuarray/type.py", line 675, in __getitem__
return _operators.__getitem__(self, *args)
File "two_tier_spkdird.py", line 384, in <module>
prev_samples,spkids,
File "two_tier_spkdird.py", line 306, in sample_level_predictor
prev_samples)
File "/home/vincent/sampleRNN_acoustic_scene_generation/lib/ops.py", line 268, in Embedding
return vectors[indices.flatten()].reshape(output_shape)
File "/home/vincent/.local/lib/python2.7/site-packages/theano/gpuarray/type.py", line 675, in __getitem__
return _operators.__getitem__(self, *args)
What do I have to modify to change the number of classes ? Thank you
Hello, I've managed to do datasets that look closely like yours but instead of having 15 sid I have only 2. So, I modified everything that was "15" to "2" in two_tier_spkdird.py. It bugs though.
What do I have to modify to change the number of classes ? Thank you