WaveNet

WaveNet copied to clipboard

When generating: RuntimeError: CUDA out of memory

1

Hi, I am running on a GeForce RTX3070 with 8GB. I can train the model if I adjust the sample_size, which I interpret as batch_size. However when generating, this parameter...

GitHubGeniusOverlord

RuntimeError: cuda runtime error (2) : out of memory

6

Hi! I'm running the following command to train the model: $ python train.py --data_dir=./test/data --output_dir=./outputs The GPU I'm using has 16276MiB. However, I get an out of memory error immediately:...

ghost

wavenet/utils/data.py there should be a time shift between the input audio and the target audio , But not so in the code

yangpeng08

The information about my computer is as below: MacBook Pro (15-inch, 2017) Processor: 2.9 GHz Intel Core i7 GPU: 16 GB 2133 MHz LPDDR3 I check the list of GPUs...

kqu7

How to use this implementation for conditioned synthesis on spectrogram?

RahulBhalley

Non-causal dilated convolution?

1

Hello, I think your dilated convolution class, as implemented, is not causal because of the 0 padding. The last comment here proposes a solution that might work better: https://github.com/pytorch/pytorch/issues/1333

Artix18

sigmoid and tanh on the same tensor?

1

Hi, thanks for the codes! Although Figure 4 in the original paper is described as in your code, Eq. 2 (which I assume to be more correct) says there are...

keunwoochoi