David Chen

I have the same problem when I try to run the examples/mnist/run.sh the dnn.training.log just show: Traceback (most recent call last): File "C:/pdnn-master/cmds/run_DNN.py", line 27, in from models.dnn import DNN...

Hi, I also have this problem while I was trying to train a simple digits speech recognition by using DNN. After I got the mfcc features from Kaldi in .scp...

Hi, I recently try to use IMU with mono version and also get bad result. After I check the scale result output from TryInitVIO, I found out it is negative...

Hi! I change the clipnorm to 1 like @aglotero said, but with more gru layers(1 convolutional, 7 gru with 1000 nodes each, 1 full connection).  I found that the...

@dylanbfox thanks for quick response! When the clipnorm is the default value(100), i get NaNs after some iterations. Then I came to read this issue, and try to train the...

@dylanbfox my learning rate is 2e-4, the default value. In my experience this is really small. Maybe I am wrong. I will try smaller value and keeping the clipnorm higher....

I set the learning rate to 2e-4, clipnorm back to 100, training with LibriSpeech-clean-100, and my model structure is 1 convolutional, 7 gru with 1000 nodes each, 1 full connection,...

I have the same error. Is this the version problem?

@amir-abdi thanks for your information, I'll give it a try!