search

cszn / DnCNN

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising (TIP, 2017)
https://cszn.github.io/
1.45k stars 536 forks source link

error 0 in training #16

Closed ghost closed 7 years ago

ghost commented 7 years ago

Hi, i'm trying to train a model with another set of images and with CPU. I replaced the images in Train400 and changed in the code GPU = []; i got error value of 0 and NAN in memory

(parameter memory|2MB (5.6e+05 parameters)| data memory|NaN (for batch size 128)| train: epoch 01 dataset 00: 1/5106:error: 0.000000 train: epoch 01 dataset 00: 2/5106:error: 0.000000 )

i cant add image , this is copy paste : 

 layer|      0|      1|      2|      3|      4|      5|      6|      7|      8|      9|     10|     11|     12|     13|     14|     15|     16|     17|
  type|  input|   conv|   relu|   conv|  bnorm|   relu|   conv|  bnorm|   relu|   conv|  bnorm|   relu|   conv|  bnorm|   relu|   conv|  bnorm|   relu|
  name|    n/a| layer1| layer2| layer3| layer4| layer5| layer6| layer7| layer8| layer9|layer10|layer11|layer12|layer13|layer14|layer15|layer16|layer17|
---------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- support n/a 3 1 3 1 1 3 1 1 3 1 1 3 1 1 3 1 1 filt dim n/a 1 n/a 64 n/a n/a 64 n/a n/a 64 n/a n/a 64 n/a n/a 64 n/a n/a filt dilat n/a 1 n/a 1 n/a n/a 1 n/a n/a 1 n/a n/a 1 n/a n/a 1 n/a n/a num filts n/a 64 n/a 64 n/a n/a 64 n/a n/a 64 n/a n/a 64 n/a n/a 64 n/a n/a stride n/a 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 pad n/a 1 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0
rf size n/a 3 3 5 5 5 7 7 7 9 9 9 11 11 11 13 13 13
rf offset n/a 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
rf stride n/a 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
---------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- -------
data size NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN NaNxNaN
data depth NaN 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64 64
data num 128 128 128 128 128 128 128 128 128 128 128 128 128 128 128 128 128 128
---------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- ------- -------
data mem NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
param mem n/a 3KB 0B 144KB 1KB 0B 144KB 1KB 0B 144KB 1KB 0B 144KB 1KB 0B 144KB 1KB 0B

Thanks!

cszn commented 7 years ago

Please make sure "Train400" works well.

ghost commented 7 years ago

thanks