Issues
- 3
- 2
- 1
Gated unit formulation
#30 opened by ngimel - 3
error while training
#26 opened by ELHoussineT - 0
- 4
- 0
How many epochs to train for CIFAR-10
#57 opened by erikdao - 0
How did you use conditional class label h
#56 opened by zhangdan8962 - 0
How to run pretrained model checkpoint.
#55 opened by CaiShilv - 7
Pretrained model download link broken
#35 opened by huanzhang12 - 39
- 2
about pixel-cnn/model.py, line38
#13 opened by Lucas2012 - 1
Object was never used (type <class 'tensorflow.python.framework.ops.Tensor'>):
#48 opened by kriz232121 - 0
How to run pretrained model checkpoint.
#54 opened by sharmayush01 - 0
why your code so weird?
#53 opened by johnzlj - 0
Pretrained model not available
#51 opened by vikmary - 0
GPU error & other things
#52 opened by cymqqqq - 0
Object_detection_image
#50 opened by kriz232121 - 0
what do these lines work for
#47 opened by share020 - 0
Dead link to pretrained model.
#46 opened by AlexanderMath - 0
about the shift operation
#45 opened by yuffon - 0
- 1
Training not starting for 64x64x3 images
#19 opened by christopherblum - 0
How are weights trained?
#41 opened by profpedromoises - 1
why has to change to discretisized loss function?
#38 opened by xfwu - 3
unexpected keyword argument 'keepdims'
#29 opened by knauth - 4
Not understand plus_in = inv_stdv * (centered_x + 1./255.) in discretized_mix_logistic_loss
#33 opened by blues-lin - 0
Training fails with other datasets
#39 opened by haihongtran - 2
The output of conv2d should be updated after g and b are updated in data dependent initialization.
#36 opened by bfs18 - 0
Understanding edge cases
#34 opened by ThomasDelteil - 3
Training results in NaN (commit 2b03725)
#27 opened by bliep - 1
Learning curve isn't discontinuous when i continued training it on a pre-trained model
#32 opened by powder21 - 2
Logistic output layer training slowly
#28 opened by weixsong - 0
7000 loss value when training cifar10
#31 opened by zoli333 - 2
How to understand the softmax sampling
#24 opened by weixsong - 0
Training pixelCNN on tiny imagenet dataset
#22 opened by ShirinVafaei - 1
- 6
- 1
- 0
PixelCNN source code
#16 opened by FuyuWang - 1
- 0
filter sizes
#15 opened by JohannesBrand - 1
Windows: No one_hot kernel in tensorflow-gpu-v1.0
#14 opened by DawyD - 0
- 3
- 0
Speeding up image generation up to 183x
#10 opened by PrajitR - 2
- 0
- 1
PixelCNN at test time
#1 opened by kolesman