rinuboney / ladder

Ladder network is a deep learning algorithm that combines supervised and unsupervised learning.

rinuboney/ladder Issues

Recurrent implementation
Closed a year ago2
z_est_bn = (z_est[l] - m) / v
Updated 5 years ago
why you put this line in code
Updated 5 years ago
PR to tensorflow
Closed 5 years ago6
Normalization & Gamma
Updated 6 years ago
Computation of running average of layer std and mean
Closed 6 years ago
Index out of range in ladder.py
Updated 6 years ago4
ladder for CIFAR-10
Updated 7 years ago
Bug Report
Updated 7 years ago
The effect of tf.train.ExponentialMovingAverage
Updated 7 years ago
Definition of prediction cost and correct prediction -- only labelled?
Closed 7 years ago1
Type error!
Updated 7 years ago1
Intermediate layers do not make use of gamma parameters
Closed 8 years ago2
Larger error for MNIST MLP baseline supervised compared to paper's results
Updated 8 years ago4
Labelled / unlabelled data split by batch_size=100 during evaluation
Updated 8 years ago2
Error when running ladder.py
Closed 8 years ago3
clean batch normalization should use batch mean and var for training
Updated 9 years ago13
incorrect mean cross entropy when batch sizes vary
Closed 9 years ago
learning rate declining faster than straight line
Closed 9 years ago1
initialization value 0 should not be multiplied by tf.Variable()
Closed 9 years ago3