GitLab

The change you added did not work with python2, it does not have encoding in .log.  This fixes it. feel free to test on Python3.  I tested python2 and I am really sue this works fine for python3.  Good catch.  sorry for not having it fixed earlier.

change cPickle to pickle for python3
specify  encoding='latin1' when call pickle.load()

Use the short path for TPUClusterResolver

The long path doesn't work for TensorFlow 1.6 because the .python package under cluster_resolver doesn't exist.

Add ClusterResolver capability to mnist_tpu.py

* Refactoring input function

* Updating name of variable

Removed "coming soon' in "Image Compression" title

lfads: Adding do_train_readin option allowing fixed readin matrices

Fix typo

Return dataset from input_fn

Fix tensorflow/models gan typos. fixes #3295

Fix file reading issue for python3 code

Fix wrong reference in ptb_word_lm.py

tcn: Fix undefined 'v' in Python 3

Small updates to DELF markdown files: dataset location and small troubleshooting fix.

For stitched models, the readin matrices and bias vectors are initialized to the
"alignment" matrix and bias specified in each dataset's .h5 file. If do_train_readin is
True, these will be trainable, and if not, they will be fixed.

`max_number_of_evaluation` -> `max_number_of_evaluations`.

Clear softmax_cross_entropy deprecation warning.

The `sparse` version is more efficient anyway.

I'm returning the labels shape [1] instead of []
because tf.accuracy fails otherwise.

Remove contacts of ex-Googler

Removed unused regularization ops