Ran
|
Files
130
|
Run time
5s
|
Badge
Embed ▾
README BADGES
|
push
travis-ci
[WIP] Upgrade to TensorFlow 1.0.0 (#426) * Make edward compatible with Tensorflow 0.12 I have changed a few things in order to make edward compatible with TF 0.12: - `tf.initialize_all_variables()` calls have been renamed to `tf.global_variables_initializer()` - `tf.Variable.ref()` calls are replaced with `value()`. See: https://github.com/tensorflow/tensorflow/commit/c46f70c06#diff-ae1a8f7b66539f000615a4ab7e4b2151R460. `ref()` is an internal function now. But I'm not sure whether `value()` is the correct function to use here. - `tf.GraphKeys.VARIABLES` is replaced with `tf.GraphKeys.GLOBAL_VARIABLES`. As a side note, even after applying these changes, edward still doesn't work because of the bug introduced in TF 0.12rc0, see: https://github.com/tensorflow/tensorflow/commit/77cfa97ac for the fix. Therefore I had to use TF master branch. There are still 5 test failing, I'm looking into them. * Update TF to 0.12 in travis environment. * Adapt to new tf.contrib.distributions.QuantizedDistribution 0.12 API. * Remove tf.Variable.value() calls. Closes #367. * Replace all tf.GraphKeys.VARIABLES with tf.GraphKeys.GLOBAL_VARIABLES * Replace all tf.initialize_all_variables() with tf.global_variables_initializer() * Use new tf.contrib.distributions.TransformedDistribution API * update setup.py; travis.yml * run tf_upgrade.py on edward/,examples/,tests/ * fix pep8 errors * reverse order of tf.concat * go through each file and make manual api changes * use variable._ref(); fix empirical.py, test_beta_log_prob.py * fix bayesian_linear_regression_tensor.py
4780 of 5568 relevant lines covered (85.85%)
0.86 hits per line
Coverage | ∆ | File | Lines | Relevant | Covered | Missed | Hits/Line |
---|