Commits · 0ffdcc44a9288057be69ef9328a16dcc9526fa69 · Shashank Suhas / seminar-breakout

10 Feb, 2018 4 commits
- Accept tf.layers compatible argument names (#635) · 0ffdcc44
  Yuxin Wu authored Feb 10, 2018
```
* Use tflayers

* some docs

* Use decorator for argument renaming

* rename parse_args
```
  0ffdcc44
- remove the use of globalns in GANs and just use argparse · adca1cd7
  Yuxin Wu authored Feb 10, 2018
  
  adca1cd7
- docs change · 8d668003
  Yuxin Wu authored Feb 10, 2018
  
  8d668003
- Remove deprecated usage of msgpack (https://github.com/tensorflow/tensorflow/issues/16906 ) · 228aead3
  Yuxin Wu authored Feb 09, 2018
  
  228aead3
09 Feb, 2018 2 commits
- stupid bugfix · 4f4b0ba3
  Yuxin Wu authored Feb 09, 2018
  
  4f4b0ba3
- Support more options on TFEventWriter (#645) · 5467e48a
  Yuxin Wu authored Feb 09, 2018
  
  5467e48a
08 Feb, 2018 4 commits
- Correct average=False && use_nccl=False · 54c5a42d
  Yuxin Wu authored Feb 08, 2018
  
  54c5a42d
- Increase GPU counting robustness (#640) · f4f41711
  Aaron Gokaslan authored Feb 08, 2018
```
* Increase GPU counting robustness

Added native Tensorflow method to check number of GPUs.

* docs

* missing import

* fix import
```
  f4f41711
- try to fix build · 1ce585be
  Yuxin Wu authored Feb 08, 2018
  
  1ce585be
- Make "builder.grads" visible to trainers. · e1a879d7
  Yuxin Wu authored Feb 08, 2018
  
  e1a879d7
07 Feb, 2018 4 commits
- Fix inference with feedinput · 9bbdf94d
  Yuxin Wu authored Feb 07, 2018
  
  9bbdf94d
- Fallback to tf.contrib.layers.variance_scaling_initializer. · 6eb4b244
  Yuxin Wu authored Feb 07, 2018
  
  6eb4b244
- [Keras] summarize total loss (fix #634) · 32c83fea
  Yuxin Wu authored Feb 06, 2018
  
  32c83fea
- more logs/checks for RandomResize · 5b7d5106
  Yuxin Wu authored Feb 05, 2018
  
  5b7d5106
06 Feb, 2018 2 commits
- [FasterRCNN] add notes about using different dataset (#632) · 271ffad1
  Yuxin Wu authored Feb 06, 2018
  
  271ffad1
- Fix buffer inconsistency in parallel_map; Change chief_only in many callbacks. · 5a9d1362
  Yuxin Wu authored Feb 05, 2018
  
  5a9d1362
05 Feb, 2018 3 commits
- Translate data_format and activation to tflayers (#627) · ea60a630
  Yuxin Wu authored Feb 05, 2018
  
  ea60a630
- [imgaug] check xrange==yrange only when using ratio (fix #628) · 0e5299bb
  Yuxin Wu authored Feb 05, 2018
  
  0e5299bb
- Prefer fused_batch_norm in bn inference. · fe049874
  Yuxin Wu authored Feb 05, 2018
  
  fe049874
04 Feb, 2018 2 commits
- docs update; use abstract sockets for a3c · a46de342
  Yuxin Wu authored Feb 04, 2018
  
  a46de342
- Update Travis Tensorflow to latest version (#626) · 35a701c0
  Aaron Gokaslan authored Feb 03, 2018
  
  35a701c0
03 Feb, 2018 2 commits
- Add dilation_rate to Conv2D (#625) · c14d1496
  Aaron Gokaslan authored Feb 03, 2018
```
Added dilation_rate to Conv2D since Tensorflow supports it easily. Requires TF 1.5 or greater.
```
  c14d1496
- deprecate more from symbolic_functions · 3c62a25c
  Yuxin Wu authored Feb 03, 2018
  
  3c62a25c
02 Feb, 2018 2 commits
- average option in replicated trainers · fe5d4984
  Yuxin Wu authored Feb 02, 2018
  
  fe5d4984
- VariableHolder for layer_norm; Do not assert replicated variables. · 70c9ba8f
  Yuxin Wu authored Feb 01, 2018
  
  70c9ba8f
01 Feb, 2018 3 commits
- Fix typo; make ParamSetter non-chief-only · bf4d8938
  Yuxin Wu authored Feb 01, 2018
  
  bf4d8938
- set_chief_only in callbacks. · b1c5766d
  Yuxin Wu authored Jan 31, 2018
  
  b1c5766d
- update docs · c3a47386
  Yuxin Wu authored Jan 31, 2018
  
  c3a47386
31 Jan, 2018 7 commits
- bind option in zmq tools. more notes about horovod. · 869bc638
  Yuxin Wu authored Jan 31, 2018
  
  869bc638
- fix bug on augmentors; bind=False in send_dataflow_zmq · 28f36c44
  Yuxin Wu authored Jan 30, 2018
  
  28f36c44
- add model badge · c7e8ba78
  Yuxin Wu authored Jan 31, 2018
  
  c7e8ba78
- bump version · 28366584
  Yuxin Wu authored Jan 31, 2018
  
  28366584
- do not load devices by TF · 61bdff40
  Yuxin Wu authored Jan 30, 2018
  
  61bdff40
- Update utils.py (#616) · 53c4db69
  yselivonchyk authored Jan 30, 2018
```
* Update utils.py

Example of deterministic weight initialization

* Update utils.py
```
  53c4db69
- fix bug in PeriodicRunHooks · 6ca57de4
  Yuxin Wu authored Jan 30, 2018
  
  6ca57de4
30 Jan, 2018 1 commit
- Fix horovod training; import pyarrow without torch · 4eb2be09
  Yuxin Wu authored Jan 29, 2018
  
  4eb2be09
29 Jan, 2018 4 commits
- LR warmup in imagenet-resnet · 0086e156
  Yuxin Wu authored Jan 29, 2018
  
  0086e156
- update maskrcnn notes · 9bfbb0a4
  Yuxin Wu authored Jan 29, 2018
  
  9bfbb0a4
- remove inceptionv3 · dc13638b
  Yuxin Wu authored Jan 28, 2018
  
  dc13638b
- Add more checks in add_moving_summary · 2e058ee1
  Yuxin Wu authored Jan 28, 2018
  
  2e058ee1