- 11 5月, 2019 3 次提交
-
-
由 Haoyu Zhang 提交于
* Fix trivial model to work properly with fp16 * Add comment on manual casting
-
由 Haoyu Zhang 提交于
Previously we had one dense layer in trivial model. The weight was [224*224*3, num_classes]. Using two dense layers, the weights are [224*224*3, 1] and [1, num_classes].
-
由 Haoyu Zhang 提交于
* Do not report metrics in performance benchmarks * Rename flag
-
- 10 5月, 2019 4 次提交
-
-
由 Haoyu Zhang 提交于
-
由 Haoyu Zhang 提交于
* Modified tweaked tests to use tensor learning rate
-
由 Toby Boyd 提交于
* Add first benchmark and return stats. * Remove print statements update training steps. * Revert print T: in print statement. * Remove print(stats) * add 2 gpu accuracy test for base. * Fixed total_batch_size when using gpu + gFile deprecations. * 8 GPU test name fix * Add 4 and 8 GPU tests. * typo fixes. * Clean up test names and methods. * bleu uncased. docstring format fix.
-
由 Haoyu Zhang 提交于
* Add learning rate tensor. This makes training slower * Improve LearningRateSchedule with better efficiency * Fix lint error * Replace constant definition with existing one
-
- 09 5月, 2019 1 次提交
-
-
由 Toby Boyd 提交于
-
- 08 5月, 2019 2 次提交
-
-
由 Menglong Zhu 提交于
246873701 by menglong: Missing __init__.py under meta_architectures/ -- 246857392 by menglong: Standardize proto namespace: lstm_object_detection.protos -- 246625127 by menglong: Internal changes. -- 246596481 by menglong: Add License -- 246580605 by menglong: Internal changes -- 246344626 by menglong: Open source interleaved mobilenet v2 model. -- 244893883 by menglong: Introduce multi_input_decoder for interleaved model. -- 244461016 by menglong: Add pre-bottleneck operation to lstm cells to support interleaved model. -- 244052176 by menglong: Update README -- 244020495 by menglong: Add test to rnn_decoder. -- 243704250 by menglong: Duplicate assignment. -- 243091836 by menglong: Move LSTMSSD meta arch into separate folder -- 242900337 by menglong: Modified mobilenet definition for LSTM-SSD -- 242773195 by menglong: Release GroupedConvLSTMCell implementation: https://arxiv.org/abs/1903.10172 -- 242574736 by menglong: Introduce module for quantizated training. -- 242544306 by menglong: lstm_ssd_meta_arch updates, added test rename: - LSTMMetaArch to LSTMSSDMetaArch - LSTMFeatureExtractor to LSTMSSDFeatureExtractor -- 241986236 by menglong: Move lstm quantization utils to 3rd party. -- 225922488 by yinxiao: Training pipeline fixes. -- 224839137 by yinxiao: Issue fix for lstm object detecion sample config. -- 224246947 by menglong: Fix logging module import -- PiperOrigin-RevId: 246873701
-
由 Haoyu Zhang 提交于
-
- 07 5月, 2019 3 次提交
-
-
由 Toby Boyd 提交于
-
由 Yukun Zhu 提交于
* deeplab quantize * Fix bug in train.py * Create quantize.md
-
由 Haoyu Zhang 提交于
-
- 04 5月, 2019 1 次提交
-
-
由 Haoyu Zhang 提交于
* Enable CuDNN BatchNorm spatial persistent by default; Remove 2nd zero padding layer * Apply scale=False and fused=True consistently to BatchNorm layers * Undo remove padding layer * Replace zero padding with padding attribute in max pooling for better performance * Resolve comments * Revert "Replace zero padding with padding attribute in max pooling for better performance" This reverts commit ad49db057c800ecac008eec1057005bd2c08ac73.
-
- 03 5月, 2019 3 次提交
-
-
由 Reed 提交于
-
由 Krishnan Srinivasan 提交于
* updated mujoco api calls to data, and tensorflow call to CriticalSection (moved from contrib.framework) * changed efficient-hrl envs so self.physics points to correct PyMj object based on mujoco_py version * corrected mujoco version number checkm in AntEnv.physics and PointEnv.physics properties * fix AntEnv.set_xy by reverting to using self.physics
-
由 Reed 提交于
-
- 02 5月, 2019 1 次提交
-
-
由 Reed 提交于
This options allows the new tf.train.experimental.enable_mixed_precision_graph_rewrite() function to be used for fp16, instead of manual casts.
-
- 01 5月, 2019 2 次提交
-
-
由 André Araujo 提交于
* Internal change. PiperOrigin-RevId: 185564155 * Fix small bug when reading DELF features when file is empty. A new test is added that catches this bug. PiperOrigin-RevId: 213839503 * Refactors DELF example code to expose a function to create a DELF feature extractor. PiperOrigin-RevId: 241492615 * Merged commit includes the following changes: 244073180 by Andre Araujo: Internal change -- 243646498 by Andre Araujo: Detect and save bounding boxes for a list of images. A new proto is added, along with auxiliary read/write functions, and tests. -- PiperOrigin-RevId: 244073180 * Merged commit includes the following changes: 244872693 by Andre Araujo: Small change. -- 244871213 by Andre Araujo: Small edit to DELF extractor function. -- 244790715 by Andre Araujo: internal automated change -- 244773146 by Andre Araujo: Add option to visualize extracted boxes, and change slightly model input. -- 244288992 by Andre Araujo: Script to cluster DELF features using K-means. -- 244275164 by Andre Araujo: Scripts to extract DELF features and boxes from Revisited Oxford/Paris datasets, used in Detect-to-Retrieve paper. -- PiperOrigin-RevId: 244872693 * Merged commit includes the following changes: PiperOrigin-RevId: 244876167 * Updates instructions to include new released models, instructions for DELF and detector extraction, etc. * Small fix * Merged commit includes the following changes: 244909336 by Andre Araujo: Update to DELF config example to use latest model. -- PiperOrigin-RevId: 244909336 * Revert "Merged commit includes the following changes:" This reverts commit f41be9f159330f62846d5d8b3d3d5d55cd1874f4. * Update delf_config_example model * Small edits to detection instructions * Specifying GLB location in README * Mention D2R code release in progress
-
由 Toby Boyd 提交于
-
- 30 4月, 2019 5 次提交
-
-
由 Igor 提交于
Replace per_device with per_replica and PerDevice with PerReplica, because the PerDevice concept was renamed and doesn't exist anymore. (#6693) * Replace per_device with per_replica and PerDevice with PerReplica, because the PerReplica concept was renamed and doesn't exist anymore.
-
由 Toby Boyd 提交于
* Add accuracy check. * Avoid double flag init, move data_dir to real data. * Comment on lower accuracy target.
-
由 Shining Sun 提交于
* bug fix * bug fix
-
由 Igor 提交于
* Add benchmarks with the --cloning flag to Resnet and NFC. * Renamed cloning to clone_model_in_keras_dist_strat. Dropped a few tests that aren't essential. * Fixed up the formatting after re-naming the flag to a much longer name. Thanks, lint. * Fixed the lint error in nfc_common.py
-
由 Songyi Blair Han 提交于
-
- 27 4月, 2019 3 次提交
-
-
由 Toby Boyd 提交于
* Combined imagenet and cifar-10 benchmarks * Comments and epochs_between_evals. * Added tuned tests and cleaned up benchmark flags * Fix names. * Return results and add images/sec hook. * updated doc strings for return values. * 128 to 256 batch for FP16 test * added more doc strings to fix lint.
-
由 pkulzc 提交于
* Merged commit includes the following changes: 244869387 by Sergio Guadarrama: This CL adds script/code to generate Visual WakeWords Dataset annotation files and TF records starting from COCO dataset. -- 244866660 by Sergio Guadarrama: Add VisualWakeWords Dataset to Slim dataset_factory to train MobileNets on it. -- 244836000 by Sergio Guadarrama: n/a -- 244104396 by Sergio Guadarrama: Add an option whether to enable / disable image cropping in inception_preprocessing. -- 242040128 by Sergio Guadarrama: Internal change 241793677 by Sergio Guadarrama: Internal change 241073081 by Sergio Guadarrama: Internal change 240131189 by Sergio Guadarrama: Internal change PiperOrigin-RevId: 244869387 * Merged commit includes the following changes: 245431876 by Sergio Guadarrama: Internal cleanup -- PiperOrigin-RevId: 245431876 * Merged commit includes the following changes: 245454983 by Sergio Guadarrama: Internal Cleanup -- PiperOrigin-RevId: 245454983
-
由 Ayush Dubey 提交于
* Add num_packs flag for MirroredStrategy's cross device ops. * fix parens * Fix lint errors and make all_reduce_alg more robust. * Set default num_packs to 1
-
- 26 4月, 2019 3 次提交
-
-
由 Gaurav Jain 提交于
tf.test.is_gpu_available() should not be called in flags since this is called before app.main() and the runtime has not yet been initialized.
-
由 Ayush Dubey 提交于
* Remove contrib AllReduceCrossDeviceOps and update all_reduce_alg options with MirroredStrategy. * cleanup
-
由 Haoyu Zhang 提交于
Reason: test failures because contrib is not available in V2 This reverts commit 325dd761.
-
- 25 4月, 2019 6 次提交
-
-
由 Toby Boyd 提交于
-
由 Shining Sun 提交于
* Added none check for output_dir * Change double quote to single
-
由 André Araujo 提交于
* Internal change. PiperOrigin-RevId: 185564155 * Fix small bug when reading DELF features when file is empty. A new test is added that catches this bug. PiperOrigin-RevId: 213839503 * Refactors DELF example code to expose a function to create a DELF feature extractor. PiperOrigin-RevId: 241492615 * Merged commit includes the following changes: 244073180 by Andre Araujo: Internal change -- 243646498 by Andre Araujo: Detect and save bounding boxes for a list of images. A new proto is added, along with auxiliary read/write functions, and tests. -- PiperOrigin-RevId: 244073180 * Merged commit includes the following changes: 244872693 by Andre Araujo: Small change. -- 244871213 by Andre Araujo: Small edit to DELF extractor function. -- 244790715 by Andre Araujo: internal automated change -- 244773146 by Andre Araujo: Add option to visualize extracted boxes, and change slightly model input. -- 244288992 by Andre Araujo: Script to cluster DELF features using K-means. -- 244275164 by Andre Araujo: Scripts to extract DELF features and boxes from Revisited Oxford/Paris datasets, used in Detect-to-Retrieve paper. -- PiperOrigin-RevId: 244872693 * Merged commit includes the following changes: PiperOrigin-RevId: 244876167 * Updates instructions to include new released models, instructions for DELF and detector extraction, etc. * Small fix * Merged commit includes the following changes: 244909336 by Andre Araujo: Update to DELF config example to use latest model. -- PiperOrigin-RevId: 244909336 * Revert "Merged commit includes the following changes:" This reverts commit f41be9f159330f62846d5d8b3d3d5d55cd1874f4. * Update delf_config_example model * Small edits to detection instructions * Specifying GLB location in README
-
由 Haoyu Zhang 提交于
-
由 Abhinav Sagar 提交于
-
由 Yuefeng Zhou 提交于
-
- 24 4月, 2019 3 次提交
-
-
由 Haoyu Zhang 提交于
-
由 Haoyu Zhang 提交于
* Introduce a short sleep before ds.prefetch in tf.data. * Further limit dataset threads to reduce CPU contention * Tuned dataset sleep time * Rename dataset sleep flag; enable it only for Keras Graph mode
-
由 Yukun Zhu 提交于
* Update colab demo * fix typo
-