- 19 5月, 2018 1 次提交
-
-
由 Younghee Kwon 提交于
* Add boosted_trees to the official models * Comments addressed from review, and a test added; using absl.flags instead of argparser. * Used help_wrap. Also added instructions for inference.
-
- 18 5月, 2018 4 次提交
-
-
由 Mark Daoust 提交于
[samples/java]: Explain where tensor names came from.
-
由 Asim Shankar 提交于
Fix #4013
-
由 Taylor Robie 提交于
* deprecate research/resnet * also add a tf warning in the code
-
-
- 17 5月, 2018 1 次提交
-
-
由 Qianli Scott Zhu 提交于
* Fix resnet missing layers. The official v1 model contains BN and Relu between input layer and pooling. * Remove the BN and Relu for V2. After some discussion with team and refer to existing implementation, those two layer seems to be only useful in V1. In V2, the first unit of the block will have a projection, that apply the BN and Relu for the shortcut. Adding a comment to make this clear. * Expand the comment section. * Remove the pre-trained checkpoint since its broken right now. Will restore it once we have new checkpoint generated.
-
- 16 5月, 2018 2 次提交
-
-
由 Katherine Wu 提交于
-
由 Haiyang Kong 提交于
* Make codes more pythonic. * Restore the indents Restore the indents.
-
- 15 5月, 2018 3 次提交
-
-
由 pkulzc 提交于
Fixed Bug regarding tfrecord shuffling in object_detection
-
由 Mark Daoust 提交于
Add lamberta to CODEOWNERS/samples
-
由 Mark Daoust 提交于
Add @lamberta to CODEOWNERS/samples
-
- 12 5月, 2018 7 次提交
-
-
由 pkulzc 提交于
Object detection export dir
-
由 Qianli Scott Zhu 提交于
* Update the wide_deep code for latest benchmark config. * Also update the transformer benchmark code.
-
由 Jonathan Huang 提交于
Release ssdlite mobilenet v2 coco trained model, add quantized training and minor fixes.
-
由 Qianli Scott Zhu 提交于
* Move the benchmark_uploader to new location. * Update benchmark logger to streaming upload. * Fix lint and unit test error. * delint. * Update the benchmark uploader test. Skip the import of benchmark_uploader when bigquery is not installed. * Merge the 2 classes of benchmark uploader into 1. * Address review comments. * delint. * Execute bigquery upload in a separate thread. * Change to use python six.moves for importing. * Address review comments and delint. * Address review comment. Adding comment for potential performance impact for model on CPU. * Fix random failure on py3. * Fix the order of flag saver to avoid the randomness. The test is broken when the benchmark_logger_type is set first, and validated when the benchmark_log_dir is not set yet.
-
由 Mark Omernick 提交于
Export @195097388.
-
由 Katherine Wu 提交于
-
由 Terry Koo 提交于
-
- 11 5月, 2018 3 次提交
-
-
由 Amit Patankar 提交于
Update mobilenet_v1.md
-
由 Zhichao Lu 提交于
196161788 by Zhichao Lu: Add eval_on_train_steps parameter. Since the number of samples in train dataset is usually different to the number of samples in the eval dataset. -- 196151742 by Zhichao Lu: Add an optional random sampling process for SSD meta arch and update mean stddev coder to use default std dev when corresponding tensor is not added to boxlist field. -- 196148940 by Zhichao Lu: Release ssdlite mobilenet v2 coco trained model. -- 196058528 by Zhichao Lu: Apply FPN feature map generation before we add additional layers on top of resnet feature extractor. -- 195818367 by Zhichao Lu: Add support for exporting detection keypoints. -- 195745420 by Zhichao Lu: Introduce include_metrics_per_category option to Object Detection eval_config. -- 195734733 by Zhichao Lu: Rename SSDLite config to be more explicit. -- 195717383 by Zhichao Lu: Add quantized training to object_detection. -- 195683542 by Zhichao Lu: Fix documentation for the interaction of fine_tune_checkpoint_type and load_all_detection_checkpoint_vars interaction. -- 195668233 by Zhichao Lu: Using batch size from params dictionary if present. -- 195570173 by Zhichao Lu: A few fixes to get new estimator API eval to match legacy detection eval binary by (1) plumbing `is_crowd` annotations through to COCO evaluator, (2) setting the `sloppy` flag in tf.contrib.data.parallel_interleave based on whether shuffling is enabled, and (3) saving the original image instead of the resized original image, which allows for small/medium/large mAP metrics to be properly computed. -- 195316756 by Zhichao Lu: Internal change -- PiperOrigin-RevId: 196161788
-
由 Qianli Scott Zhu 提交于
The final BN and ReLU layer is only need for v2 model since it was doing preactivation in each block.
-
- 10 5月, 2018 3 次提交
-
-
由 Ilya Mironov 提交于
Addressing reviewers' comments. Added correct handling of order=np.inf.
-
由 Ilya Mironov 提交于
-
由 Qianli Scott Zhu 提交于
* Add benchmark logging for wide_deep. * Fix lint error.
-
- 09 5月, 2018 3 次提交
-
-
由 Tobias Hinz 提交于
Update mnist_eager.py to use GPU if available and not specified otherwise. At the moment it uses the CPU even if a GPU is available, since "if flags_obj.no_gpu or tf.test.is_gpu_available():" in "def run_mnist_eager(flags_obj):" evaluates to True if a GPU is available.
-
由 Taylor Robie 提交于
* forbid resnet v1 fp16 * address PR comments
-
由 Ilya Mironov 提交于
Updates to differential_privacy/pate following the Banff workshop.
-
- 08 5月, 2018 8 次提交
-
-
由 Vlad Paunescu 提交于
-
由 Ilya Mironov 提交于
-
由 Ilya Mironov 提交于
-
由 Ilya Mironov 提交于
-
由 Ilya Mironov 提交于
-
-
由 Ilya Mironov 提交于
-
由 Ilya Mironov 提交于
-
- 07 5月, 2018 2 次提交
-
-
由 Mark Daoust 提交于
Collapse license cell
-
由 Mark Daoust 提交于
Fix typo in iris_data.py
-
- 06 5月, 2018 1 次提交
-
-
由 BMDan 提交于
-
- 05 5月, 2018 2 次提交
-
-
由 Billy Lamberta 提交于
-
由 Billy Lamberta 提交于
-