- 02 5月, 2018 3 次提交
-
-
由 Asim Shankar 提交于
[samples/core/get_starter/eager]: Update with API simplifications in 1.8
-
由 Asim Shankar 提交于
-
由 Frank Chen 提交于
* Add new import paths to mnist_tpu.py This adds new import paths to mnist_tpu.py so individuals can check out the git repository and run `python mnist_tpu.py` without finagling with python import search paths. * Fix lint errors * Add disable= to pylint instructions * Remove parent path * Update comments
-
- 12 4月, 2018 1 次提交
-
-
由 Neal Wu 提交于
Pull doc fixes into version branch.
-
- 11 4月, 2018 3 次提交
-
-
由 Mark Daoust 提交于
-
由 Akshay Agrawal 提交于
-
由 Akshay Agrawal 提交于
* "Machine Learning" --> Machine learning. * "steepest the ascent" --> "steepest ascent" * "standard gradient descent" --> "stochastic gradient descent"
-
- 04 4月, 2018 5 次提交
-
-
由 Anna R 提交于
-
由 pkulzc 提交于
Internal changes for object detection
-
由 pkulzc 提交于
-
由 Qianli Scott Zhu 提交于
-
由 Lukasz Kaiser 提交于
First commit of learning_unsupervised_learning
-
- 03 4月, 2018 28 次提交
-
-
由 Nick Johnston 提交于
Image encoder: fix binary files parsed as UTF-8
-
由 pkulzc 提交于
-
由 Luke Metz 提交于
-
由 dlibenzi 提交于
* Fix MNIST to allow master=local to work again. * Update mnist_tpu.py
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191377056
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191369912
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191358138
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191342074
-
由 Zhichao Lu 提交于
- Fix externally broken test input_test.py - Add Cython to object detection REQUIRED_PACKAGES PiperOrigin-RevId: 191338837
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191248237
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191112757
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191014662
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190996725
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190972737
-
由 Zhichao Lu 提交于
Updating model to infer when labels are padded, rather than relying solely on the mode. This is necessary for evaluating on train data. PiperOrigin-RevId: 190960173
-
由 Zhichao Lu 提交于
Use keyword arguments in model builder for ssd feature extractor. This was incorreclty setting reuse argument. PiperOrigin-RevId: 190940462
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190790653
-
由 Zhichao Lu 提交于
Removing flag defaults for train steps and eval steps. If not provided, will revert to the values specified in the config. PiperOrigin-RevId: 190706470
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190705850
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190688309
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190682119
-
由 Zhichao Lu 提交于
Add warning if shuffle=false but num_readers > 1. The latter would cause the input data stream to still be shuffled. PiperOrigin-RevId: 190678792
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190651873
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190636417
-
由 Zhichao Lu 提交于
Cython is must installed package for building cocoapi. else the build will not initiate. PiperOrigin-RevId: 190553060
-
由 Zhichao Lu 提交于
Tfmini requires positive values for the dimensions. Therefore, we need to test for no paddings in both width and height direction. PiperOrigin-RevId: 190520529
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 190505306
-
由 Zhichao Lu 提交于
Switch line orders in trainer so that restore_map is called after moving average variables are created. Moving averages are now properly loaded during fine-tuning, instead of being recreated. PiperOrigin-RevId: 190496046
-