- 06 Aug, 2020 1 commit
-
-
Hongkun Yu authored
Move mock_task to utils/testing/ PiperOrigin-RevId: 325275356
-
- 05 Aug, 2020 4 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 325115012
-
Allen Wang authored
PiperOrigin-RevId: 325093611
-
Dan Holtmann-Rice authored
PiperOrigin-RevId: 325088513
-
Hongkun Yu authored
PiperOrigin-RevId: 325073831
-
- 04 Aug, 2020 1 commit
-
-
A. Unique TensorFlower authored
For object detection models, removing rpn_head.anchors_per_location from config and using anchor.num_scales * len(anchor.aspect_ratios) to compute anchors_per_location PiperOrigin-RevId: 324878153
-
- 03 Aug, 2020 10 commits
-
-
Chen Chen authored
tensorflow_models' colabs. PiperOrigin-RevId: 324704216
-
xinliupitt authored
-
Chen Chen authored
PiperOrigin-RevId: 324703765
-
Chen Chen authored
PiperOrigin-RevId: 324700979
-
Hongkun Yu authored
PiperOrigin-RevId: 324693165
-
Hongkun Yu authored
PiperOrigin-RevId: 324659665
-
Chen Chen authored
Change the version of tf models nightly pip to 2.3.0 and the required tensorflow package for non-nightly pip to tensorflow>=2.3.0. PiperOrigin-RevId: 324656938
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 324656070
-
Chen Chen authored
PiperOrigin-RevId: 324630399
-
Francois Chollet authored
PiperOrigin-RevId: 324625967
-
- 01 Aug, 2020 2 commits
-
-
xinliupitt authored
-
xinliupitt authored
-
- 31 Jul, 2020 5 commits
-
-
xinliupitt authored
-
Hongkun Yu authored
PiperOrigin-RevId: 324258617
-
Srihari Humbarwadi authored
-
Abdullah Rashwan authored
PiperOrigin-RevId: 324140487
-
Hongkun Yu authored
PiperOrigin-RevId: 324137297
-
- 30 Jul, 2020 7 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 324084377
-
xinliupitt authored
-
xinliupitt authored
-
xinliupitt authored
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 323948101
-
Tomer Kaftan authored
Pre-emptively disable the KerasTensors refactoring for the detection models in tensorflow models/official/vision/detection, because they rely on several unsupported things that will stop working entirely when the refactoring goes live. Specifically: * The custom layers implement `__call__` instead of `call` and rely on manually enter the keras backend graph * The vision models try to use `tf.while_loop` as Keras op layers during functional API construction, which is unsupported. Updating the models to avoid this would subtly change the variable names and break the pre-existing tf1-style name-based checkpoints, so for now we will just disable the KerasTensors refactoring for these models. PiperOrigin-RevId: 323937426
-
Hongkun Yu authored
PiperOrigin-RevId: 323927843
-
- 29 Jul, 2020 7 commits
-
-
xinliupitt authored
-
xinliupitt authored
-
xinliupitt authored
-
André Susano Pinto authored
PiperOrigin-RevId: 323827690
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 323732686
-
Hongkun Yu authored
PiperOrigin-RevId: 323711016
-
Hongkun Yu authored
PiperOrigin-RevId: 323685656
-
- 28 Jul, 2020 3 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 323653596
-
Hongkun Yu authored
PiperOrigin-RevId: 323644473
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 323627192
-