提交 · 27fb855b027ead16d2616dcb59c67409a2176b7f · 曾经的那一瞬间 / Models

“33dca040b7c72c5bce52faa6c6e6594fd4b10cf9”上不存在“dataset/mot/image_lists/prw.train”

19 1月, 2022 2 次提交
- L
  Internal change · 27fb855b
  由 Le Hou 提交于 1月 18, 2022
```
PiperOrigin-RevId: 422638136
```
  27fb855b
- A
  Internal change · 871c4e0a
  由 A. Unique TensorFlower 提交于 1月 18, 2022
```
PiperOrigin-RevId: 422612138
```
  871c4e0a
09 12月, 2021 1 次提交

Replace dynamic Squeeze with reshape to 1d · ce7d8065

由 Smit Hinsu 提交于 12月 08, 2021

TPUs don't support dynamic Squeeze as it won't be able to identify the dimensions to remove if the input shape is dynamic. For example, [<=16, 1] could return a vector or scalar result depending on the size of the dynamic dim.

PiperOrigin-RevId: 415083667

ce7d8065

18 11月, 2021 1 次提交
- H
  Deprecate the graph rewrite path for fp16. This is no longer a TF2 api and there is no usage. · 51cb03b0
  由 Hongkun Yu 提交于 11月 17, 2021
```
PiperOrigin-RevId: 410629444
```
  51cb03b0
13 4月, 2021 1 次提交

Use nonexperimental mixed precision API for official models. · 0d8f9807

由 Reed Wanderman-Milne 提交于 4月 12, 2021

For all modified calls to set_mixed_precision_policy(), the loss_scale argument was removed, as it cannot be passed if the nonexperimental API is used. For all such callers, the loss_scale is later used to explicitly create a LossScaleOptimizer, so removing the argument has no impact.

Switching to the non-experimental LossScaleOptimizer has no effect, as it has near identical behavior and all isinstance checks within the official models check for the non-experimental version.

PiperOrigin-RevId: 368101975

0d8f9807

10 3月, 2021 1 次提交
- F
  Internal change · 5a898973
  由 Frederick Liu 提交于 3月 09, 2021
```
PiperOrigin-RevId: 361957289
```
  5a898973
15 12月, 2020 1 次提交
- C
  Internal change · 690e44ed
  由 Chen Chen 提交于 12月 14, 2020
```
PiperOrigin-RevId: 347505505
```
  690e44ed
16 11月, 2020 1 次提交
- H
  experimental_steps_per_execution -> steps_per_execution for TF 2.4 release. · a1b04a45
  由 Hongkun Yu 提交于 11月 15, 2020
```
PiperOrigin-RevId: 342566888
```
  a1b04a45
14 11月, 2020 1 次提交
- R
  Internal change · f409e4d0
  由 Reed Wanderman-Milne 提交于 11月 13, 2020
```
PiperOrigin-RevId: 342390260
```
  f409e4d0
27 10月, 2020 1 次提交
- H
  Py3 cleanup for nlp/ · cd5e1204
  由 Hongkun Yu 提交于 10月 26, 2020
```
PiperOrigin-RevId: 339071563
```
  cd5e1204
30 9月, 2020 1 次提交
- C
  Internal change · 8f345563
  由 Chenkai Kuang 提交于 9月 29, 2020
```
PiperOrigin-RevId: 334517779
```
  8f345563
13 9月, 2020 1 次提交
- H
  [Clean up] Consolidate distribution utils. · 33a4c207
  由 Hongkun Yu 提交于 9月 12, 2020
```
PiperOrigin-RevId: 331359058
```
  33a4c207
22 8月, 2020 2 次提交
- M
  Add a flag to control the number of train examples. · 1c89b792
  由 Maxim Neumann 提交于 8月 21, 2020
```
PiperOrigin-RevId: 327838493
```
  1c89b792
- M
  Add a flag to control the number of train examples. · 2454e6ea
  由 Maxim Neumann 提交于 8月 21, 2020
```
PiperOrigin-RevId: 327838493
```
  2454e6ea
15 8月, 2020 1 次提交
- K
  
  Skip softmax activation if num_labels <= 1 · fbf3470f
  由 Kevin 提交于 8月 14, 2020
  
  fbf3470f
13 8月, 2020 2 次提交
- H
  Internal change · 999fae62
  由 Hongkun Yu 提交于 8月 12, 2020
```
PiperOrigin-RevId: 326286926
```
  999fae62
- H
  Internal change · 88253ce5
  由 Hongkun Yu 提交于 8月 12, 2020
```
PiperOrigin-RevId: 326286926
```
  88253ce5
07 8月, 2020 1 次提交
- K
  
  Consider regression model in predict mode · 83f355bb
  由 Kevin 提交于 8月 07, 2020
  
  83f355bb
23 7月, 2020 2 次提交
- H
  Allow export only path takes the hub module. · 6a446b64
  由 Hongkun Yu 提交于 7月 22, 2020
```
PiperOrigin-RevId: 322594580
```
  6a446b64
- H
  Allow export only path takes the hub module. · acaef198
  由 Hongkun Yu 提交于 7月 22, 2020
```
PiperOrigin-RevId: 322594580
```
  acaef198
13 6月, 2020 1 次提交
- M
  Allow to provide custom metrics to run_classifier. · 40fdb50c
  由 Maxim Neumann 提交于 6月 12, 2020
```
PiperOrigin-RevId: 316134518
```
  40fdb50c
12 6月, 2020 1 次提交
- M
  Support using sample weights for finetuning regression. · 2dfd1e63
  由 Maxim Neumann 提交于 6月 12, 2020
```
Replace as well the regression loss with Keras loss object.

PiperOrigin-RevId: 316062800
```
  2dfd1e63
05 6月, 2020 1 次提交
- T
  Internal cleanup. · 9b219f04
  由 Tianqi Liu 提交于 6月 04, 2020
```
PiperOrigin-RevId: 314822016
```
  9b219f04
04 6月, 2020 1 次提交
- M
  Adjust run_classification to support fine-tuning regression tasks. · b2e422b0
  由 Maxim Neumann 提交于 6月 03, 2020
```
PiperOrigin-RevId: 314607393
```
  b2e422b0
29 5月, 2020 1 次提交

Deprecate old customized training loop for run_classifier.py as compile/fit... · abf60128

由 Hongkun Yu 提交于 5月 28, 2020

Deprecate old customized training loop for run_classifier.py as compile/fit fully satisfy needs/performance.

PiperOrigin-RevId: 313660745

abf60128

26 5月, 2020 1 次提交

Add --sub_model_export_name to run_classifier and run_squad. · 38800db3

由 André Susano Pinto 提交于 5月 26, 2020

This allows one to finetune a BERT model into a task before using it for
another task. E.g. SQuAD before finetune another QA type of tasks.

PiperOrigin-RevId: 313145768

38800db3

24 5月, 2020 1 次提交
- H
  Internal change · ca2e6ae0
  由 Hongkun Yu 提交于 5月 23, 2020
```
PiperOrigin-RevId: 312923051
```
  ca2e6ae0
20 5月, 2020 1 次提交
- H
  [Clean up] Remove enable_eager in the session config: Model garden is TF2 only now. · c2666cea
  由 Hongkun Yu 提交于 5月 19, 2020
```
Remove is_v2_0

PiperOrigin-RevId: 312336907
```
  c2666cea
19 5月, 2020 1 次提交
- A
  Internal change · 457064e1
  由 A. Unique TensorFlower 提交于 5月 18, 2020
```
PiperOrigin-RevId: 312096778
```
  457064e1
05 5月, 2020 1 次提交
- H
  Move gin flags to hyperparams_flags are they are in the same category and we... · d93a1206
  由 Hongkun Yu 提交于 5月 04, 2020
```
Move gin flags to hyperparams_flags are they are in the same category and we will use them more widely.

PiperOrigin-RevId: 309779408
```
  d93a1206
22 4月, 2020 1 次提交
- L
  Internal change · 04165099
  由 Le Hou 提交于 4月 21, 2020
```
PiperOrigin-RevId: 307712303
```
  04165099
21 4月, 2020 1 次提交
- A
  Internal change · e289c8c7
  由 A. Unique TensorFlower 提交于 4月 20, 2020
```
PiperOrigin-RevId: 307416709
```
  e289c8c7
17 4月, 2020 2 次提交
- C
  Internal change · 01c4ee81
  由 Chen Chen 提交于 4月 16, 2020
```
PiperOrigin-RevId: 306986791
```
  01c4ee81
- J
  Thread custom callbacks through BERT classifier. This allows injection of... · b523f160
  由 Jeremiah Harmsen 提交于 4月 16, 2020
```
Thread custom callbacks through BERT classifier.  This allows injection of custom callbacks for library users of run_classifier.

PiperOrigin-RevId: 306966758
```
  b523f160
16 4月, 2020 1 次提交
- L
  Internal change · 89599a23
  由 Le Hou 提交于 4月 15, 2020
```
PiperOrigin-RevId: 306664983
```
  89599a23
11 4月, 2020 1 次提交
- A
  Add end learning rate (end_lr) FLAG for linear learning rate decay. · 22de957a
  由 A. Unique TensorFlower 提交于 4月 10, 2020
```
PiperOrigin-RevId: 305948522
```
  22de957a
28 3月, 2020 1 次提交
- A
  Support Lamb optimizer within BERT. · da5860f2
  由 A. Unique TensorFlower 提交于 3月 27, 2020
```
PiperOrigin-RevId: 303356961
```
  da5860f2
26 3月, 2020 2 次提交
- H
  Move modeling_training_utils.py to bert folder · 6ef71d49
  由 Hongkun Yu 提交于 3月 25, 2020
```
PiperOrigin-RevId: 303050809
```
  6ef71d49
- H
  Internal change · 38486725
  由 Hongkun Yu 提交于 3月 25, 2020
```
PiperOrigin-RevId: 302977474
```
  38486725
18 3月, 2020 1 次提交
- H
  Adds scale_loss to run_customized_training_loop, which should be the correct treatment. · bf6d6f6f
  由 Hongkun Yu 提交于 3月 17, 2020
```
PiperOrigin-RevId: 301441181
```
  bf6d6f6f

曾经的那一瞬间 / Models 大约 1 年 前同步成功

曾经的那一瞬间 / Models
大约 1 年前同步成功