Commits · 589f0269608f2199c4881e4a29d6ed494229b674 · ModelZoo / ResNet50_tensorflow

23 Oct, 2020 1 commit

Shift all attribute assignments in functional subclass models below the... · 589f0269

A. Unique TensorFlower authored Oct 23, 2020

Shift all attribute assignments in functional subclass models below the super().__init__ call, and enable setattr tracking on all functional subclasses.

PiperOrigin-RevId: 338759531

589f0269

22 Oct, 2020 1 commit
- Internal change · 414b7b74
  Poorva Potdar authored Oct 22, 2020
```
PiperOrigin-RevId: 338521847
```
  414b7b74
21 Oct, 2020 1 commit
- Change Seq2SeqTransformer inputs to dictionary. · f3641f23
  Hongkun Yu authored Oct 21, 2020
```
PiperOrigin-RevId: 338309524
```
  f3641f23
16 Oct, 2020 1 commit
- Log transformer_scaffold configs. · 992a5bfa
  Le Hou authored Oct 16, 2020
```
PiperOrigin-RevId: 337540237
```
  992a5bfa
14 Oct, 2020 1 commit
- Implement XLNet QA model. · 7a6a8741
  Allen Wang authored Oct 14, 2020
```
PiperOrigin-RevId: 337134894
```
  7a6a8741
13 Oct, 2020 1 commit
- Slice Transformer outputs with a TF Op layer instead of a Lambda layer. · cd39fd58
  A. Unique TensorFlower authored Oct 13, 2020
```
Lambda layers are fundamentally non-portable since they serialize
Python bytecode.

PiperOrigin-RevId: 336917172
```
  cd39fd58
12 Oct, 2020 1 commit
- Implement SpanLabeler for XLNet. · fcb43c38
  Allen Wang authored Oct 12, 2020
```
PiperOrigin-RevId: 336709640
```
  fcb43c38
07 Oct, 2020 1 commit
- Minor bug fixes · d4557be2
  Le Hou authored Oct 07, 2020
```
PiperOrigin-RevId: 335944731
```
  d4557be2
06 Oct, 2020 1 commit

Clean up internal seq2seqTransformer usage. · 662a40be

Hongkun Yu authored Oct 05, 2020

Add it to export. TODO: move the encoder and decoder layer out to layers/ or decoders/.

PiperOrigin-RevId: 335560860

662a40be

05 Oct, 2020 1 commit

PR #42625: Add training call argument for MultiHeadAttention · 229f1dd5

Tzu-Wei Sung authored Oct 05, 2020

Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/42625

Miss `training` call argument for dropout layer. /cc @tanzhenyu for visibility.
Copybara import of the project:

--
ba2198fc735a2deca08c68a3c6266d02e01dfe3b by Tzu-Wei Sung <windqaq@gmail.com>:

Add training call argument for MultiHeadAttention

Remove trailing space

PiperOrigin-RevId: 335507127

229f1dd5

02 Oct, 2020 1 commit
- Implement XLNet Classifier model. · 44a5367a
  Allen Wang authored Oct 02, 2020
```
PiperOrigin-RevId: 335069348
```
  44a5367a
25 Sep, 2020 2 commits
- Internal change · 2a5c349d
  Chen Chen authored Sep 25, 2020
```
PiperOrigin-RevId: 333784610
```
  2a5c349d
- Internal change · 07a07f6a
  Chen Chen authored Sep 25, 2020
```
PiperOrigin-RevId: 333772508
```
  07a07f6a
24 Sep, 2020 1 commit
- Add norm_first to Transformer Scaffold; Add an option in gated_feedword to... · eaff981c
  A. Unique TensorFlower authored Sep 24, 2020
```
Add norm_first to Transformer Scaffold; Add an option in gated_feedword to disable the output layer_norm.

PiperOrigin-RevId: 333591020
```
  eaff981c
22 Sep, 2020 3 commits
- Implement XLNetBase. This is the underlying network shared between XLNet models. · 0864d66e
  Allen Wang authored Sep 22, 2020
```
PiperOrigin-RevId: 333183470
```
  0864d66e
- Internal Change · 4c226604
  Chen Chen authored Sep 22, 2020
```
PiperOrigin-RevId: 333163906
```
  4c226604
- Internal change · c3d53565
  Hongkun Yu authored Sep 21, 2020
```
PiperOrigin-RevId: 332992539
```
  c3d53565
21 Sep, 2020 4 commits
- Internal change · 5b9feb6d
  A. Unique TensorFlower authored Sep 21, 2020
```
PiperOrigin-RevId: 332953126
```
  5b9feb6d
- Internal change · 976ba237
  Hongkun Yu authored Sep 21, 2020
```
PiperOrigin-RevId: 332891157
```
  976ba237
- Internal change · 52b16a1a
  Chen Chen authored Sep 21, 2020
```
PiperOrigin-RevId: 332806032
```
  52b16a1a
- Clean keras_nlp.BertEncoder: · 71a2fc91
  Chen Chen authored Sep 21, 2020
```
Remove return_all_encoder_outputs argument from keras_nlp.BertEncoder as it is unused and we will always return all encoder outputs in the returned dict.

PiperOrigin-RevId: 332802761
```
  71a2fc91
20 Sep, 2020 1 commit

Migrate all TF-NLP tasks to consume models with dictionary outputs and... · 3ce3a5c5

Hongkun Yu authored Sep 20, 2020

Migrate all TF-NLP tasks to consume models with dictionary outputs and encoders with dictionary outputs.

PiperOrigin-RevId: 332740616

3ce3a5c5

18 Sep, 2020 5 commits
- Use self.get_temp_dir and simplify the test · 674506c4
  Hongkun Yu authored Sep 18, 2020
```
PiperOrigin-RevId: 332512380
```
  674506c4
- Add MaskedLM for keras_nlp · 96ca6e05
  Zhenyu Tan authored Sep 18, 2020
```
PiperOrigin-RevId: 332506544
```
  96ca6e05
- Implement Transformer XL · 2c98b4b0
  Allen Wang authored Sep 18, 2020
```
PiperOrigin-RevId: 332486600
```
  2c98b4b0
- Internal change · c7e31961
  A. Unique TensorFlower authored Sep 18, 2020
```
PiperOrigin-RevId: 332476566
```
  c7e31961
- Alias nlp/modeling BertEncoder with keras_nlp BertEncoder. · ba206271
  Zhenyu Tan authored Sep 17, 2020
```
PiperOrigin-RevId: 332386511
```
  ba206271
16 Sep, 2020 2 commits
- Internal change · e6ffa057
  Hongkun Yu authored Sep 16, 2020
```
PiperOrigin-RevId: 332092620
```
  e6ffa057
- Implement `TransformerXLBlock.` · 81fa82ef
  Allen Wang authored Sep 16, 2020
```
PiperOrigin-RevId: 332055561
```
  81fa82ef
15 Sep, 2020 1 commit
- Implement `TwoStreamRelativeAttention`. · 3325205b
  Allen Wang authored Sep 15, 2020
```
PiperOrigin-RevId: 331787357
```
  3325205b
14 Sep, 2020 2 commits
- Move `MultiHeadRelativeAttention` to relative_attention.py. · 2e41d8ca
  Allen Wang authored Sep 14, 2020
```
PiperOrigin-RevId: 331652519
```
  2e41d8ca
- Replace keras_nlp.module with keras_nlp.layers.module · fbfa69c8
  Zhenyu Tan authored Sep 14, 2020
```
PiperOrigin-RevId: 331620370
```
  fbfa69c8
11 Sep, 2020 1 commit
- aliasing OnDeviceEmbedding inside tensorflow_models. · 1a273519
  Zhenyu Tan authored Sep 11, 2020
```
PiperOrigin-RevId: 331173006
```
  1a273519
10 Sep, 2020 1 commit
- Internal change · 9b8b13e8
  Hongkun Yu authored Sep 10, 2020
```
PiperOrigin-RevId: 330982547
```
  9b8b13e8
09 Sep, 2020 6 commits
- Add dict outputs to BertEncoder and EncoderScaffold · f1b9d43f
  Hongkun Yu authored Sep 09, 2020
```
PiperOrigin-RevId: 330817736
```
  f1b9d43f
- BertEncoder to ingest OnDeviceEmbedding from keras_nlp. · fcff6f65
  Zhenyu Tan authored Sep 09, 2020
```
PiperOrigin-RevId: 330767398
```
  fcff6f65
- Move OnDeviceEmbedding to keras_nlp. · 9392652e
  Zhenyu Tan authored Sep 09, 2020
```
PiperOrigin-RevId: 330754739
```
  9392652e
- MultiHeadRelativeAttention compatibility changes with XLNet · b7f6e079
  Allen Wang authored Sep 09, 2020
```
PiperOrigin-RevId: 330751568
```
  b7f6e079
- Have BertEncoder ingest keras_nlp.SelfAttentionMask · 7ae450a1
  Zhenyu Tan authored Sep 08, 2020
```
PiperOrigin-RevId: 330623395
```
  7ae450a1
- BertEncoder to ingest TransformerEncoderBlock · 2f737e1e
  Zhenyu Tan authored Sep 08, 2020
```
PiperOrigin-RevId: 330622339
```
  2f737e1e