Skip to content

feat: Support training of RNN and LSTM.#1110

Merged
SanftMonster merged 19 commits intoSciSharp:masterfrom
Wanglongzhi2001:master
Jun 21, 2023
Merged

feat: Support training of RNN and LSTM.#1110
SanftMonster merged 19 commits intoSciSharp:masterfrom
Wanglongzhi2001:master

Conversation

@Wanglongzhi2001
Copy link
Contributor

@Wanglongzhi2001 Wanglongzhi2001 commented Jun 18, 2023 •
edited
Loading

  1. Support the model building of RNN.
  2. Support the model building of LSTM.
  3. Support the training of RNN.
  4. Support the training of LSTM.
  5. Add test.
  6. Refactor Tensors class to deal with implicit conversion from Tensors to Tensor and support Nest Tensors.
  7. Add Nest class to handle nested Tensor cases, and add corresponding Flatten and Mapstructure API.
  8. Use IOptionalArgs in Layer.Apply to prevent the inconsistency of Call method parameters in different layers, resulting in the need to set too many Call method parameters.
  9. Add the implemention of while_loop in graph mode.

SanftMonster and others added 18 commits June 6, 2023 11:12
@SanftMonster
feat: add RNN basic framework.
46e190d
@SanftMonster
feat: add Nestable and Nest<T>.
4939105
@SanftMonster
feat: support simple RNN.
537b3e1
@SanftMonster
Merge pull request #1097 from AsakusaRinne/rnn-dev
9da157f 
feat: add rnn basic modules
@SanftMonster
fix: some possible errors of RNN.
dcaa0f4
@SanftMonster
Merge pull request #1098 from AsakusaRinne/rnn-dev
81a9d23 
fix: some possible errors of RNN.
@Wanglongzhi2001
Add feature(not completed):add SimpleRNNCell, StackedRNNCell, RNN and…
db8e43b 
… test
@SanftMonster
Merge pull request #1100 from Wanglongzhi2001/rnn-dev
1d97b71 
Add feature(not completed):add SimpleRNNCell, StackedRNNCell, RNN and test.
@SanftMonster
feat: support model building with RNN.
f1fbcf2
@SanftMonster
fix: error when training SimpleRNN.
07ea656
@SanftMonster
feat: add exception catch to code generator.
5bfe098
@Wanglongzhi2001 @SanftMonster
Add new feature: add LSTMCell and test
df7d700
@SanftMonster
fix: error after merging LSTM support.
6b30902
@SanftMonster
feat: update some gen_ops.
0114885
@SanftMonster
fix: none gradient error when training LSTM.
675b93a
@SanftMonster
fix: training LSTM does not align with tensorflow.
a0df810
@SanftMonster
Merge pull request #1109 from AsakusaRinne/rnn-dev
edbf89b 
feat: support training of RNN.
@Wanglongzhi2001
feat: support training of RNN
cb31cf4
@Wanglongzhi2001
Copy link
Contributor Author

Wanglongzhi2001 commented Jun 18, 2023

It builds on windows successfully, but fails to build on linux. According to the error message, it seems that the problem is on protobuf. Is this matters? @AsakusaRinne

@SanftMonster
Copy link
Collaborator

SanftMonster commented Jun 18, 2023

It builds on windows successfully, but fails to build on linux. According to the error message, it seems that the problem is on protobuf. Is this matters? @AsakusaRinne

I'll test it on linux later

@SanftMonster SanftMonster added enhancement New feature or request api New API or change API keras auto-release and removed auto-release labels Jun 18, 2023
@SanftMonster
Copy link
Collaborator

SanftMonster commented Jun 21, 2023

The problem in ci has been located, which will be solved after upgrading the redist packages later.

@SanftMonster SanftMonster merged commit 0454c7b into SciSharp:master Jun 21, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@Oceania2018 Oceania2018 Awaiting requested review from Oceania2018 Oceania2018 is a code owner

Assignees

No one assigned

Labels

api New API or change API enhancement New feature or request keras

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@Wanglongzhi2001 @SanftMonster