Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

TL2.0 #35

Merged
merged 41 commits into from May 25, 2019
Merged

TL2.0 #35

merged 41 commits into from May 25, 2019

Conversation

@ArnoldLIULJ
Copy link
Member

@ArnoldLIULJ ArnoldLIULJ commented May 20, 2019

Update model compatible with TensorLayer2.0.
Rewrite the loss. cross_entropy_seq_with_mask and cross_entropy_seq.
Need to run to see if it converges and produce desirable results

@ArnoldLIULJ ArnoldLIULJ changed the title Tf2.0 TL2.0 May 20, 2019
@ArnoldLIULJ
Copy link
Member Author

@ArnoldLIULJ ArnoldLIULJ commented May 22, 2019

Update the model with deeper structure
n_units->256, n_layers->3(for both encoding and decoding)

Lingjun Liu and others added 4 commits May 22, 2019
Lingjun Liu Lingjun Liu
Lingjun Liu Lingjun Liu
Lingjun Liu added 2 commits May 23, 2019
Lingjun Liu Lingjun Liu
Lingjun Liu Lingjun Liu
@ArnoldLIULJ
Copy link
Member Author

@ArnoldLIULJ ArnoldLIULJ commented May 23, 2019

FIX multilayer RNN bugs. states propagation.
ADD inference test

@ArnoldLIULJ
Copy link
Member Author

@ArnoldLIULJ ArnoldLIULJ commented May 23, 2019

PACK multiple layers into a list layers with self-defined layers
CHANGE batch_size to decoder_length
ADD top_n method at inference time

Lingjun Liu and others added 10 commits May 24, 2019
@zsdonghao zsdonghao merged commit b631ce8 into master May 25, 2019
1 check passed
1 check passed
WIP ready for review
Details
@zsdonghao zsdonghao deleted the tf2.0 branch May 25, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Linked issues

Successfully merging this pull request may close these issues.

None yet

3 participants
You can’t perform that action at this time.