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.

Sign up for GitHub

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

Jump to bottom

fix(pu): polish all mlp model and related configs #26

Merged
merged 16 commits into from
May 7, 2023
Merged

fix(pu): polish all mlp model and related configs #26

merged 16 commits into from
May 7, 2023

Conversation

puyuan1996
Copy link
Collaborator

@puyuan1996 puyuan1996 commented May 7, 2023 •
edited
Loading

  • fix the norm, activation, and initialization in the mlp model
  • add the option res_connection_in_dynamics. If it's True, we add skip connections of the latent state when learning next_latent_state in the dynamics network which can make the learning of next_latent_state more stable and smooth.

image

@puyuan1996
feature(pu): add discrete_action_encoding_type option
bdbb857
@puyuan1996
Merge branch 'main' of https://github.com/opendilab/LightZero into de…
fae8fdf 
…v-mlp
@puyuan1996
fix(pu): don't use last_linear_layer_init_zero in the middle layers i…
ecffa3f 
…n mlp model
@puyuan1996
fix(pu): add activation in the final layers of common fcn and represe…
95aa1e2 
…ntation network
@puyuan1996
feature(pu): add norm type option in ez mlp model
0203da7
@puyuan1996
Merge branch 'main' of https://github.com/opendilab/LightZero into de…
e5f2bdd 
…v-mlp
@puyuan1996
polish(pu): polish mz mlp model
03c43c4
@puyuan1996
debug(pu): use the same common and muzero_model_mlp with branch main-…
b5b8fa9 
…before-release
@puyuan1996
Merge branch 'main' of https://github.com/opendilab/LightZero into de…
b252da9 
…v-mlp
@puyuan1996
polish(pu): set output_norm=True for all MLP model
b72420b
@puyuan1996
polish(pu): polish output_norm and output_activation
b42e0c7
@puyuan1996
polish(pu): add res_connection_in_dynamics option in muzero_mlp_model
c020800
@puyuan1996
fix(pu): fix output_norm, output_act and init_zero in mlp model
b2ba385
@puyuan1996
polish(pu): rename hidden_state to latent_state in policy, polish mlp…
ac3dc2f 
… config
@puyuan1996 puyuan1996 added bug Something isn't working enhancement New feature or request labels May 7, 2023
@puyuan1996 puyuan1996 merged commit 1d2b8f2 into main May 7, 2023
@puyuan1996 puyuan1996 deleted the dev-mlp branch May 7, 2023 06:42
@puyuan1996 puyuan1996 restored the dev-mlp branch May 7, 2023 06:42
@puyuan1996 puyuan1996 deleted the dev-mlp branch May 7, 2023 06:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
bug Something isn't working enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
1 participant
@puyuan1996