dodox/dppo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

squash commits

Browse Source
This commit is contained in:
allenzren
2024-09-11 21:09:17 -04:00
parent 8ce0aa1485
commit 2ddf63b8f5
200 changed files with 1240 additions and 1186 deletions
Download Patch File Download Diff File Expand all files Collapse all files
cfg/gym/finetune/halfcheetah-v2/ft_awr_diffusion_mlp.yaml
+2 -3
View File
@@ -83,20 +83,19 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
residual_style: True
critic:
_target_: model.common.critic.CriticObs
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
transition_dim: ${transition_dim}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_dipo_diffusion_mlp.yaml
+2 -3
View File
@@ -82,7 +82,7 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
@@ -91,13 +91,12 @@ model:
_target_: model.common.critic.CriticObsAct
action_dim: ${action_dim}
action_steps: ${act_steps}
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
transition_dim: ${transition_dim}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_dql_diffusion_mlp.yaml
+2 -3
View File
@@ -81,7 +81,7 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
@@ -90,13 +90,12 @@ model:
_target_: model.common.critic.CriticObsAct
action_dim: ${action_dim}
action_steps: ${act_steps}
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
transition_dim: ${transition_dim}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_idql_diffusion_mlp.yaml
+3 -4
View File
@@ -84,7 +84,7 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
@@ -93,19 +93,18 @@ model:
_target_: model.common.critic.CriticObsAct
action_dim: ${action_dim}
action_steps: ${act_steps}
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
critic_v:
_target_: model.common.critic.CriticObs
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
transition_dim: ${transition_dim}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_ppo_diffusion_mlp.yaml
+2 -4
View File
@@ -91,20 +91,18 @@ model:
mlp_dims: [512, 512, 512]
activation_type: ReLU
residual_style: True
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
critic:
_target_: model.common.critic.CriticObs
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
ft_denoising_steps: ${ft_denoising_steps}
transition_dim: ${transition_dim}
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
cond_steps: ${cond_steps}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_ppo_exact_diffusion_mlp.yaml
+2 -4
View File
@@ -99,20 +99,18 @@ model:
mlp_dims: [512, 512, 512]
activation_type: ReLU
residual_style: True
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
critic:
_target_: model.common.critic.CriticObs
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
ft_denoising_steps: ${ft_denoising_steps}
transition_dim: ${transition_dim}
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
cond_steps: ${cond_steps}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_qsm_diffusion_mlp.yaml
+2 -3
View File
@@ -82,7 +82,7 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
@@ -91,13 +91,12 @@ model:
_target_: model.common.critic.CriticObsAct
action_dim: ${action_dim}
action_steps: ${act_steps}
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
transition_dim: ${transition_dim}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ft_rwr_diffusion_mlp.yaml
+1 -2
View File
@@ -74,7 +74,7 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
@@ -82,6 +82,5 @@ model:
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
transition_dim: ${transition_dim}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ppo_diffusion_mlp.yaml
+2 -4
View File
@@ -87,22 +87,20 @@ model:
_target_: model.diffusion.mlp_diffusion.DiffusionMLP
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
time_dim: 16
mlp_dims: [512, 512, 512]
activation_type: ReLU
residual_style: True
critic:
_target_: model.common.critic.CriticObs
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
ft_denoising_steps: ${ft_denoising_steps}
transition_dim: ${transition_dim}
horizon_steps: ${horizon_steps}
obs_dim: ${obs_dim}
action_dim: ${action_dim}
cond_steps: ${cond_steps}
denoising_steps: ${denoising_steps}
device: ${device}
cfg/gym/finetune/halfcheetah-v2/ppo_gaussian_mlp.yaml
+2 -3
View File
@@ -80,15 +80,14 @@ model:
mlp_dims: [512, 512, 512]
activation_type: ReLU
residual_style: True
cond_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
horizon_steps: ${horizon_steps}
transition_dim: ${transition_dim}
critic:
_target_: model.common.critic.CriticObs
obs_dim: ${obs_dim}
cond_dim: ${eval:'${obs_dim} * ${cond_steps}'}
mlp_dims: [256, 256, 256]
activation_type: Mish
residual_style: True
horizon_steps: ${horizon_steps}
cond_steps: ${cond_steps}
device: ${device}