Finetune: AWAC

configs/finetune/awac/antmaze/large_diverse_v2.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: antmaze-large-diverse-v2
+eval_frequency: 50000
+gamma: 0.99
+group: awac-antmaze-large-diverse-v2-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 100
+normalize_reward: true
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/antmaze/large_play_v2.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: antmaze-large-play-v2
+eval_frequency: 50000
+gamma: 0.99
+group: awac-antmaze-large-play-v2-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 100
+normalize_reward: true
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/antmaze/medium_diverse_v2.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: antmaze-medium-diverse-v2
+eval_frequency: 50000
+gamma: 0.99
+group: awac-antmaze-medium-diverse-v2-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 100
+normalize_reward: true
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/antmaze/medium_play_v2.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: antmaze-medium-play-v2
+eval_frequency: 50000
+gamma: 0.99
+group: awac-antmaze-medium-play-v2-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 100
+normalize_reward: true
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/antmaze/umaze_diverse_v2.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: antmaze-umaze-diverse-v2
+eval_frequency: 50000
+gamma: 0.99
+group: awac-antmaze-umaze-diverse-v2-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 100
+normalize_reward: true
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/antmaze/umaze_v2.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: antmaze-umaze-v2
+eval_frequency: 50000
+gamma: 0.99
+group: awac-antmaze-umaze-v2-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 100
+normalize_reward: true
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/door/cloned_v1.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: door-cloned-v1
+eval_frequency: 5000
+gamma: 0.99
+group: awac-door-cloned-v1-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 10
+normalize_reward: false
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/hammer/cloned_v1.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: hammer-cloned-v1
+eval_frequency: 5000
+gamma: 0.99
+group: awac-hammer-cloned-v1-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 10
+normalize_reward: false
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/pen/cloned_v1.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: pen-cloned-v1
+eval_frequency: 5000
+gamma: 0.99
+group: awac-pen-cloned-v1-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 10
+normalize_reward: false
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69

configs/finetune/awac/relocate/cloned_v1.yaml

@@ -0,0 +1,20 @@
+awac_lambda: 0.1
+batch_size: 256
+buffer_size: 10000000
+checkpoints_path: null
+deterministic_torch: false
+device: cuda
+env_name: relocate-cloned-v1
+eval_frequency: 5000
+gamma: 0.99
+group: awac-relocate-cloned-v1-multiseed-v0
+hidden_dim: 256
+learning_rate: 0.0003
+n_test_episodes: 10
+normalize_reward: false
+offline_iterations: 1000000
+online_iterations: 1000000
+project: CORL
+seed: 42
+tau: 0.005
+test_seed: 69