added trained policy files

osudrl · Aug 21, 2020 · b0992fb · b0992fb
1 parent d9d414f
commit b0992fb
Show file tree

Hide file tree

Showing 58 changed files with 304 additions and 0 deletions.
diff --git a/trained_models/aslip_unified_0_v6.pt b/trained_models/aslip_unified_0_v6.pt
diff --git a/trained_models/aslip_unified_0_v7.pt b/trained_models/aslip_unified_0_v7.pt
diff --git a/trained_models/aslip_unified_task0_v6.pt b/trained_models/aslip_unified_task0_v6.pt
diff --git a/trained_models/aslip_unified_task0_v7.pt b/trained_models/aslip_unified_task0_v7.pt
diff --git a/trained_models/comparison/aslip_delta_policies/avg_GRFs_speed0.0.npy b/trained_models/comparison/aslip_delta_policies/avg_GRFs_speed0.0.npy
diff --git a/trained_models/comparison/aslip_delta_policies/avg_GRFs_speed1.0.npy b/trained_models/comparison/aslip_delta_policies/avg_GRFs_speed1.0.npy
diff --git a/trained_models/comparison/aslip_delta_policies/avg_GRFs_speed2.0.npy b/trained_models/comparison/aslip_delta_policies/avg_GRFs_speed2.0.npy
diff --git a/...ed_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/actor.pt b/...ed_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/actor.pt
diff --git a/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/critic.pt b/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/critic.pt
diff --git a/...ies/traj-aslip_aslip_old_2048_12288_seed-0/events.out.tfevents.1590545745.vsl065.107110.0 b/...ies/traj-aslip_aslip_old_2048_12288_seed-0/events.out.tfevents.1590545745.vsl065.107110.0
diff --git a/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/experiment.info b/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: False
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 40
+num_steps: 307
+previous: None
+recurrent: False
+redis_address: None
+reward: aslip_old
+state_est: True
+traj: aslip
+use_gae: True
+viz_port: 8097
diff --git a/...els/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/experiment.pkl b/...els/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-0/experiment.pkl
diff --git a/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/actor.pt b/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/actor.pt
diff --git a/..._models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/critic.pt b/..._models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/critic.pt
diff --git a/...es/traj-aslip_aslip_old_2048_12288_seed-10/events.out.tfevents.1589497244.vsl074.160017.0 b/...es/traj-aslip_aslip_old_2048_12288_seed-10/events.out.tfevents.1589497244.vsl074.160017.0
diff --git a/...es/traj-aslip_aslip_old_2048_12288_seed-10/events.out.tfevents.1590545751.vsl065.107111.0 b/...es/traj-aslip_aslip_old_2048_12288_seed-10/events.out.tfevents.1590545751.vsl065.107111.0
diff --git a/...s/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/experiment.info b/...s/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: False
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 40
+num_steps: 307
+previous: None
+recurrent: False
+redis_address: None
+reward: aslip_old
+state_est: True
+traj: aslip
+use_gae: True
+viz_port: 8097
diff --git a/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/experiment.pkl b/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-10/experiment.pkl
diff --git a/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/actor.pt b/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/actor.pt
diff --git a/..._models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/critic.pt b/..._models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/critic.pt
diff --git a/...es/traj-aslip_aslip_old_2048_12288_seed-20/events.out.tfevents.1590545749.vsl066.226791.0 b/...es/traj-aslip_aslip_old_2048_12288_seed-20/events.out.tfevents.1590545749.vsl066.226791.0
diff --git a/...s/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/experiment.info b/...s/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: False
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 40
+num_steps: 307
+previous: None
+recurrent: False
+redis_address: None
+reward: aslip_old
+state_est: True
+traj: aslip
+use_gae: True
+viz_port: 8097
diff --git a/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/experiment.pkl b/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-20/experiment.pkl
diff --git a/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/actor.pt b/...d_models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/actor.pt
diff --git a/..._models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/critic.pt b/..._models/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/critic.pt
diff --git a/...es/traj-aslip_aslip_old_2048_12288_seed-30/events.out.tfevents.1590545747.vsl066.226792.0 b/...es/traj-aslip_aslip_old_2048_12288_seed-30/events.out.tfevents.1590545747.vsl066.226792.0
diff --git a/...s/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/experiment.info b/...s/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: False
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 40
+num_steps: 307
+previous: None
+recurrent: False
+redis_address: None
+reward: aslip_old
+state_est: True
+traj: aslip
+use_gae: True
+viz_port: 8097
diff --git a/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/experiment.pkl b/...ls/comparison/aslip_delta_policies/traj-aslip_aslip_old_2048_12288_seed-30/experiment.pkl
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task0_v7/actor.pt b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task0_v7/actor.pt
diff --git a/...elta_policies_RSS/aslip_unified_task0_v7/events.out.tfevents.1591220454.schnauzer.28628.0 b/...elta_policies_RSS/aslip_unified_task0_v7/events.out.tfevents.1591220454.schnauzer.28628.0
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task0_v7/experiment.info b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task0_v7/experiment.info
@@ -0,0 +1,22 @@
+clip: 0.2
+entropy_coeff: 0.0
+epochs: 3
+eps: 1e-05
+gamma: 0.99
+input_norm_steps: 10000
+lam: 0.95
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 64
+mirror: False
+n_itr: 10000
+num_procs: 22
+num_steps: 231
+policy_name: aslip_unified_task0_v7
+previous: None
+redis_address: None
+speed: 0.0
+state_est: True
+use_gae: True
+viz_port: 8097
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task0_v7/experiment.pkl b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task0_v7/experiment.pkl
diff --git a/...ls/comparison/aslip_delta_policies_RSS/aslip_unified_task10_v7/aslip_unified_task10_v7.pt b/...ls/comparison/aslip_delta_policies_RSS/aslip_unified_task10_v7/aslip_unified_task10_v7.pt
diff --git a/...elta_policies_RSS/aslip_unified_task10_v7/events.out.tfevents.1591245813.schnauzer.3063.0 b/...elta_policies_RSS/aslip_unified_task10_v7/events.out.tfevents.1591245813.schnauzer.3063.0
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task10_v7/experiment.info b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task10_v7/experiment.info
@@ -0,0 +1,22 @@
+clip: 0.2
+entropy_coeff: 0.0
+epochs: 3
+eps: 1e-05
+gamma: 0.99
+input_norm_steps: 10000
+lam: 0.95
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 64
+mirror: False
+n_itr: 10000
+num_procs: 10
+num_steps: 509
+policy_name: aslip_unified_task10_v7
+previous: None
+redis_address: None
+speed: 0.0
+state_est: True
+use_gae: True
+viz_port: 8097
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task10_v7/experiment.pkl b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task10_v7/experiment.pkl
diff --git a/...ls/comparison/aslip_delta_policies_RSS/aslip_unified_task20_v7/aslip_unified_task20_v7.pt b/...ls/comparison/aslip_delta_policies_RSS/aslip_unified_task20_v7/aslip_unified_task20_v7.pt
diff --git a/...elta_policies_RSS/aslip_unified_task20_v7/events.out.tfevents.1591245815.schnauzer.3066.0 b/...elta_policies_RSS/aslip_unified_task20_v7/events.out.tfevents.1591245815.schnauzer.3066.0
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task20_v7/experiment.info b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task20_v7/experiment.info
@@ -0,0 +1,22 @@
+clip: 0.2
+entropy_coeff: 0.0
+epochs: 3
+eps: 1e-05
+gamma: 0.99
+input_norm_steps: 10000
+lam: 0.95
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 64
+mirror: False
+n_itr: 10000
+num_procs: 10
+num_steps: 509
+policy_name: aslip_unified_task20_v7
+previous: None
+redis_address: None
+speed: 0.0
+state_est: True
+use_gae: True
+viz_port: 8097
diff --git a/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task20_v7/experiment.pkl b/trained_models/comparison/aslip_delta_policies_RSS/aslip_unified_task20_v7/experiment.pkl
diff --git a/trained_models/comparison/iros_retrain_policies/avg_GRFs_speed0.0.npy b/trained_models/comparison/iros_retrain_policies/avg_GRFs_speed0.0.npy
diff --git a/trained_models/comparison/iros_retrain_policies/avg_GRFs_speed1.0.npy b/trained_models/comparison/iros_retrain_policies/avg_GRFs_speed1.0.npy
diff --git a/trained_models/comparison/iros_retrain_policies/avg_GRFs_speed2.0.npy b/trained_models/comparison/iros_retrain_policies/avg_GRFs_speed2.0.npy
diff --git a/...comparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/actor.pt b/...comparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/actor.pt
diff --git a/...omparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/critic.pt b/...omparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/critic.pt
diff --git a/..._traj-walking_iros_paper_2048_12288_seed-0/events.out.tfevents.1590828111.vsl040.306554.0 b/..._traj-walking_iros_paper_2048_12288_seed-0/events.out.tfevents.1590828111.vsl040.306554.0
diff --git a/...son/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/experiment.info b/...son/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: True
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 40
+num_steps: 307
+previous: None
+recurrent: False
+redis_address: None
+reward: iros_paper
+state_est: True
+traj: walking
+use_gae: True
+viz_port: 8097
diff --git a/...ison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/experiment.pkl b/...ison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-0/experiment.pkl
diff --git a/...omparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/actor.pt b/...omparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/actor.pt
diff --git a/...mparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/critic.pt b/...mparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/critic.pt
diff --git a/...traj-walking_iros_paper_2048_12288_seed-10/events.out.tfevents.1590828116.vsl040.306555.0 b/...traj-walking_iros_paper_2048_12288_seed-10/events.out.tfevents.1590828116.vsl040.306555.0
diff --git a/...on/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/experiment.info b/...on/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: True
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 40
+num_steps: 307
+previous: None
+recurrent: False
+redis_address: None
+reward: iros_paper
+state_est: True
+traj: walking
+use_gae: True
+viz_port: 8097
diff --git a/...son/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/experiment.pkl b/...son/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-10/experiment.pkl
diff --git a/...omparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/actor.pt b/...omparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/actor.pt
diff --git a/...mparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/critic.pt b/...mparison/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/critic.pt
diff --git a/...traj-walking_iros_paper_2048_12288_seed-20/events.out.tfevents.1590827679.vsl025.129043.0 b/...traj-walking_iros_paper_2048_12288_seed-20/events.out.tfevents.1590827679.vsl025.129043.0
diff --git a/...on/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/experiment.info b/...on/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/experiment.info
@@ -0,0 +1,34 @@
+algo_name: ppo
+clip: 0.2
+clock_based: True
+dyn_random: False
+entropy_coeff: 0.0
+env_name: Cassie-v0
+epochs: 3
+eps: 1e-05
+exchange_reward: None
+fixed_speed: None
+gamma: 0.99
+history: 0
+hold_level: None
+ik_baseline: False
+input_norm_steps: 10000
+lam: 0.95
+learn_gains: False
+lr: 0.0001
+max_grad_norm: 0.05
+max_traj_len: 400
+minibatch_size: 2048
+mirror: False
+n_itr: 10000
+no_delta: False
+num_procs: 50
+num_steps: 245
+previous: None
+recurrent: False
+redis_address: None
+reward: iros_paper
+state_est: True
+traj: walking
+use_gae: True
+viz_port: 8097
diff --git a/...son/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/experiment.pkl b/...son/iros_retrain_policies/clock_traj-walking_iros_paper_2048_12288_seed-20/experiment.pkl