Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl

lerobot/configs/policy/act.yaml

@@ -10,7 +10,6 @@ log_freq: 250
 
 horizon: 100
 n_obs_steps: 1
-n_latency_steps: 0
 # when temporal_agg=False, n_action_steps=horizon
 n_action_steps: ${horizon}
 
@@ -57,3 +56,8 @@ policy:
 
   state_dim: ???
   action_dim: ???
+
+  delta_timestamps:
+    observation.image: [0.0]
+    observation.state: [0.0]
+    action: [0.0, 0.02, 0.04, 0.06, 0.08, 0.1, 0.12, 0.14, 0.16, 0.18, 0.2, 0.22, 0.24, 0.26, 0.28, 0.3, 0.32, 0.34, 0.36, 0.38, 0.4, 0.42, 0.44, 0.46, 0.48, 0.5, 0.52, 0.54, 0.56, 0.58, 0.6, 0.62, 0.64, 0.66, 0.68, 0.70, 0.72, 0.74, 0.76, 0.78, 0.8, 0.82, 0.84, 0.86, 0.88, 0.9, 0.92, 0.94, 0.96, 0.98, 1.0, 1.02, 1.04, 1.06, 1.08, 1.1, 1.12, 1.14, 1.16, 1.18, 1.2, 1.22, 1.24, 1.26, 1.28, 1.3, 1.32, 1.34, 1.36, 1.38, 1.40, 1.42, 1.44, 1.46, 1.48, 1.5, 1.52, 1.54, 1.56, 1.58, 1.6, 1.62, 1.64, 1.66, 1.68, 1.7, 1.72, 1.74, 1.76, 1.78, 1.8, 1.82, 1.84, 1.86, 1.88, 1.90, 1.92, 1.94, 1.96, 1.98]

lerobot/configs/policy/diffusion.yaml

@@ -16,7 +16,6 @@ seed: 100000
 horizon: 16
 n_obs_steps: 2
 n_action_steps: 8
-n_latency_steps: 0
 dataset_obs_steps: ${n_obs_steps}
 past_action_visible: False
 keypoint_visible_rate: 1.0
@@ -38,7 +37,6 @@ policy:
   shape_meta: ${shape_meta}
 
   horizon: ${horizon}
-  # n_action_steps: ${eval:'${n_action_steps}+${n_latency_steps}'}
   n_obs_steps: ${n_obs_steps}
   num_inference_steps: 100
   obs_as_global_cond: ${obs_as_global_cond}
@@ -64,6 +62,11 @@ policy:
   lr_warmup_steps: 500
   grad_clip_norm: 10
 
+  delta_timestamps:
+    observation.image: [-.1, 0]
+    observation.state: [-.1, 0]
+    action: [-.1, 0, .1, .2, .3, .4, .5, .6, .7, .8, .9, 1.0, 1.1, 1.2, 1.3, 1.4]
+
 noise_scheduler:
   _target_: diffusers.schedulers.scheduling_ddpm.DDPMScheduler
   num_train_timesteps: 100

lerobot/configs/policy/tdmpc.yaml

@@ -77,3 +77,9 @@ policy:
   num_q: 5
   mlp_dim: 512
   latent_dim: 50
+
+  delta_timestamps:
+    observation.image: "[i / ${fps} for i in range(6)]"
+    observation.state: "[i / ${fps} for i in range(6)]"
+    action: "[i / ${fps} for i in range(5)]"
+    next.reward: "[i / ${fps} for i in range(5)]"