at237299966/ODE_policy_microwave

ODE Policy — R1 Lite "microwave" (two arms)

Checkpoints of the ODE policy trained on real-robot demonstrations of the Galaxea R1 Lite using both arms: the right arm puts a bowl into the microwave, the left arm works the door. The policy learns a velocity field ȧ = f(c, s, a) over the 12 joint targets of both arms and produces each 8-step action chunk by integrating that field (RK4) from the current target, instead of denoising a chunk from noise.

Files

Path	Content
checkpoints/{50,100,150,200,250,300}.ckpt	Full training checkpoints (698 MB each): model, ema_model, optimizer state dicts + the pickled Hydra config. Serve the EMA weights (the loader does this by default).
config/config.yaml, config/overrides.yaml	Resolved training config and the command-line overrides of this run
logs/logs.json.txt	Per-step / per-epoch training log (loss, lr, epoch time, val_loss every 5 epochs)
SHA256SUMS	Checksums of every uploaded file

Checkpoints are torch.save(..., pickle_module=dill) payloads written by the ODE_Policy training runner. Loading them needs that codebase at a version with two-arm support (branch r1lite_exp, commit e9cc9c0 or later: ODEImagePolicy with n_grippers / state_pose_index) and torch.load(..., pickle_module=dill, weights_only=False) — only load checkpoints you trust.

Data

at237299966/r1lite_microwave: LeRobot v2.1, 34 episodes / 14,432 frames at 15 Hz, single task "microwave".

Input / output	Definition
Cameras	head_rgb (720×1280), right_wrist_rgb and left_wrist_rgb (360×640), padded to square then resized to 256×256 (same geometry as the robot client)
agent_pos	14-D, the robot client's order: [L1..L6 (rad), L_grip (mm), R1..R6 (rad), R_grip (mm)]
ODE state a	the 12 absolute joint targets [L1..L6, R1..R6] (rad). Row convention: a[t] = commanded target at t−1 (episode start: measured joints)
Gripper head	binary per gripper (left, right), +1 close / −1 open; the label flips where the operator's gripper ramp starts
Window	8 observation frames, 8-step action chunk

Training

Model	ResNet18 (GroupNorm, random crop) per camera (3 cameras), 8 frames → condition c (512); vector field = 4-layer FiLM-MLP, width 512; proprioception enters the field only (cond_use_s=false); separate gripper head (2 logits per step)
Integration	RK4, 8 control steps per chunk, vel_scale=0.05; observation fixed over the chunk (obs_update=fixed)
Start perturbation	50% of samples, per-joint std 0.005 rad except R2 0.0414 and R3 0.0276 rad, decay 0.7 per step
Optimisation	AdamW, lr 1.5e-4, cosine schedule over 300 epochs, warm-up 500 steps, EMA
Batch	2 × A100-40GB (DDP), 40 per GPU = 80 global; 175 steps/epoch, 52,500 steps in total
Wall clock	9 h 0 min (108 s/epoch)

epoch	0	10	50	100	150	200	250	300
train loss	1.38	0.28	0.080	0.046	0.029	0.018	0.013	0.011
val loss	0.85	0.98	1.14	1.31	1.55	1.74	1.93	2.00 (ep 295)

The validation loss is a weak signal: one held-out episode, raw (non-EMA) weights, with the training start perturbation still active. Its minimum is at epoch 5 (0.80) and it rises steadily afterwards (34 demos), so compare the early and late checkpoints on the robot before picking one.

Deployment

The robot side is the unmodified openpi-r1lite closed-loop client (main_r1lite_closedloop.py, 15 Hz, executes row 0 of every (8, 14) reply; it must send all three cameras). The server reads the two-arm layout from the checkpoint, re-integrates a chunk every 8 ticks from the last target it sent and walks the client through it; both arms are commanded (no left-arm echo):

python -m roboverse_learn.il.deploy.serve_policy \
    --checkpoint checkpoints/100.ckpt --host 0.0.0.0 --port 8000 --device cuda:0 \
    --gripper_close 0 --gripper_open 100 --gripper_ramp_ticks 28 24   # gripper ramps: left right
python -m roboverse_learn.il.deploy.test_client --host 127.0.0.1 --port 8000 -n 40 --bimanual --gripper_range 0 100   # pre-flight

Set the torso to the demonstrations' pose [-0.6845, 1.4986, 0.8031] rad before a run (it is not part of the policy's state). Measured on an A100: ~21 ms on re-planning ticks, < 0.1 ms on the others; loopback round trip p95 35 ms (budget 66.7 ms). Code: ODE_Policy repository, neural_ode/real/ (data + training) and roboverse_learn/il/deploy/ (server).