FlowPush checkpoints

Evaluation-only checkpoints (EMA weights, fitted normalizer, resolved config, git SHA) for FlowPush: flow-matching and diffusion visuomotor policies on Push-T, a 1-step distilled student, and Push-F transfer policies. Each file is training seed 0 of the corresponding row in the repository README; evaluating it on the 50 test episodes reproduces that seed's number (the README reports 3-seed means).

file	model	success@0.95, seed 0
pusht_ddpm_unet_seed0.pt	DDPM U-Net, state obs, 100 steps	0.44
pusht_fm_unet_seed0.pt	flow-matching U-Net, state obs, 8 Euler steps (the teacher)	0.60
pusht_fm_dit_seed0.pt	flow-matching DiT (7.8M params), state obs, 8 steps	0.34
pusht_fm_unet_image_seed0.pt	flow-matching U-Net, 96×96 RGB obs, 8 steps	0.30
pusht_2rf_seed0.pt	2-rectified flow, 1 step	0.46
pusht_1step_distilled_seed0.pt	1-step distilled student	0.52
pushf_scratch_n100_seed0.pt	Push-F, trained from scratch on 100 demos, 8 steps	0.66
pushf_finetune_n100_seed0.pt	Push-F, fine-tuned from the Push-T teacher on 100 demos (lr 1e-4), 8 steps	0.78

eval_records.tgz holds the per-episode evaluation records behind the repository's results/*.csv.

Usage

git clone https://github.com/WilliamHangXu/FlowPush.git && cd FlowPush
uv sync && make data && make checkpoints
uv run python evaluate.py --ckpt checkpoints/pusht_fm_unet_seed0.pt --seeds test --points euler:8
# Push-F policies: add --env pushf

Samplers and step counts are evaluation-time choices (--points euler:1, ddim:16, …). Checkpoints are self-contained: no side files are needed.

License

MIT.