task stringclasses 1
value | env_id stringclasses 1
value | env_kwargs dict | seed int64 2.03B 2.03B | episode_seed int64 3.03B 3.03B | episode int64 0 11.1k | extra_state dict | timestamp stringdate 2026-05-04 01:27:30 2026-05-04 03:33:07 | history listlengths 2 150 | stats dict | init_state dict | init_state_hash stringlengths 64 64 | solve_kwargs dict | version stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,042,400 | 0 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:30.636623 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":53,"reward":29.426101232799375,"terminated":true,"truncated":false,"success":true,"coverage"(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[315.73188294875774,348.(...TRUNCATED) | e7c491aef37fe63529cc188336fe61c9b3646d912ad96ed1cc24eafa7be4e436 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,043,424 | 1 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:36.270958 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":31,"reward":12.21488808625793,"terminated":true,"truncated":false,"success":true,"coverage":(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[278.37819175580785,329.(...TRUNCATED) | 22bdc06e153a3b11d6af68c52e10a2f590e71cd3149106d165f3d889ffb8ae15 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,044,192 | 2 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:40.714073 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":54,"reward":23.2499860037571,"terminated":true,"truncated":false,"success":true,"coverage":0(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[233.80003111654153,414.(...TRUNCATED) | c32e3c80551c9683cee88d8edc2fec35eb60f23e909ba2486ea2d9213a4c40d4 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,045,728 | 3 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:48.872818 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":30,"reward":10.763230187958182,"terminated":true,"truncated":false,"success":true,"coverage"(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[254.04368478847945,333.(...TRUNCATED) | ba8d2cbd54f3c0637efde2db3496a229818b750d5f12024d0464ea684bfb9a59 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,047,008 | 4 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:56.432926 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":65,"reward":32.19941545441009,"terminated":true,"truncated":false,"success":true,"coverage":(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[270.86750235698935,414.(...TRUNCATED) | e15e8da48470308ea39cb67152ae667c884e51f3412c5d6c6b56902b7abf8b81 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,047,264 | 5 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:57.997046 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":53,"reward":26.62774372497335,"terminated":true,"truncated":false,"success":true,"coverage":(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[307.35579971749945,394.(...TRUNCATED) | b6ccf4ea1b542ad6c43c7b0a982be8ad61ce718238cfc5d7aac54b49fd50faee | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,047,520 | 6 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:27:59.704847 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":56,"reward":29.407266829476754,"terminated":true,"truncated":false,"success":true,"coverage"(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[236.9424954322709,320.6(...TRUNCATED) | 873bae156564d778a0b9967350590945c9090fcda17cb54f443e3b9afc90f1b5 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,048,032 | 7 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:28:01.814044 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":21,"reward":6.860463387248062,"terminated":true,"truncated":false,"success":true,"coverage":(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[169.12797424188747,376.(...TRUNCATED) | c3fe125e4a1ef90f568689937d620a29a47686473c5500bcee03b3d086d8ed2d | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,051,360 | 8 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:28:21.089630 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":32,"reward":11.423976041783476,"terminated":true,"truncated":false,"success":true,"coverage"(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[270.66506285982325,306.(...TRUNCATED) | 9373fcd25703c84c67a71d36f840803ca39ec54cf0cb03ab9d92182f7d318f33 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
pusht_ppo_10hz_success_150step_norm4_v0 | pusht/easy | {"max_steps":150,"success_threshold":0.9,"init_distribution":"maniskill_like","control_hz":10,"easy_(...TRUNCATED) | 2,026,042,400 | 3,026,051,616 | 9 | {"checkpoint_root":"/home/raychai/VisGym_3d/visgym_training/ppo_pusht/runs/pusht_ppo_cpu256_10m_10hz(...TRUNCATED) | 2026-05-04T01:28:23.100302 | [{"step":0,"prompt":"You are controlling a 2D PushT environment from a top-down image.\n\nVisual ele(...TRUNCATED) | {"step":53,"reward":16.635341925475466,"terminated":true,"truncated":false,"success":true,"coverage"(...TRUNCATED) | {"agent_position":[157.0,486.4],"agent_velocity":[0.0,0.0],"block_position":[171.71567934297852,331.(...TRUNCATED) | 821e1671c65ae826324ef39f388d7b34ab7a33a7f6ac1792bdb7bdc0f9a3ea56 | {
"strategy": "ppo",
"checkpoint": "best_model.pt",
"max_total_steps": 150
} | visgym_pusht_ppo_success_150step_parallel_v1 |
End of preview. Expand in Data Studio
pusht_96_norm4_10hz
96px PushT PPO successful trajectory dataset.
The trajectories are generated by a 10Hz PPO PushT solver with action codec norm4, rendered images in history[*].image, image_prev, and image_next at 96x96 and JPEG quality 85.
Each episode is success-only and capped at 150 environment steps.
Splits
| split | records | format |
|---|---|---|
| train | 500,000 | gzip-compressed JSONL |
| test | 1,000 | gzip-compressed JSONL |
Train/test initial states are filtered to be disjoint by init_state_hash; see metadata/.
Coordinates in move actions use normalized [0, 1] coordinates rounded to exactly four decimal places; execution decodes them back to PushT 512x512 physical coordinates. Rendered images are 96px.
Target Hub repo: https://huggingface.co/datasets/novastar112/pusht_96_norm4_10hz
Remap
This dataset is remapped from novastar112/pusht_96_norm4_10hz. Each history step has image_prev rewritten to the same frame as image; the last history step of each episode also has image_next rewritten to that final image. Records are batched into fewer .jsonl.gz shards for faster snapshot downloads.
Remapped Hub repo: https://huggingface.co/datasets/novastar112/pusht_96_norm4_10hz_remap
- Downloads last month
- 311