quant-control-bench-policies / checkpoint_training.json
happynood's picture
Upload folder using huggingface_hub
288db43 verified
Raw
History Blame Contribute Delete
2.16 kB
{
"env": "Go1JoystickFlatTerrain",
"seed": 0,
"num_timesteps_requested": 200000000,
"num_timesteps_actual": 206438400,
"num_timesteps_tuned": 200000000,
"num_envs": 8192,
"num_envs_tuned": 8192,
"episode_length": 1000,
"obs_dim": 48,
"action_dim": 12,
"obs_key": "state",
"activation": "swish",
"hidden_sizes": [
512,
256,
128
],
"wall_clock_s": 8399.055955592998,
"peak_vram_mb": 2225,
"final_reward": 31.772350311279297,
"final_reward_std": 1.8097243309020996,
"extraction_max_abs_error": 4.466623067855835e-06,
"restored_from": null,
"progress": [
{
"step": 0,
"reward": 14.834701538085938,
"reward_std": 7.491441249847412,
"wall_clock_s": 91.15248306300055
},
{
"step": 22937600,
"reward": 20.874135971069336,
"reward_std": 5.617393970489502,
"wall_clock_s": 1066.7189738640009
},
{
"step": 45875200,
"reward": 27.87371826171875,
"reward_std": 4.395743370056152,
"wall_clock_s": 1988.3939709080005
},
{
"step": 68812800,
"reward": 30.308570861816406,
"reward_std": 3.4507648944854736,
"wall_clock_s": 2912.9838387829986
},
{
"step": 91750400,
"reward": 30.687503814697266,
"reward_std": 2.2425899505615234,
"wall_clock_s": 3820.470397779998
},
{
"step": 114688000,
"reward": 30.80521583557129,
"reward_std": 2.158846378326416,
"wall_clock_s": 4770.100509854998
},
{
"step": 137625600,
"reward": 31.07474708557129,
"reward_std": 1.9040088653564453,
"wall_clock_s": 5735.6359301500015
},
{
"step": 160563200,
"reward": 31.509870529174805,
"reward_std": 2.184321165084839,
"wall_clock_s": 6703.119132361997
},
{
"step": 183500800,
"reward": 31.633445739746094,
"reward_std": 1.9835591316223145,
"wall_clock_s": 7492.1453145390005
},
{
"step": 206438400,
"reward": 31.772350311279297,
"reward_std": 1.8097243309020996,
"wall_clock_s": 8398.231447928
}
],
"budget_was_reduced": false
}