Robotics
LeRobot
Safetensors
reward-model
rewind
failure-aware
farm
dexmimicgen
bimanual
farm_rewind_state_dexmg_two_arm_threading / policy_postprocessor.json
chomeed's picture
Upload checkpoint from /scratch2/gwanwoo13/chomeed/outputs/rewind/rewind_state_ours_variant7_dexmg_threading_v1_blend_mono_anchorstack_eps0.2_succmse/checkpoints/005000/pretrained_model
16e4a7c verified
Raw History Blame Contribute Delete
695 Bytes
{
"name": "policy_postprocessor",
"steps": [
{
"registry_name": "unnormalizer_processor",
"config": {
"eps": 1e-08,
"features": {
"action": {
"type": "ACTION",
"shape": [
14
]
}
},
"norm_map": {
"VISUAL": "IDENTITY",
"STATE": "MEAN_STD",
"LANGUAGE": "IDENTITY",
"REWARD": "IDENTITY"
}
},
"state_file": "policy_postprocessor_step_0_unnormalizer_processor.safetensors"
},
{
"registry_name": "device_processor",
"config": {
"device": "cpu",
"float_dtype": null
}
}
]
}