robocasa_gr1.yaml

type: robocasa_gr1

# Input must already contain simulator-backed EEF33 columns (the public NVIDIA
# joint44 files do not; use the converted EEF33 LeRobot v3 release). Prompts
# resolve task_index through the converted meta/tasks.parquet; missing/blank
# prompt metadata is an error.
dataset_dir: /path/to/robocasa_gr1_data

# ── Action space ─────────────────────────────────────────────────────────────
# Only EEF is supported. Raw EEF33:
#   [L xyz3+rot6d6+hand6, R xyz3+rot6d6+hand6, waist3].
# Normalize EEF33 first, then scatter arm poses, Fourier hands, and waist into
# the shared EEF/dex-hand/reserved slots. Gripper slots 9/43 stay unmapped.
# All 33 raw dims are simulator-backed and always valid, so no per-dim
# state/action mask is needed; the unified mask marks unmapped slots false.
action_mode: eef
unify_action: true
unify_action_map: ["0-8", "10-15", "34-42", "44-49", "68-70"]
unify_state_map: null  # null = state reuses unify_action_map

# ── Frame sampling ───────────────────────────────────────────────────────────
# num_frames = state/action window length; action horizon = num_frames - 1.
# video_stride subsamples video only; state/action stay at the raw rate.
# window_stride = sliding window step (1 = every frame starts a window).
num_frames: 33
video_stride: 4
window_stride: 1
split: train

# ── Rendering (3-slot L-shape canvas) ────────────────────────────────────────
# camera_layout = [head, left/bottom-left, right/bottom-right]; null slots are
# rendered as black tiles. At 384x320: top 256Hx320W, bottom halves 128Hx160W,
# each tile BILINEAR-stretched to its slot. target_camera drives single-view
# mode and is ignored when multiview=true.
height: 384
width: 320
multiview: true
camera_layout: [observation.images.ego_view, null, null]
target_camera: observation.images.ego_view

# ── Normalization ────────────────────────────────────────────────────────────
# min-max / z-score / quantile, or null to disable in-reader normalization.
# The pooled stats file is fixed at
# <dataset_dir>/meta/robocasa_gr1_normalization_stats.npy and auto-built there
# on first use (rank 0 scans the whole root, other ranks wait) — every task
# bucket shares that one file. Pose and waist dims pool action/state because
# they share physical units; Fourier-hand actions are discrete commands while
# hand state is continuous joint angle, so those 12 dims use directional
# `eef` / `eef_state` blocks. Delete the file to force a rebuild, or build it
# up front with robocasa_gr1_stats_computation.
normalize_mode: min-max

# ── Video augmentation (optional) ────────────────────────────────────────────
# Load-time color jitter applied by the reader to each clip with the SAME
# random factors across all frames (temporal consistency), TRAIN split only.
# enabled: false keeps video byte-identical while preserving the knobs.
color_jitter:
  enabled: true
  brightness: 0.2
  contrast: 0.2
  saturation: 0.2
  hue: 0.0