Upload folder using huggingface_hub

.gitattributes

@@ -58,3 +58,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+simulation/Maniskill_gen_new/asset/car.obj filter=lfs diff=lfs merge=lfs -text
+simulation/eval_simulation/simulation/mani_skill/utils/visualization/UbuntuSansMono-Regular.ttf filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,67 @@
+# Manipulation VLA 评测全集 (GR00T N1.7 · pi0 · Genie-Envisioner)
+
+ManiSkill 仿真上对多个 VLA 模型做的 **pairwise-OOD / all-factor / conflict / VLM** 评测的
+**可复现全集**:代码 + 仿真 + 全部评测结果(逐 ckpt 逐 seed)+ 复现步骤。**不含 ckpt 权重**(见下方 HF 链接)。
+
+怎么跑 → 见 [`REPRODUCE.md`](REPRODUCE.md)。
+
+---
+
+## 目录结构
+
+```
+README.md                 ← 本文件:总览 + 结果 + 导航
+REPRODUCE.md              ← 怎么跑每一类评测(环境/ckpt/命令)
+code/                     ← 全部评测代码(client / runner / dispatcher)
+results/
+├─ gr00t/
+│  ├─ pairgrid_hard/      6 实验 × 9 ckpt,硬口径(同色同形干扰项)
+│  │   {color_size, color_spatial, verb_spatial, spatial_size, spatial_object, verb_size}/
+│  ├─ pairgrid_easy/color_size/   软口径对照(1 干扰项, ~2× 尺寸对比)
+│  ├─ all_factor/         16 ckpt,full-factor(对齐 pi0.5 协议)
+│  └─ conflict_env/       10 实验,双轴 env success(seed42)
+├─ pi0/pairgrid_hard/     color_spatial(9 ckpt)+ color_size(3 ckpt),硬口径
+├─ genie/conflict_env/    10 实验,双轴 env success
+└─ vlm_gemini-2.5-flash/  Gemini-2.5-flash 看视频判 factor_followed + FDR_summary.txt
+simulation/
+├─ eval_simulation/       ManiSkill 仿真代码(VerbObjectColor-v1 环境)
+├─ Maniskill_gen_new/     数据采集/环境定义(get_env_id_and_color)
+└─ eval_simulation_ASSETS.tar.gz   二进制 3D 资源(解开覆盖回 eval_simulation/)
+```
+
+每个 ckpt 目录里:`*_seed42.txt / _seed40.txt / _seed41.txt`(逐 episode 明细)+ `SUMMARY.txt`(3-seed 均)。
+
+## checkpoint 权重(本仓库不含)
+- pair-grid(f6/f12/f18):HF `yqi19/gr00t_public_pair_ckpt`
+- all_factor(f50/n*):HF `yqi19/gr00t_all_factor_evaluation`
+
+---
+
+## 评测协议(一句话)
+- **pairgrid 硬口径**:同色(或同形)cube 干扰项,强制模型靠目标因子区分;task_difficulty=1.5,gpu-sim,150 步,replan=10,210 ep/seed,seeds 42/40/41。
+- **pairgrid 软口径(EASY)**:仅 color_size,1 干扰项 + ~2× 尺寸对比。
+- **all_factor**:单指令全因子采样(pi0.5 对齐),200 ep,500 步,no_distractor=0.70。
+- **conflict env**:双轴成功率(因子A / 因子B 各自命中)。
+- **VLM**:Gemini-2.5-flash 看视频判模型跟了哪个因子,算 FDR。
+> 每个实验的 target/distractor 尺寸、spatial 坐标等**精确配置**见本文件末尾「附:干扰项明细」。
+
+---
+
+## 关键结果(3-seed 均,各实验最强 ckpt)
+
+| 实验 | 模型 | 最强 ckpt | 成功率 |
+|---|---|---|---|
+| color_spatial | pi0 | stair_f18 | **54.9%** |
+| color_spatial | GR00T | stair_f18 | 51.1% |
+| spatial_size | GR00T | stair_f18 | 48.1% |
+| all_factor | GR00T | Lrandom_f50_n400 | 47.0% |
+| spatial_object | GR00T | Lrandom_f12 | 39.5% |
+| color_size (EASY) | GR00T | Lrandom_f18 | 37.2% |
+| verb_spatial | GR00T | stair_f18 | 35.5% |
+| color_size (HARD) | GR00T | Lrandom_f18 | 34.9% |
+| verb_size | GR00T | random_f12 | 24.2% |
+
+**结论:** 任务难度 color_spatial > spatial_size > spatial_object > verb_spatial > color_size ≈ verb_size;
+**含 size 的任务对 VLA 最难**(8–35%);空间类任务里 **stair_f18 普遍最强**;`random_f18` 多实验偏弱。
+
+完整逐 ckpt 表见 [`result.md`](result.md)(或各 `results/.../SUMMARY.txt`)。

REPRODUCE.md

@@ -0,0 +1,102 @@
+# 可复现评测全集 — 使用说明 (REPRODUCE)
+
+本仓库是 GR00T N1.7 / pi0 / Genie-Envisioner 在 ManiSkill 上的 pairwise-OOD、
+all-factor、conflict、VLM 评测的**可复现全集**(除 checkpoint 权重外的全部:
+代码、仿真、评测结果、复现步骤)。
+
+## 目录结构
+```
+.
+├── README.md                  # 全部实验结果总表(7 节)+ 附录:干扰项/口径明细
+├── result.md                  # GR00T pair-grid + all_factor 结果(精简版)
+├── REPRODUCE.md               # 本文件:怎么跑
+├── code/                   # 全部评测代码(client + runner + dispatcher)
+├── eval_results/              # 每个 ckpt 的逐 seed 明细 txt/jsonl
+│   ├── results_gr00t_pair_*/  #   GR00T pair-grid 硬口径(+color_size_easy)
+│   ├── results_af/            #   GR00T all_factor full-factor
+│   ├── results_pi0_pairwise/  #   pi0 pair-grid
+│   ├── conflict_env_{gr00t,genie}/  # conflict 双轴 env success
+│   └── vlm_eval/              #   Gemini-2.5-flash VLM 判定 + FDR_summary
+└── simulation/
+    ├── eval_simulation_src.tar.gz   # ManiSkill 仿真(VerbObjectColor-v1 环境)
+    ├── Maniskill_gen_new.tar.gz     # 数据采集定义(get_env_id_and_color 等)
+    └── VERSION.txt                   # 仿真 git commit
+```
+
+## 0. 准备
+
+### 0.1 仿真就位(代码已是文件夹,可在线浏览;仅二进制 3D 资源是 tar)
+```bash
+# 代码已在 simulation/eval_simulation/ 与 simulation/Maniskill_gen_new/(文件夹,可直接看)
+# 把二进制资源(网格/贴图/hdr,382 个)解开,覆盖回 eval_simulation/ 即补全可跑:
+tar xzf simulation/eval_simulation_ASSETS.tar.gz -C simulation/   # 解出 eval_simulation/.../*.stl,*.glb,*.hdr ...
+# 然后把整个 simulation/eval_simulation 与 simulation/Maniskill_gen_new 放到 /workspace/ 下
+cp -r simulation/eval_simulation   /workspace/
+cp -r simulation/Maniskill_gen_new /workspace/
+```
+> 说明:代码(.py/.urdf/.xml/.mtl/.json/.sh,864 文件 7.9M)放成文件夹便于浏览/复现;
+> 二进制 3D 资源(.stl/.glb/.obj/.png/.hdr 等 382 文件 110M)单独 tar,因为它们本就不可读、
+> 且文件数多会撞 HF 提交限流。解压后 eval_simulation 即与原始一致(1246 文件)。
+
+### 0.2 venv(三套,互相隔离)
+- `/workspace/groot_eval/.venv_ms`   ManiSkill 物理/渲染 + 评测 client(numpy/torch/mani_skill/zmq)
+- `/workspace/groot_eval/.venv_groot` GR00T N1.7 推理服务器(gr00t 包 + transformers)
+- `/venv/pi0_eval`                    pi0 openpi 推理服务器(仅 pi0 评测需要)
+
+### 0.3 checkpoint(本仓库**不含**权重,从 HF 拉)
+- pair-grid(f6/f12/f18):`yqi19/gr00t_public_pair_ckpt` → 放到 `/workspace/gr00t_pair_ckpt/<ckpt>/`
+- all_factor(f50/n*):`yqi19/gr00t_all_factor_evaluation` → `/workspace/groot_eval/gr00t_af_ckpts/<ckpt>/checkpoint-10000/`
+- pi0:`/workspace/pi0_ckpt/<ckpt>/17999/`
+
+## 1. GR00T pair-grid(硬口径,3-seed)
+单卡单 ckpt(内部跑 seed 42/40/41,起一次 GR00T zmq server):
+```bash
+cd /workspace/groot_eval
+SEEDS_OVERRIDE="42 40 41" bash code/run_groot_one_ckpt.sh <ckpt_name> <gpu> <port>
+# 例: ... color_spatial_stair_f18 0 5700
+# color_size EASY 口径: 额外 COLOR_SIZE_MODE=easy
+```
+- 实现:`code/groot_grid_eval.py`(env 构建 + HARD/EASY 干扰项 + 评分)
+- 输出:`results_gr00t_pair_<experiment>/<ckpt>/gr00t_<exp>_<ckpt>_seed<sd>.txt`
+- 实验由 ckpt 名前缀解析(color_size/color_spatial/verb_spatial/spatial_size/spatial_object/verb_size)
+
+### 多 ckpt 并行(8 卡常驻调度器)
+```bash
+tmux new-session -d -s gpair_dispatcher "bash code/gpair_dispatcher.sh"
+# 把 "<ckpt>:<seed>[:easy]" 逐行写进 logs/gr00t_pair/vs_queue.txt 即可;
+# all_factor 用 "af:<ckpt>" 写进 af_queue.txt
+```
+
+## 2. GR00T all_factor(full-factor,对齐 pi0.5)
+```bash
+bash code/run_af_one_ckpt_fast.sh <af_ckpt> <gpu> <port>
+# 内部:run_full_factor_groot.sh + groot_full_factor_main.py,seeds 40/41/42,
+# sample_n=200, 200ep, max_steps=500, no_distractor=0.70
+# 输出: results_af/<ckpt>/full_factor_<ckpt>_seed<sd>.txt
+```
+
+## 3. pi0 pair-grid
+```bash
+SEEDS_OVERRIDE="42 40 41" bash code/run_pi0_one_ckpt.sh <ckpt> <gpu> <port>
+# sim_backend=cpu(pi0 server 与 GPU-sim 冲突);实现 code/pi0_grid_eval.py
+# 输出: results_pi0_pairwise/<ckpt>/pi0_<exp>_<ckpt>_seed<sd>.txt
+```
+
+## 4. conflict-experiment(双轴 env success)
+```bash
+# GR00T: groot_main.py + run_ood_groot_inference.sh(起 GR00T server)
+# 输出 overall_<factorA>_success / overall_<factorB>_success
+```
+
+## 5. VLM 判定(Gemini-2.5-flash)+ FDR
+- 用 Gemini-2.5-flash 看 conflict 视频判 factor_followed,产出 `vlm_eval/<model>/vlm_eval_<exp>.jsonl`
+- FDR=(S_f1−S_f2)/(S_f1+S_f2),三方对照见 `eval_results/vlm_eval/FDR_summary.txt`
+
+## 6. 干扰项 / 口径明细(可复现关键)
+**见 `README.md` 末尾「附:可复现配置」** —— 每个实验的 target/distractor 尺寸、
+spatial 锚点坐标、num_distractors、指令模板、task_difficulty / max_steps / replan
+全部列清。照着即可逐 cell 复现。
+
+## 7. 结果速览
+最强:color_spatial_stair_f18(GR00T 51.1% / pi0 54.9%);含 size 的任务(color_size/verb_size)
+对 VLA 最难(~8–35%)。完整 73 ckpt × 3-seed 表见 `README.md`。

code/gpair_dispatcher.sh

@@ -0,0 +1,90 @@
+#!/usr/bin/env bash
+# gpair_dispatcher.sh — long-running tmux-resident dispatcher for GR00T pair eval.
+# Polls every 30s: for each free GPU (no gpair_g<g> tmux session), pop the next
+# job from a priority list of queue files and launch it.  Independent of Claude
+# wakeups — runs until all queue files are empty (and then continues idle, in
+# case new queues get added).
+#
+# Queue file priority order (first non-empty wins):
+#   1) cs_queue.txt           — HARD color_size残余 (highest)
+#   2) vs_queue.txt           — verb_spatial
+#   3) ecs_queue.txt          — EASY color_size
+#   4) any *_queue.txt below LOG_DIR (lex-sorted)
+#
+# Each line: "<ckpt>:<seed>".  After successful dispatch, line is removed.
+# Logs each dispatch to dispatcher.log.
+
+set -u
+ROOT=/workspace/groot_eval
+H="$ROOT/harness"
+LD="$ROOT/logs/gr00t_pair"
+DLOG="$LD/dispatcher.log"
+PRIORITY=(cs_queue.txt vs_queue.txt ecs_queue.txt af_queue.txt)
+INTERVAL=30
+PORT_BASE=5700
+
+mkdir -p "$LD"
+echo "[$(date +%F\ %H:%M:%S)] dispatcher start (pid $$, interval ${INTERVAL}s)" >> "$DLOG"
+
+# log-rotation: cap at 5000 lines
+trim_log() { [[ -f "$DLOG" ]] && [[ "$(wc -l < "$DLOG")" -gt 5000 ]] && tail -3000 "$DLOG" > "$DLOG.tmp" && mv "$DLOG.tmp" "$DLOG"; }
+
+pop_one() {
+  # echo "<ckpt>:<seed> <queue_file>" or nothing
+  local f
+  for f in "${PRIORITY[@]}"; do
+    local qf="$LD/$f"
+    [[ -s "$qf" ]] || continue
+    local line; line=$(head -1 "$qf")
+    [[ -n "$line" ]] && { echo "$line $qf"; return 0; }
+  done
+  # any other *_queue.txt
+  for qf in "$LD"/*_queue.txt; do
+    [[ -s "$qf" ]] || continue
+    case "$(basename "$qf")" in cs_queue.txt|vs_queue.txt|ecs_queue.txt) continue;; esac
+    local line; line=$(head -1 "$qf")
+    [[ -n "$line" ]] && { echo "$line $qf"; return 0; }
+  done
+  return 1
+}
+
+while true; do
+  for g in 0 1 2 3 4 5 6 7; do
+    if tmux has-session -t "gpair_g$g" 2>/dev/null; then continue; fi
+    # GPU free — try to grab next job
+    pair=$(pop_one) || continue
+    job=${pair%% *}; qf=${pair##* }
+    port=$((PORT_BASE + g))
+    # all-factor job: "af:<ckpt>"  → runs run_af_one_ckpt_fast.sh (3 seeds in one launch)
+    if [[ "$job" == af:* ]]; then
+      ck=${job#af:}
+      if [[ ! -d "/workspace/groot_eval/gr00t_af_ckpts/$ck" ]]; then
+        echo "[$(date +%H:%M:%S)] WAIT af $ck not present in gr00t_af_ckpts/" >> "$DLOG"
+        continue
+      fi
+      tmux new-session -d -s "gpair_g$g" \
+        "bash $H/run_af_one_ckpt_fast.sh $ck $g $port > $LD/r_af_${ck}.log 2>&1"
+      sed -i '1d' "$qf"
+      echo "[$(date +%H:%M:%S)] DISPATCH AF $ck → g$g (port $port) from $(basename $qf)" >> "$DLOG"
+      continue
+    fi
+    # pair-grid job: "ckpt:seed" or "ckpt:seed:mode"
+    IFS=':' read -r ck sd mode <<<"$job"
+    mode=${mode:-hard}
+    if [[ ! -d "/workspace/gr00t_pair_ckpt/$ck" ]]; then
+      echo "[$(date +%H:%M:%S)] WAIT $ck:$sd not yet downloaded (skip this round)" >> "$DLOG"
+      continue
+    fi
+    if [[ "$mode" == "easy" ]]; then
+      tag="${ck}_EASY_seed${sd}"
+    else
+      tag="${ck}_seed${sd}"
+    fi
+    tmux new-session -d -s "gpair_g$g" \
+      "COLOR_SIZE_MODE=$mode SEEDS_OVERRIDE=$sd bash $H/run_groot_one_ckpt.sh $ck $g $port > $LD/r_${tag}.log 2>&1"
+    sed -i '1d' "$qf"
+    echo "[$(date +%H:%M:%S)] DISPATCH $ck:$sd:$mode → g$g (port $port) from $(basename $qf)" >> "$DLOG"
+  done
+  trim_log
+  sleep "$INTERVAL"
+done

code/gpair_dispatcher_watchdog.sh

@@ -0,0 +1,11 @@
+#!/usr/bin/env bash
+# 每 5min 检测 gpair_dispatcher tmux 是否存活,死了就重起
+ROOT=/workspace/groot_eval
+while true; do
+  if ! tmux has-session -t gpair_dispatcher 2>/dev/null; then
+    tmux new-session -d -s gpair_dispatcher \
+      "while true; do bash $ROOT/harness/gpair_dispatcher.sh; echo \"[\$(date +%F\\ %H:%M:%S)] dispatcher exited rc=\$? — restart in 10s\" >> $ROOT/logs/gr00t_pair/dispatcher.log; sleep 10; done"
+    echo "[$(date +%F\ %H:%M:%S)] watchdog re-spawned gpair_dispatcher" >> $ROOT/logs/gr00t_pair/dispatcher.log
+  fi
+  sleep 300
+done

code/groot_client.py

@@ -0,0 +1,98 @@
+#!/usr/bin/env python3
+"""
+Minimal stand-alone client for the GR00T zmq PolicyServer
+(gr00t/policy/server_client.py :: PolicyServer).
+
+Lives in the ManiSkill venv, which deliberately does NOT have the heavy
+`gr00t` package installed.  It re-implements exactly the wire format used by
+`gr00t.policy.server_client.MsgSerializer`:
+
+  * msgpack for the envelope
+  * numpy arrays serialised with ``np.save`` into ``{"__ndarray_class__": True,
+    "as_npy": <bytes>}``
+
+Only the ``get_action`` / ``reset`` / ``ping`` endpoints are used here, none of
+which return a ``ModalityConfig``, so we do not need to model that class.
+"""
+from __future__ import annotations
+
+import io
+from typing import Any
+
+import msgpack
+import numpy as np
+import zmq
+
+
+def _encode(obj: Any):
+    if isinstance(obj, np.ndarray):
+        buf = io.BytesIO()
+        np.save(buf, obj, allow_pickle=False)
+        return {"__ndarray_class__": True, "as_npy": buf.getvalue()}
+    return obj
+
+
+def _decode(obj):
+    if isinstance(obj, dict) and "__ndarray_class__" in obj:
+        return np.load(io.BytesIO(obj["as_npy"]), allow_pickle=False)
+    return obj
+
+
+class GrootClient:
+    """REQ-socket client mirroring gr00t.policy.server_client.PolicyClient."""
+
+    def __init__(self, host: str = "127.0.0.1", port: int = 5555,
+                 timeout_ms: int = 120_000):
+        self._ctx = zmq.Context.instance()
+        self._host = host
+        self._port = port
+        self._timeout_ms = timeout_ms
+        self._connect()
+
+    def _connect(self):
+        self._sock = self._ctx.socket(zmq.REQ)
+        self._sock.setsockopt(zmq.RCVTIMEO, self._timeout_ms)
+        self._sock.setsockopt(zmq.SNDTIMEO, self._timeout_ms)
+        self._sock.setsockopt(zmq.LINGER, 0)
+        self._sock.connect(f"tcp://{self._host}:{self._port}")
+
+    def _call(self, endpoint: str, data: dict | None = None,
+              requires_input: bool = True):
+        req: dict = {"endpoint": endpoint}
+        if requires_input:
+            req["data"] = data
+        try:
+            self._sock.send(msgpack.packb(req, default=_encode))
+            msg = self._sock.recv()
+        except zmq.error.Again:
+            self._sock.close()
+            self._connect()
+            raise
+        resp = msgpack.unpackb(msg, object_hook=_decode, raw=False)
+        if isinstance(resp, dict) and "error" in resp:
+            raise RuntimeError(f"GR00T server error: {resp['error']}")
+        return resp
+
+    def ping(self) -> bool:
+        try:
+            self._call("ping", requires_input=False)
+            return True
+        except zmq.error.ZMQError:
+            self._sock.close()
+            self._connect()
+            return False
+
+    def get_action(self, observation: dict, options: dict | None = None):
+        """Returns (action_dict, info_dict)."""
+        resp = self._call("get_action",
+                          {"observation": observation, "options": options})
+        return tuple(resp)  # msgpack list -> (action, info)
+
+    def reset(self, options: dict | None = None):
+        return self._call("reset", {"options": options})
+
+    def kill_server(self):
+        try:
+            self._call("kill", requires_input=False)
+        except Exception:
+            pass

code/groot_full_factor_batch.py

@@ -0,0 +1,201 @@
+#!/usr/bin/env python3
+"""
+GR00T full-factor eval — SINGLE-PROCESS batch (method A, conflict-style).
+
+Functionally identical to running run_full_factor_groot.sh + groot_full_factor_main.py
+per cell, but the 200 cells run in ONE python process: torch/mani_skill import,
+GR00T zmq connection and GPU-sim context are initialised ONCE instead of 200×.
+All per-cell logic (cell sampling, RNG sequence, env build, success, video,
+result-line / header / overall_success format) is reused VERBATIM from
+groot_full_factor_main.py so results match the per-cell harness exactly.
+
+Usage:
+  groot_full_factor_batch.py --host H --port P --results-txt PATH --video-root DIR \
+      [--sample-n 200] [--sample-seed 42] [--seed-base 40] [--total-episodes 200] \
+      [--max-episode-steps 500] [--no-distractor-prob 0.70] [--replan-steps 5] \
+      [--sim-backend gpu] [--render-backend gpu]
+"""
+from __future__ import annotations
+
+import argparse
+import itertools
+import math
+import pathlib
+import random
+import sys
+
+import gymnasium as gym
+import imageio.v2 as imageio
+import mani_skill.envs  # noqa: F401
+import numpy as np
+
+# Reuse EVERY piece of per-cell logic from the per-cell harness → guaranteed parity.
+from groot_full_factor_main import (
+    SIZE_CONFIG,
+    COLOR_TO_ID,
+    _query_groot,
+    _spatial_xy,
+    _state8,
+    _success,
+    _to_numpy_hwc,
+    make_instruction,
+)
+from groot_client import GrootClient
+
+# Identical ordering to run_full_factor_groot.sh's sampler.
+VERBS    = ["lift", "grasp", "push", "pull", "rotate", "slide"]
+COLORS   = ["red", "yellow", "blue", "orange", "green", "black"]
+SHAPES   = ["cube", "sphere", "cup", "car", "pyramid", "star"]
+SPATIALS = ["left", "right", "middle", "front", "behind"]
+SIZES    = ["small", "large", "smaller", "larger"]
+
+
+def sample_cells(sample_n: int, sample_seed: int):
+    all_tasks = list(itertools.product(VERBS, COLORS, SHAPES, SPATIALS, SIZES))
+    if sample_n > 0:
+        rng = random.Random(sample_seed)
+        rng.shuffle(all_tasks)
+        all_tasks = all_tasks[:sample_n]
+    return all_tasks
+
+
+def run_cell(client, verb, color, shape, spatial, size, cell_seed, n_eps,
+             no_distractor_prob, max_steps, replan_steps, sim_backend,
+             render_backend, video_dir, save_wrist=True):
+    """Mirror groot_full_factor_main.eval_full_factor for ONE cell, 1 process."""
+    prompt = make_instruction(verb, size, color, shape, spatial)
+    size_cfg = SIZE_CONFIG[size]
+    object_color_id = COLOR_TO_ID[color]
+    has_comparison = size_cfg["distractor_size_scales"] is not None
+    distractor_max = 1 if has_comparison else 0
+
+    make_kw = dict(
+        obs_mode="rgb",
+        control_mode="pd_joint_pos",
+        sim_backend=sim_backend,
+        render_backend=render_backend,
+        max_episode_steps=max_steps,
+        verb=verb,
+        object_shape=shape,
+        object_color_id=object_color_id,
+        distractor_max=distractor_max,
+        object_size_jiggle=0.0,
+    )
+    env = gym.make("VerbObjectColor-v1", **make_kw)
+    video_dir.mkdir(parents=True, exist_ok=True)
+
+    rng = random.Random(cell_seed)  # same as per-cell main.py (rng=Random(args.seed))
+    successes = 0
+    try:
+        for ep in range(n_eps):
+            no_distractor = rng.random() < no_distractor_prob
+            reset_options = {
+                "obj_xy": _spatial_xy(spatial, rng),
+                "target_size_scale": size_cfg["target_size_scale"],
+            }
+            if size_cfg["distractor_size_scales"] is not None:
+                reset_options["distractor_size_scales"] = size_cfg["distractor_size_scales"]
+            if no_distractor:
+                reset_options["num_distractors"] = 0
+
+            obs, _ = env.reset(seed=cell_seed + ep, options=reset_options)
+            client.reset()
+            plan = []
+            base_w = imageio.get_writer(video_dir / f"ep{ep:03d}.mp4", fps=30)
+            wrist_w = imageio.get_writer(video_dir / f"ep{ep:03d}_wrist.mp4", fps=30) if save_wrist else None
+            done = False
+            ep_ok = False
+            try:
+                while not done:
+                    rgb_b = _to_numpy_hwc(obs["sensor_data"]["base_camera"]["rgb"])
+                    rgb_h = _to_numpy_hwc(obs["sensor_data"]["hand_camera"]["rgb"])
+                    base_w.append_data(rgb_b)
+                    if wrist_w is not None:
+                        wrist_w.append_data(rgb_h)
+                    if not plan:
+                        chunk = _query_groot(client, rgb_b, rgb_h, _state8(env), prompt)
+                        nn = min(replan_steps, len(chunk))
+                        if nn < 1:
+                            break
+                        plan = list(chunk[:nn])
+                    action = np.asarray(plan.pop(0), dtype=np.float32).ravel()[:8]
+                    obs, _r, term, trunc, info = env.step(action)
+                    if _success(info):
+                        ep_ok = True
+                    done = bool(term or trunc) or ep_ok
+            finally:
+                base_w.close()
+                if wrist_w is not None:
+                    wrist_w.close()
+            if ep_ok:
+                successes += 1
+    finally:
+        env.close()
+    return successes, n_eps, prompt
+
+
+def main():
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--host", default="127.0.0.1")
+    ap.add_argument("--port", type=int, default=5555)
+    ap.add_argument("--results-txt", required=True)
+    ap.add_argument("--video-root", required=True)
+    ap.add_argument("--sample-n", type=int, default=200)
+    ap.add_argument("--sample-seed", type=int, default=42)
+    ap.add_argument("--seed-base", type=int, default=40)
+    ap.add_argument("--total-episodes", type=int, default=200)
+    ap.add_argument("--max-episode-steps", type=int, default=500)
+    ap.add_argument("--no-distractor-prob", type=float, default=0.70)
+    ap.add_argument("--replan-steps", type=int, default=5)
+    ap.add_argument("--sim-backend", default="gpu")
+    ap.add_argument("--render-backend", default="gpu")
+    a = ap.parse_args()
+
+    cells = sample_cells(a.sample_n, a.sample_seed)
+    total_cells = len(cells)
+    n_eps = max(1, math.ceil(a.total_episodes / total_cells))
+
+    rt = pathlib.Path(a.results_txt)
+    rt.parent.mkdir(parents=True, exist_ok=True)
+    with rt.open("w") as f:
+        f.write("# Full-factor inference (GR00T N1.7) [single-process batch]\n")
+        f.write(f"sample_n={a.sample_n}  sample_seed={a.sample_seed}  total_cells={total_cells}\n")
+        f.write(f"total_episodes_target={a.total_episodes}  num_episodes_per_cell={n_eps}\n")
+        f.write(f"total_episodes_actual={total_cells * n_eps}\n")
+        f.write(f"host={a.host}  port={a.port}\n")
+        f.write(f"sim_backend={a.sim_backend}  render_backend={a.render_backend}\n")
+        f.write(f"max_episode_steps={a.max_episode_steps}  seed_base={a.seed_base}\n")
+        f.write(f"no_distractor_prob={a.no_distractor_prob}  replan_steps={a.replan_steps}\n\n")
+        f.write("index verb color shape spatial size prompt successes/total\n")
+
+    client = GrootClient(a.host, a.port)
+    tot_s = tot_n = 0
+    for i, (verb, color, shape, spatial, size) in enumerate(cells, start=1):
+        cell_seed = a.seed_base + i
+        vdir = pathlib.Path(a.video_root) / f"{verb}_{size}_{color}_{shape}_{spatial}"
+        print(f"[{i}/{total_cells}] {make_instruction(verb,size,color,shape,spatial)}", flush=True)
+        try:
+            s, n, prompt = run_cell(
+                client, verb, color, shape, spatial, size, cell_seed, n_eps,
+                a.no_distractor_prob, a.max_episode_steps, a.replan_steps,
+                a.sim_backend, a.render_backend, vdir)
+            cell_res = f"{s}/{n}"
+            tot_s += s
+            tot_n += n
+        except Exception as e:  # noqa: BLE001
+            print(f"  !! cell {i} failed: {e}", flush=True)
+            prompt = make_instruction(verb, size, color, shape, spatial)
+            cell_res = "NA"
+        with rt.open("a") as f:
+            f.write(f'{i} {verb} {color} {shape} {spatial} {size} "{prompt}" {cell_res}\n')
+
+    rate = 100.0 * tot_s / tot_n if tot_n else 0.0
+    with rt.open("a") as f:
+        f.write(f"\noverall_success={tot_s}/{tot_n} ({rate:.1f}%)\n")
+    print(f"\nDone: {tot_n} episodes across {total_cells} cells")
+    print(f"Overall: {tot_s}/{tot_n} ({rate:.1f}%)")
+    print(f"Results: {rt}")
+
+
+if __name__ == "__main__":
+    main()

code/groot_full_factor_main.py

@@ -0,0 +1,285 @@
+#!/usr/bin/env python3
+"""
+GR00T N1.7 full-factor eval — aligned 1:1 with the pi0.5 protocol in
+eval_pi0_5/examples/maniskill_full_factor/main.py.
+
+The ENVIRONMENT, cell vocabulary, instruction format, distractor / size /
+spatial logic, success criterion and the `Success rate: X / Y (Z%)` stdout
+line are COPIED VERBATIM from the pi0.5 harness so the numbers are directly
+comparable.  The only thing that differs is the policy boundary: instead of
+the openpi websocket client we drive a fine-tuned GR00T N1.7 checkpoint
+served over zmq by gr00t.eval.run_gr00t_server (same wire format the conflict
+harness uses — see groot_main.py::_query_groot).
+"""
+from __future__ import annotations
+
+import collections
+import dataclasses
+import logging
+import pathlib
+import random
+
+import gymnasium as gym
+import imageio.v2 as imageio
+import mani_skill.envs  # noqa: F401
+import numpy as np
+import torch
+import tqdm
+import tyro
+
+from groot_client import GrootClient
+
+# ── Vocabularies (VERBATIM from pi0.5 main.py) ────────────────────────────────
+TRAINING_VERBS    = ("lift", "grasp", "push", "pull", "rotate", "slide")
+TRAINING_COLORS   = ("red", "yellow", "blue", "orange", "green", "black")
+TRAINING_SHAPES   = ("cube", "sphere", "cup", "car", "pyramid", "star")
+TRAINING_SPATIALS = ("left", "right", "middle", "front", "behind")
+TRAINING_SIZES    = ("small", "large", "smaller", "larger")
+
+COLOR_TO_ID = {c: i for i, c in enumerate(TRAINING_COLORS)}
+
+VERB_TO_EN = {
+    "lift": "Lift", "grasp": "Grasp", "push": "Push",
+    "pull": "Pull", "rotate": "Rotate", "slide": "Slide",
+}
+SPATIAL_TO_PHRASE = {
+    "left": "on the left", "right": "on the right", "middle": "in the middle",
+    "front": "in front", "behind": "at the back",
+}
+SPATIAL_XY_ANCHOR = {
+    "left":   (-0.10,  0.00),
+    "right":  ( 0.10,  0.00),
+    "middle": ( 0.00,  0.00),
+    "front":  ( 0.00,  0.10),
+    "behind": ( 0.00, -0.10),
+}
+SIZE_CONFIG = {
+    "small":   dict(target_size_scale=0.72, distractor_size_scales=None),
+    "large":   dict(target_size_scale=1.34, distractor_size_scales=None),
+    "smaller": dict(target_size_scale=0.82, distractor_size_scales=[1.08]),
+    "larger":  dict(target_size_scale=1.18, distractor_size_scales=[0.92]),
+}
+
+
+def make_instruction(verb: str, size: str, color: str, shape: str, spatial: str) -> str:
+    return f"{VERB_TO_EN[verb]} the {size} {color} {shape} {SPATIAL_TO_PHRASE[spatial]}."
+
+
+@dataclasses.dataclass
+class Args:
+    # GR00T zmq policy server
+    host: str = "127.0.0.1"
+    port: int = 5555
+    replan_steps: int = 5
+
+    # Task specification (5 factors)
+    verb: str = "lift"
+    color: str = "red"
+    shape: str = "cube"
+    spatial: str = "left"
+    size: str = "small"
+    prompt: str = ""
+    """Override language instruction; if empty, auto-built from the 5 factors."""
+
+    no_distractor_prob: float = 0.70
+    """Probability per episode of forcing num_distractors=0 (pi0.5: 0.70)."""
+
+    # ManiSkill
+    num_episodes: int = 50
+    max_episode_steps: int = 500
+    sim_backend: str = "cpu"
+    render_backend: str = "cpu"
+    obs_mode: str = "rgb"
+    render_mode: str | None = None
+    seed: int = 0
+
+    # Output
+    video_out_path: str = "data/maniskill_full_factor/videos"
+    save_wrist_video: bool = True
+
+
+# ── Helpers (VERBATIM from pi0.5 main.py) ─────────────────────────────────────
+def _to_numpy_hwc(x: np.ndarray | torch.Tensor) -> np.ndarray:
+    if torch.is_tensor(x):
+        x = x.detach().float().cpu().numpy()
+    x = np.asarray(x)
+    if x.ndim == 4:
+        x = x[0]
+    if x.shape[0] in (1, 3) and x.shape[-1] != 3 and x.ndim == 3:
+        x = np.transpose(x, (1, 2, 0))
+    if np.issubdtype(x.dtype, np.floating) and x.max() <= 1.0:
+        x = (np.clip(x, 0, 1) * 255).astype(np.uint8)
+    else:
+        x = x.astype(np.uint8)
+    return np.ascontiguousarray(x)
+
+
+def _state8(env: gym.Env) -> np.ndarray:
+    qpos = env.unwrapped.agent.robot.get_qpos()
+    if torch.is_tensor(qpos):
+        qpos = qpos[0].detach().cpu().numpy()
+    qpos = np.asarray(qpos, dtype=np.float32).ravel()
+    if qpos.size >= 8:
+        return qpos[:8].copy()
+    out = np.zeros(8, dtype=np.float32)
+    out[: qpos.size] = qpos
+    return out
+
+
+def _success(info: dict) -> bool:
+    if "success" not in info:
+        return False
+    s = info["success"]
+    if torch.is_tensor(s):
+        return bool(s.squeeze().item())
+    return bool(np.asarray(s).squeeze())
+
+
+def _spatial_xy(spatial: str, rng: random.Random) -> list[float]:
+    ax, ay = SPATIAL_XY_ANCHOR[spatial]
+    return [ax + rng.uniform(-0.012, 0.012), ay + rng.uniform(-0.012, 0.012)]
+
+
+# ── GR00T policy boundary (VERBATIM from groot_main.py::_query_groot) ──────────
+def _query_groot(client: GrootClient, img_base: np.ndarray, img_wrist: np.ndarray,
+                  state8: np.ndarray, instruction: str) -> np.ndarray:
+    """Build the nested GR00T observation, query the server, return an
+    (action_horizon, 8) float32 chunk = [7 joint-pos targets, 1 gripper]."""
+    obs = {
+        "video": {
+            "image":       img_base[None, None, ...],
+            "wrist_image": img_wrist[None, None, ...],
+        },
+        "state": {
+            "arm":     state8[:7][None, None, :].astype(np.float32),
+            "gripper": state8[7:8][None, None, :].astype(np.float32),
+        },
+        "language": {
+            "annotation.human.task_description": [[instruction]],
+        },
+    }
+    action, _info = client.get_action(obs)
+    arm = np.asarray(action["arm"], dtype=np.float32)       # (1, Th, 7)
+    grip = np.asarray(action["gripper"], dtype=np.float32)  # (1, Th, 1)
+    return np.concatenate([arm[0], grip[0]], axis=-1)       # (Th, 8)
+
+
+# ── Main eval loop (env / sampling logic VERBATIM from pi0.5 main.py) ──────────
+def eval_full_factor(args: Args) -> None:
+    logging.basicConfig(level=logging.INFO, force=True)
+
+    verb    = args.verb.lower().strip()
+    color   = args.color.lower().strip()
+    shape   = args.shape.lower().strip()
+    spatial = args.spatial.lower().strip()
+    size    = args.size.lower().strip()
+
+    for val, vocab, name in [
+        (verb,    TRAINING_VERBS,    "verb"),
+        (color,   TRAINING_COLORS,   "color"),
+        (shape,   TRAINING_SHAPES,   "shape"),
+        (spatial, TRAINING_SPATIALS, "spatial"),
+        (size,    TRAINING_SIZES,    "size"),
+    ]:
+        if val not in vocab:
+            raise ValueError(f"{name}={val!r} not in {vocab}")
+
+    prompt = args.prompt.strip() or make_instruction(verb, size, color, shape, spatial)
+    logging.info("prompt=%r", prompt)
+
+    size_cfg = SIZE_CONFIG[size]
+    object_color_id = COLOR_TO_ID[color]
+
+    has_comparison = size_cfg["distractor_size_scales"] is not None
+    distractor_max = 1 if has_comparison else 0
+
+    make_kw: dict = dict(
+        obs_mode=args.obs_mode,
+        control_mode="pd_joint_pos",
+        sim_backend=args.sim_backend,
+        render_backend=args.render_backend,
+        max_episode_steps=args.max_episode_steps,
+        verb=verb,
+        object_shape=shape,
+        object_color_id=object_color_id,
+        distractor_max=distractor_max,
+        object_size_jiggle=0.0,
+    )
+    if args.render_mode is not None:
+        make_kw["render_mode"] = args.render_mode
+
+    env = gym.make("VerbObjectColor-v1", **make_kw)
+
+    video_out_path = pathlib.Path(args.video_out_path)
+    video_out_path.mkdir(parents=True, exist_ok=True)
+
+    client = GrootClient(args.host, args.port)
+    rng = random.Random(args.seed)
+
+    successes = 0
+    for ep in tqdm.tqdm(range(args.num_episodes)):
+        no_distractor = rng.random() < args.no_distractor_prob
+        reset_options: dict = {
+            "obj_xy": _spatial_xy(spatial, rng),
+            "target_size_scale": size_cfg["target_size_scale"],
+        }
+        if size_cfg["distractor_size_scales"] is not None:
+            reset_options["distractor_size_scales"] = size_cfg["distractor_size_scales"]
+        if no_distractor:
+            reset_options["num_distractors"] = 0
+
+        obs, _ = env.reset(seed=args.seed + ep, options=reset_options)
+        client.reset()
+        plan: collections.deque = collections.deque()
+
+        base_path  = video_out_path / f"ep{ep:03d}.mp4"
+        wrist_path = video_out_path / f"ep{ep:03d}_wrist.mp4"
+        writer = imageio.get_writer(base_path, fps=30)
+        wrist_writer = imageio.get_writer(wrist_path, fps=30) if args.save_wrist_video else None
+
+        done = False
+        ep_success = False
+        try:
+            while not done:
+                rgb_b = _to_numpy_hwc(obs["sensor_data"]["base_camera"]["rgb"])
+                rgb_h = _to_numpy_hwc(obs["sensor_data"]["hand_camera"]["rgb"])
+                writer.append_data(rgb_b)
+                if wrist_writer is not None:
+                    wrist_writer.append_data(rgb_h)
+
+                if not plan:
+                    st = _state8(env)
+                    chunk = _query_groot(client, rgb_b, rgb_h, st, prompt)
+                    n = min(args.replan_steps, len(chunk))
+                    if n < 1:
+                        logging.warning("Empty action chunk from policy")
+                        break
+                    plan.extend(chunk[:n])
+
+                action = np.asarray(plan.popleft(), dtype=np.float32).ravel()[:8]
+                obs, _reward, term, trunc, info = env.step(action)
+                if _success(info):
+                    ep_success = True
+                done = bool(term or trunc) or ep_success
+        finally:
+            try:
+                writer.close()
+            finally:
+                if wrist_writer is not None:
+                    wrist_writer.close()
+
+        if ep_success:
+            successes += 1
+        logging.info("Episode %d success=%s no_distractor=%s", ep, ep_success, no_distractor)
+
+    env.close()
+    rate = successes / max(args.num_episodes, 1)
+    logging.info("Success rate: %d / %d (%.1f%%)", successes, args.num_episodes, 100.0 * rate)
+
+
+def main() -> None:
+    eval_full_factor(tyro.cli(Args))
+
+
+if __name__ == "__main__":
+    main()

code/groot_grid_eval.py

@@ -0,0 +1,433 @@
+#!/usr/bin/env python3
+"""
+GR00T PAIRWISE-GRID eval (HARD口径) — mirrors pi0_grid_eval.py exactly, with
+the policy boundary swapped from openpi-websocket to GR00T zmq (GrootClient).
+Same HARD color_size / color_spatial cell construction; same env reset
+options; same per-cell scoring & results-txt format; same 3-seed protocol.
+"""
+from __future__ import annotations
+
+import argparse
+import os
+import pathlib
+import random
+import sys
+
+import numpy as np
+
+MGEN_ROOT = os.environ.get("MGEN_ROOT", "/workspace/Maniskill_gen_new")
+SIM_ROOT = os.environ.get("SIM_ROOT", "/workspace/eval_simulation/simulation")
+HARN_DIR = pathlib.Path(__file__).resolve().parent
+for _p in (str(HARN_DIR), SIM_ROOT, MGEN_ROOT):
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
+
+import gymnasium as gym  # noqa: E402
+import mani_skill.envs  # noqa: E402,F401  (registers VerbObjectColor-v1)
+from groot_client import GrootClient  # noqa: E402
+# Repo's own canonical env-id/color resolver (handles legacy routing exactly).
+from scripts.run_verb_color_shape_motion_planning import get_env_id_and_color  # noqa: E402
+
+COLORS = ("red", "yellow", "blue", "orange", "green", "black")
+SIZES = ("small", "large", "smaller", "larger", "smallest", "largest")
+SPATIALS = ("left", "right", "middle", "front", "behind")
+SHAPES = ("cube", "sphere", "cup", "car", "pyramid", "star")
+VERB_POOL = ("lift", "grasp", "push")
+VERB_CAP = {"lift": "Lift", "grasp": "Grasp", "push": "Push",
+            "pull": "Pull", "rotate": "Rotate", "slide": "Slide"}
+SPATIAL_PHRASE = {"left": "on the left", "right": "on the right",
+                  "middle": "in the middle", "front": "in front",
+                  "behind": "at the back"}
+SPATIAL_ANCHOR = {"left": (-0.10, 0.0), "right": (0.10, 0.0),
+                  "middle": (0.0, 0.0), "front": (0.0, 0.10),
+                  "behind": (0.0, -0.10)}
+
+
+def _spatial_xy(spatial, rng):
+    ax, ay = SPATIAL_ANCHOR[spatial]
+    return [ax + rng.uniform(-0.012, 0.012), ay + rng.uniform(-0.012, 0.012)]
+
+
+def _to_hwc_uint8(x):
+    import torch
+    if torch.is_tensor(x):
+        x = x.detach().float().cpu().numpy()
+    x = np.asarray(x)
+    if x.ndim == 4:
+        x = x[0]
+    if x.ndim == 3 and x.shape[0] in (1, 3) and x.shape[-1] != 3:
+        x = np.transpose(x, (1, 2, 0))
+    if np.issubdtype(x.dtype, np.floating) and x.max() <= 1.0 + 1e-6:
+        x = (np.clip(x, 0, 1) * 255).astype(np.uint8)
+    else:
+        x = x.astype(np.uint8)
+    return np.ascontiguousarray(x)
+
+
+def _state8(env):
+    import torch
+    q = env.unwrapped.agent.robot.get_qpos()
+    if torch.is_tensor(q):
+        q = q[0].detach().cpu().numpy()
+    q = np.asarray(q, dtype=np.float32).ravel()
+    out = np.zeros(8, dtype=np.float32)
+    out[: min(8, q.size)] = q[: min(8, q.size)]
+    return out
+
+
+def _success(info):
+    import torch
+    s = info.get("success", False)
+    if torch.is_tensor(s):
+        return bool(s.squeeze().item())
+    return bool(np.asarray(s).squeeze())
+
+
+# HARD口径 — same as pi0_grid_eval.HARD_SIZE
+HARD_SIZE = {
+    "small":    (0.72, [1.20], 1),
+    "large":    (1.34, [0.80], 1),
+    "smaller":  (0.82, [1.08], 1),
+    "larger":   (1.18, [0.92], 1),
+    "smallest": (0.78, [1.00, 1.24], 2),
+    "largest":  (1.26, [1.00, 0.80], 2),
+}
+
+# EASY口径 — 1 distractor only, ~2x contrast, drops smallest/largest double-distractor
+EASY_SIZE = {
+    "small":    (0.65, [1.35], 1),
+    "large":    (1.40, [0.65], 1),
+    "smaller":  (0.70, [1.30], 1),
+    "larger":   (1.30, [0.70], 1),
+    "smallest": (0.65, [1.40], 1),
+    "largest":  (1.40, [0.65], 1),
+}
+
+
+def build_color_size_cell(color, size_label, *, distractor_max_arg, task_difficulty, mode="hard"):
+    verb, shape = "lift", "cube"
+    table = HARD_SIZE if mode == "hard" else EASY_SIZE
+    t_scale, d_scales, n_d = table[size_label]
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=max(int(distractor_max_arg), n_d),
+        task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [("cube", int(color_id))] * n_d + \
+            [None] * (3 - n_d)
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": int(n_d),
+                  "target_size_scale": float(t_scale),
+                  "distractor_size_scales": [float(x) for x in d_scales]}
+    instruction = f"Lift the {size_label} {color} cube."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def build_color_spatial_cell(color, spatial, *, rng, distractor_max_arg, task_difficulty):
+    verb = rng.choice(VERB_POOL)
+    shape = "cube"
+    others = [s for s in SPATIALS if s != spatial]
+    d_spatial = rng.choice(others)
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=1, task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [("cube", int(color_id)), None, None]
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": 1, "target_size_scale": 1.0,
+                  "obj_xy": _spatial_xy(spatial, rng),
+                  "distractor_xy": [_spatial_xy(d_spatial, rng)],
+                  "distractor_size_scales": [1.0]}
+    instruction = f"{VERB_CAP[verb]} the {color} cube {SPATIAL_PHRASE[spatial]}."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def build_verb_spatial_cell(verb, spatial, *, rng, distractor_max_arg, task_difficulty):
+    """HARD: 2 same-color same-shape(cube) cubes at DIFFERENT spatial anchors.
+    Color sampled per-cell (third factor); instruction uses verb + spatial."""
+    color = rng.choice(COLORS)
+    shape = "cube"
+    others = [s for s in SPATIALS if s != spatial]
+    d_spatial = rng.choice(others)
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=1, task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [("cube", int(color_id)), None, None]
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": 1, "target_size_scale": 1.0,
+                  "obj_xy": _spatial_xy(spatial, rng),
+                  "distractor_xy": [_spatial_xy(d_spatial, rng)],
+                  "distractor_size_scales": [1.0]}
+    instruction = f"{VERB_CAP[verb]} the cube {SPATIAL_PHRASE[spatial]}."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def build_spatial_object_cell(spatial, shape, *, rng, distractor_max_arg, task_difficulty):
+    """HARD: target shape at spatial; distractor = DIFFERENT shape, same color,
+    at a DIFFERENT spatial.  Instruction uses shape + spatial phrase."""
+    color = rng.choice(COLORS)
+    verb = rng.choice(VERB_POOL)
+    d_shape = rng.choice([s for s in SHAPES if s != shape])
+    d_spatial = rng.choice([s for s in SPATIALS if s != spatial])
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=1, task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [(d_shape, int(color_id)), None, None]
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": 1, "target_size_scale": 1.0,
+                  "obj_xy": _spatial_xy(spatial, rng),
+                  "distractor_xy": [_spatial_xy(d_spatial, rng)],
+                  "distractor_size_scales": [1.0]}
+    instruction = f"{VERB_CAP[verb]} the {shape} {SPATIAL_PHRASE[spatial]}."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def build_verb_size_cell(verb, size_label, *, rng, distractor_max_arg, task_difficulty):
+    """HARD: target cube of size; distractor(s) = same-color cube differing
+    ONLY in size.  Instruction uses verb + size word."""
+    color = rng.choice(COLORS)
+    shape = "cube"
+    t_scale, d_scales, n_d = HARD_SIZE[size_label]
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=max(int(distractor_max_arg), n_d),
+        task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [("cube", int(color_id))] * n_d + \
+            [None] * (3 - n_d)
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": int(n_d),
+                  "target_size_scale": float(t_scale),
+                  "distractor_size_scales": [float(x) for x in d_scales]}
+    instruction = f"{VERB_CAP[verb]} the {size_label} cube."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def build_spatial_size_cell(spatial, size_label, *, rng, distractor_max_arg, task_difficulty):
+    """HARD: 2 same-color same-shape(cube) cubes at DIFFERENT spatial AND
+    DIFFERENT size.  Instruction uses both spatial phrase + size word."""
+    color = rng.choice(COLORS)
+    verb = rng.choice(VERB_POOL)
+    shape = "cube"
+    other_spatials = [s for s in SPATIALS if s != spatial]
+    d_spatial = rng.choice(other_spatials)
+    t_scale, _, _ = HARD_SIZE[size_label]
+    # distractor: opposite-extreme scale
+    if size_label in ("small", "smaller", "smallest"):
+        d_scale = 1.25
+    else:
+        d_scale = 0.75
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=1, task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [("cube", int(color_id)), None, None]
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": 1,
+                  "target_size_scale": float(t_scale),
+                  "obj_xy": _spatial_xy(spatial, rng),
+                  "distractor_xy": [_spatial_xy(d_spatial, rng)],
+                  "distractor_size_scales": [float(d_scale)]}
+    instruction = f"{VERB_CAP[verb]} the {size_label} cube {SPATIAL_PHRASE[spatial]}."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def _groot_action_chunk(client: GrootClient, img_base, img_wrist, state8, instruction):
+    """Build GR00T obs, call get_action, return (Th, 8) float32 chunk."""
+    obs = {
+        "video": {
+            "image":       img_base[None, None, ...],
+            "wrist_image": img_wrist[None, None, ...],
+        },
+        "state": {
+            "arm":     state8[:7][None, None, :].astype(np.float32),
+            "gripper": state8[7:8][None, None, :].astype(np.float32),
+        },
+        "language": {
+            "annotation.human.task_description": [[instruction]],
+        },
+    }
+    action, _info = client.get_action(obs)
+    arm = np.asarray(action["arm"], dtype=np.float32)        # (1, Th, 7) or (Th, 7)
+    grip = np.asarray(action["gripper"], dtype=np.float32)   # (1, Th, 1) or (Th, 1)
+    if arm.ndim == 3:
+        arm = arm[0]
+    if grip.ndim == 3:
+        grip = grip[0]
+    if grip.ndim == 1:
+        grip = grip[:, None]
+    return np.concatenate([arm, grip], axis=-1).astype(np.float32)  # (Th, 8)
+
+
+def run_cell(client, env_id, make_kw, reset_opts, instruction, *,
+             seed, sim_backend, max_steps, replan):
+    mk = dict(make_kw)
+    mk["sim_backend"] = sim_backend
+    mk["render_backend"] = sim_backend
+    env = gym.make(env_id, **mk)
+    obs, _ = env.reset(seed=seed, options=reset_opts)
+    plan = []
+    done = ok = False
+    steps = 0
+    try:
+        while not done and steps < max_steps:
+            b = _to_hwc_uint8(obs["sensor_data"]["base_camera"]["rgb"])
+            h = _to_hwc_uint8(obs["sensor_data"]["hand_camera"]["rgb"])
+            if not plan:
+                chunk = _groot_action_chunk(client, b, h, _state8(env), instruction)
+                n = min(replan, len(chunk))
+                if n < 1:
+                    break
+                plan = list(chunk[:n])
+            act = np.asarray(plan.pop(0), dtype=np.float32).ravel()[:8]
+            obs, _r, term, trunc, info = env.step(act)
+            steps += 1
+            if _success(info):
+                ok = True
+            done = bool(term or trunc) or ok
+    finally:
+        env.close()
+    return ok
+
+
+def main():
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--experiment", required=True,
+                    choices=["color_size", "color_spatial", "verb_spatial",
+                             "spatial_size", "spatial_object", "verb_size"])
+    ap.add_argument("--host", default="127.0.0.1")
+    ap.add_argument("--port", type=int, default=5600)
+    ap.add_argument("--seed", type=int, default=42)
+    ap.add_argument("--results-txt", required=True)
+    ap.add_argument("--sim-backend", default="gpu")
+    ap.add_argument("--task-difficulty", type=float, default=1.5)
+    ap.add_argument("--max-episode-steps", type=int, default=300)
+    ap.add_argument("--replan-steps", type=int, default=5)
+    ap.add_argument("--distractor-max", type=int, default=2)
+    ap.add_argument("--max-cells", type=int, default=0)
+    ap.add_argument("--target-episodes", type=int, default=0)
+    ap.add_argument("--color-size-mode", choices=["hard", "easy"], default="hard")
+    a = ap.parse_args()
+
+    if a.experiment == "color_size":
+        f1vals, f1name = COLORS, "color"
+        f2vals, f2name = SIZES, "size"
+        meta = f"fixed verb=lift shape=cube  {a.color_size_mode.upper()} same-color cube distractor"
+    elif a.experiment == "color_spatial":
+        f1vals, f1name = COLORS, "color"
+        f2vals, f2name = SPATIALS, "spatial"
+        meta = "fixed shape=cube; verb~{lift,grasp,push}; HARD same-color cube distractor"
+    elif a.experiment == "verb_spatial":
+        f1vals, f1name = VERB_POOL, "verb"
+        f2vals, f2name = SPATIALS, "spatial"
+        meta = "fixed shape=cube; color~{COLORS}; HARD same-color cube distractor at different spatial"
+    elif a.experiment == "spatial_size":
+        f1vals, f1name = SPATIALS, "spatial"
+        f2vals, f2name = SIZES, "size"
+        meta = "fixed shape=cube; color&verb~random; HARD same-color cube distractor at different spatial & different size"
+    elif a.experiment == "spatial_object":
+        f1vals, f1name = SPATIALS, "spatial"
+        f2vals, f2name = SHAPES, "shape"
+        meta = "color&verb~random; HARD same-color different-shape distractor at different spatial"
+    else:  # verb_size
+        f1vals, f1name = VERB_POOL, "verb"
+        f2vals, f2name = SIZES, "size"
+        meta = "fixed shape=cube; color~random; HARD same-color cube distractor differing only in size"
+    cells = [(c, x) for c in f1vals for x in f2vals]
+    if a.max_cells > 0:
+        cells = cells[: a.max_cells]
+
+    client = GrootClient(a.host, a.port)
+    # one ping to fail fast if server isn't up
+    if not client.ping():
+        raise SystemExit(f"GR00T server not reachable at {a.host}:{a.port}")
+    rt = pathlib.Path(a.results_txt)
+    rt.parent.mkdir(parents=True, exist_ok=True)
+    with rt.open("w") as f:
+        f.write(f"# gr00t {a.experiment} grid  {meta}  "
+                f"seed={a.seed} task_difficulty={a.task_difficulty} "
+                f"sim={a.sim_backend} cells={len(cells)}\n")
+        f.write(f"idx {f1name} {f2name} success prompt\n")
+
+    import math as _m
+    ncells = len(cells)
+    reps = max(1, _m.ceil(a.target_episodes / ncells)) if a.target_episodes > 0 else 1
+    print(f"cells={ncells} reps/cell={reps} → {ncells*reps} episodes/seed", flush=True)
+
+    succ = tot = 0
+    for idx, (f1, f2) in enumerate(cells, 1):
+        for r in range(reps):
+            rng = random.Random(a.seed * 100003 + idx * 131 + r)
+            try:
+                if a.experiment == "color_size":
+                    env_id, mk, ro, instr = build_color_size_cell(
+                        f1, f2, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty, mode=a.color_size_mode)
+                elif a.experiment == "color_spatial":
+                    env_id, mk, ro, instr = build_color_spatial_cell(
+                        f1, f2, rng=rng, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                elif a.experiment == "verb_spatial":
+                    env_id, mk, ro, instr = build_verb_spatial_cell(
+                        f1, f2, rng=rng, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                elif a.experiment == "spatial_size":
+                    env_id, mk, ro, instr = build_spatial_size_cell(
+                        f1, f2, rng=rng, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                elif a.experiment == "spatial_object":
+                    env_id, mk, ro, instr = build_spatial_object_cell(
+                        f1, f2, rng=rng, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                else:  # verb_size
+                    env_id, mk, ro, instr = build_verb_size_cell(
+                        f1, f2, rng=rng, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                ok = run_cell(client, env_id, mk, ro, instr,
+                              seed=a.seed + idx * 1000 + r,
+                              sim_backend=a.sim_backend,
+                              max_steps=a.max_episode_steps, replan=a.replan_steps)
+            except Exception as e:  # noqa: BLE001
+                print(f"[{idx}/{ncells} r{r}] FAIL {f1},{f2}: {e}", flush=True)
+                ok, instr = False, f"ERROR:{e}"
+            succ += int(ok)
+            tot += 1
+            with rt.open("a") as f:
+                f.write(f'{idx} {f1} {f2} rep{r} {int(ok)} "{instr}"\n')
+        print(f"[{idx}/{ncells}] {f1} {f2} {reps}reps → cum {succ}/{tot}",
+              flush=True)
+
+    rate = 100.0 * succ / tot if tot else 0.0
+    with rt.open("a") as f:
+        f.write(f"\noverall_success={succ}/{tot} ({rate:.1f}%)\n")
+    print(f"\nDone {a.experiment}: overall_success={succ}/{tot} ({rate:.1f}%)")
+
+
+if __name__ == "__main__":
+    main()

code/groot_main.py

@@ -0,0 +1,746 @@
+#!/usr/bin/env python3
+"""
+Roll out a fine-tuned GR00T N1.7 policy on OOD pairwise conflict experiments.
+
+This is the GR00T counterpart to genie-inference-maniskill's
+``genie_envisioner/main.py``.  The environment-construction logic
+(`_build_env_and_instruction`, all 10 experiment types incl. size / spatial),
+the rollout loop, dual-success metrics, video saving and results-file format
+are kept *verbatim* from the Genie-Envisioner version so results are directly
+comparable.  The only difference is the policy: instead of an in-process
+MVActor we query an out-of-process GR00T inference server
+(`gr00t/eval/run_gr00t_server.py`) over zmq, which keeps GR00T's heavy
+dependency set isolated from ManiSkill's.
+
+Supports all 10 experiment types:
+  verb_color | verb_object | color_object
+  verb_size  | verb_spatial
+  size_object | color_size | color_spatial | spatial_size | spatial_object
+
+Batch mode (used by run_ood_groot_inference.sh): a single process, GR00T server
+loaded once, all jobs from a JSON file executed sequentially.
+"""
+from __future__ import annotations
+
+import collections
+import dataclasses
+import datetime as _dt
+import logging
+import os
+import pathlib
+import sys
+
+import gymnasium as gym
+import numpy as np
+import tqdm
+import tyro
+
+# ── repo roots ─────────────────────────────────────────────────────────────────
+# maniskill_conflict provides both the `mani_skill` package (pip-installed) and
+# the top-level `collection_strategy` package (NOT installed; needs sys.path).
+_MANISKILL_CONFLICT_ROOT = pathlib.Path(
+    os.environ.get(
+        "MANISKILL_CONFLICT_ROOT",
+        "/workspace/groot_eval/genie_repo/maniskill_conflict",
+    )
+).resolve()
+
+# Same meta_path redirect trick as openpi / genie (harmless if no such finder).
+for _f in sys.meta_path:
+    _fmod = sys.modules.get(getattr(type(_f), "__module__", ""), None)
+    if _fmod is not None and "mani_skill" in getattr(_fmod, "MAPPING", {}):
+        _fmod.MAPPING["mani_skill"] = str(_MANISKILL_CONFLICT_ROOT / "mani_skill")
+        break
+
+if _MANISKILL_CONFLICT_ROOT.exists():
+    _s = str(_MANISKILL_CONFLICT_ROOT)
+    if _s in sys.path:
+        sys.path.remove(_s)
+    sys.path.insert(0, _s)
+
+# groot_client.py lives next to this file
+sys.path.insert(0, str(pathlib.Path(__file__).resolve().parent))
+
+import mani_skill.envs  # noqa: F401 — registers VerbObjectColor-v1
+
+from collection_strategy.lib.pairwise_task_language import VERB_TO_EN
+from collection_strategy.lib.training_vocab import (
+    THIRD_COLORS_FOR_VERB_OBJECT,
+    THIRD_OBJECTS_FOR_VERB_COLOR,
+    THIRD_VERBS_FOR_COLOR_OBJECT,
+    TRAINING_COLORS,
+    TRAINING_SHAPES,
+    TRAINING_VERBS,
+)
+from groot_client import GrootClient
+
+COLOR_TO_ID = {c: i for i, c in enumerate(TRAINING_COLORS)}
+
+# ── size / spatial vocabularies (verbatim from genie main.py) ──────────────────
+SIZES: tuple[str, ...] = ("small", "large", "smaller", "larger", "smallest", "largest")
+SIZE_SCALES: dict[str, float] = {
+    "small": 0.72, "large": 1.34,
+    "smaller": 0.82, "larger": 1.18,
+    "smallest": 0.78, "largest": 1.26,
+}
+SPATIALS: tuple[str, ...] = ("left", "right", "middle", "front", "behind")
+SPATIAL_ANCHORS: dict[str, tuple[float, float]] = {
+    "left": (-0.10, 0.00), "right": (0.10, 0.00),
+    "middle": (0.00, 0.00), "front": (0.00, 0.10), "behind": (0.00, -0.10),
+}
+SPATIAL_TO_PHRASE: dict[str, str] = {
+    "left": "on the left", "right": "on the right",
+    "middle": "in the middle", "front": "in front", "behind": "at the back",
+}
+_VERB_CAPS: dict[str, str] = {
+    "lift": "Lift", "grasp": "Grasp", "push": "Push",
+    "pull": "Pull", "rotate": "Rotate", "slide": "Slide",
+}
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Args
+# ──────────────────────────────────────────────────────────────────────────────
+
+@dataclasses.dataclass
+class Args:
+    # ── GR00T inference server ──
+    host: str = "127.0.0.1"
+    port: int = 5555
+    replan_steps: int = 5
+    """Execute this many env steps before querying the model again."""
+
+    # ── OOD pair spec ──
+    experiment: str = "verb_color"
+    """verb_color | verb_object | color_object | verb_size | verb_spatial |
+    size_object | color_size | color_spatial | spatial_size | spatial_object."""
+    pair_i: int = 0
+    pair_j: int = 1
+    run_type: str = "verb"
+    third_seed: int = 0
+    third_indices: str = "0,1"
+    """Comma-separated indices into THIRD_* list (default '0,1' matches conflict training)."""
+
+    # ── Episode settings ──
+    num_episodes: int = 20
+    max_episode_steps: int = 300
+    sim_backend: str = "gpu"
+    seed: int = 0
+
+    # ── Output ──
+    experiment_root: str = "data/conflict_groot/experiments"
+    experiment_name: str = ""
+    save_wrist_video: bool = True
+
+    # ── Batch mode (server loaded once, all runs executed in-process) ──
+    batch_jobs_file: str = ""
+    batch_results_txt: str = ""
+    batch_skip_to: int = 0
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Helpers (verbatim from genie main.py)
+# ──────────────────────────────────────────────────────────────────────────────
+
+def _to_hwc_uint8(x) -> np.ndarray:
+    try:
+        import torch
+        if torch.is_tensor(x):
+            x = x.detach().float().cpu().numpy()
+    except Exception:
+        pass
+    x = np.asarray(x)
+    if x.ndim == 4:
+        x = x[0]
+    if x.ndim == 3 and x.shape[0] in (1, 3) and x.shape[-1] != 3:
+        x = np.transpose(x, (1, 2, 0))
+    if np.issubdtype(x.dtype, np.floating) and x.max() <= 1.0 + 1e-6:
+        x = (np.clip(x, 0.0, 1.0) * 255).astype(np.uint8)
+    else:
+        x = x.astype(np.uint8)
+    return np.ascontiguousarray(x)
+
+
+def _state8(env: gym.Env) -> np.ndarray:
+    qpos = env.unwrapped.agent.robot.get_qpos()
+    try:
+        import torch
+        if torch.is_tensor(qpos):
+            qpos = qpos[0].detach().cpu().numpy()
+    except Exception:
+        pass
+    qpos = np.asarray(qpos, dtype=np.float32).ravel()
+    out = np.zeros(8, dtype=np.float32)
+    out[: min(8, len(qpos))] = qpos[: min(8, len(qpos))]
+    return out
+
+
+def _bool_info(info: dict, key: str) -> bool:
+    v = info.get(key, False)
+    try:
+        import torch
+        if torch.is_tensor(v):
+            return bool(v.squeeze().item())
+    except Exception:
+        pass
+    return bool(np.asarray(v).squeeze())
+
+
+def _parse_third_pool(full: tuple, spec: str) -> tuple:
+    idxs = [int(x.strip()) for x in spec.split(",") if x.strip()]
+    return tuple(full[i] for i in idxs)
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Environment factory  (VERBATIM from genie_envisioner/main.py)
+# ──────────────────────────────────────────────────────────────────────────────
+
+def _build_env_and_instruction(args: Args) -> tuple[gym.Env, str]:
+    """Create VerbObjectColor-v1 for the given OOD pair. Returns (env, instruction)."""
+    import random as _random
+    rng = _random.Random(args.third_seed)
+
+    i, j = args.pair_i, args.pair_j
+    assert i != j, f"pair_i must differ from pair_j, got ({i}, {j})"
+
+    verb_i = TRAINING_VERBS[i]
+    verb_j = TRAINING_VERBS[j]
+
+    _shape_pool = _parse_third_pool(THIRD_OBJECTS_FOR_VERB_COLOR, args.third_indices)
+    _color_pool = _parse_third_pool(THIRD_COLORS_FOR_VERB_OBJECT, args.third_indices)
+    _verb_pool  = _parse_third_pool(THIRD_VERBS_FOR_COLOR_OBJECT, args.third_indices)
+
+    make_kw: dict = dict(
+        obs_mode="rgb",
+        control_mode="pd_joint_pos",
+        sim_backend=args.sim_backend,
+        render_backend=args.sim_backend,
+        max_episode_steps=args.max_episode_steps,
+    )
+
+    if args.experiment == "verb_color":
+        shape = rng.choice(_shape_pool)
+        color_i = TRAINING_COLORS[i];  color_j = TRAINING_COLORS[j]
+        instruction = VERB_TO_EN[verb_i].format(color=color_j, shape=shape)
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=shape, object_color_id=COLOR_TO_ID[color_i],
+                distractor_max=3, distractor_specs=[(shape, COLOR_TO_ID[color_j]), None, None],
+            )
+        elif args.run_type == "color":
+            make_kw.update(
+                verb=verb_j, object_shape=shape, object_color_id=COLOR_TO_ID[color_j],
+                distractor_max=3, distractor_specs=[(shape, COLOR_TO_ID[color_i]), None, None],
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'color' for verb_color, got {args.run_type!r}")
+
+    elif args.experiment == "verb_object":
+        color = rng.choice(_color_pool)
+        shape_i = TRAINING_SHAPES[i];  shape_j = TRAINING_SHAPES[j]
+        instruction = VERB_TO_EN[verb_i].format(color=color, shape=shape_j)
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=shape_i, object_color_id=COLOR_TO_ID[color],
+                distractor_max=3, distractor_specs=[(shape_j, COLOR_TO_ID[color]), None, None],
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=verb_j, object_shape=shape_j, object_color_id=COLOR_TO_ID[color],
+                distractor_max=3, distractor_specs=[(shape_i, COLOR_TO_ID[color]), None, None],
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'shape' for verb_object, got {args.run_type!r}")
+
+    elif args.experiment == "color_object":
+        third_verb = rng.choice(_verb_pool)
+        color_i, shape_i = TRAINING_COLORS[i], TRAINING_SHAPES[i]
+        color_j, shape_j = TRAINING_COLORS[j], TRAINING_SHAPES[j]
+        instruction = VERB_TO_EN[third_verb].format(color=color_i, shape=shape_j)
+        if args.run_type == "color":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_i, object_color_id=COLOR_TO_ID[color_i],
+                distractor_max=3, distractor_specs=[(shape_j, COLOR_TO_ID[color_j]), None, None],
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_j, object_color_id=COLOR_TO_ID[color_j],
+                distractor_max=3, distractor_specs=[(shape_i, COLOR_TO_ID[color_i]), None, None],
+            )
+        else:
+            raise ValueError(f"run_type must be 'color' or 'shape' for color_object, got {args.run_type!r}")
+
+    elif args.experiment == "verb_size":
+        size_i = SIZES[i];   size_j = SIZES[j]
+        scale_i = SIZE_SCALES[size_i]; scale_j = SIZE_SCALES[size_j]
+        _superlative = (i // 2 == 2)
+        third_shape = rng.choice(_shape_pool)
+        instruction = f"{_VERB_CAPS[verb_i]} the {size_j} {third_shape}."
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=third_shape, object_color_id=0,
+                target_size_scale=scale_i,
+                distractor_specs=[(third_shape, 0), (third_shape, 0) if _superlative else None, None],
+                distractor_size_scales=[scale_j, 1.00] if _superlative else [scale_j],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "size":
+            make_kw.update(
+                verb=verb_j, object_shape=third_shape, object_color_id=0,
+                target_size_scale=scale_j,
+                distractor_specs=[(third_shape, 0), (third_shape, 0) if _superlative else None, None],
+                distractor_size_scales=[scale_i, 1.00] if _superlative else [scale_i],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'size' for verb_size, got {args.run_type!r}")
+
+    elif args.experiment == "verb_spatial":
+        spatial_i = SPATIALS[i]; spatial_j = SPATIALS[j]
+        third_shape = rng.choice(_shape_pool)
+        instruction = f"{_VERB_CAPS[verb_i]} the {third_shape} {SPATIAL_TO_PHRASE[spatial_j]}."
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=third_shape, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "spatial":
+            make_kw.update(
+                verb=verb_j, object_shape=third_shape, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'spatial' for verb_spatial, got {args.run_type!r}")
+
+    elif args.experiment == "size_object":
+        size_i = SIZES[i];   size_j = SIZES[j]
+        shape_i = TRAINING_SHAPES[i]; shape_j = TRAINING_SHAPES[j]
+        _superlative = (i // 2 == 2)
+        third_verb = rng.choice(_verb_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {size_i} {shape_j}."
+        if args.run_type == "size":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_i, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_i],
+                distractor_specs=[(shape_j, 0), (shape_i, 0) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_j], 1.00] if _superlative else [SIZE_SCALES[size_j]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_j, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_j],
+                distractor_specs=[(shape_i, 0), (shape_i, 0) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_i], 1.00] if _superlative else [SIZE_SCALES[size_i]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'size' or 'shape' for size_object, got {args.run_type!r}")
+
+    elif args.experiment == "color_size":
+        color_i = TRAINING_COLORS[i]; color_j = TRAINING_COLORS[j]
+        size_i = SIZES[i];            size_j = SIZES[j]
+        _superlative = (i // 2 == 2)
+        third_verb = rng.choice(_verb_pool)
+        if _superlative:
+            _neutral_color_pool = [c for c in TRAINING_COLORS if c not in (color_i, color_j)]
+            _neutral_color = rng.choice(_neutral_color_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {color_i} {size_j} cube."
+        if args.run_type == "color":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_i],
+                target_size_scale=SIZE_SCALES[size_i],
+                distractor_specs=[("cube", COLOR_TO_ID[color_j]),
+                                   ("cube", COLOR_TO_ID[_neutral_color]) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_j], 1.00] if _superlative else [SIZE_SCALES[size_j]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "size":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_j],
+                target_size_scale=SIZE_SCALES[size_j],
+                distractor_specs=[("cube", COLOR_TO_ID[color_i]),
+                                   ("cube", COLOR_TO_ID[_neutral_color]) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_i], 1.00] if _superlative else [SIZE_SCALES[size_i]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'color' or 'size' for color_size, got {args.run_type!r}")
+
+    elif args.experiment == "color_spatial":
+        color_i = TRAINING_COLORS[i]; color_j = TRAINING_COLORS[j]
+        third_verb = rng.choice(_verb_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {color_i} cube {SPATIAL_TO_PHRASE[SPATIALS[j]]}."
+        if args.run_type == "color":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_i],
+                target_size_scale=1.0,
+                distractor_specs=[("cube", COLOR_TO_ID[color_j]), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "spatial":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_j],
+                target_size_scale=1.0,
+                distractor_specs=[("cube", COLOR_TO_ID[color_i]), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'color' or 'spatial' for color_spatial, got {args.run_type!r}")
+
+    elif args.experiment == "spatial_size":
+        size_i = SIZES[i]; size_j = SIZES[j]
+        third_shape = rng.choice(_shape_pool)
+        instruction = f"Lift the {size_j} {third_shape} {SPATIAL_TO_PHRASE[SPATIALS[i]]}."
+        if args.run_type == "spatial":
+            make_kw.update(
+                verb="lift", object_shape=third_shape, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_i],
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_size_scales=[SIZE_SCALES[size_j]],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "size":
+            make_kw.update(
+                verb="lift", object_shape=third_shape, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_j],
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_size_scales=[SIZE_SCALES[size_i]],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'spatial' or 'size' for spatial_size, got {args.run_type!r}")
+
+    elif args.experiment == "spatial_object":
+        shape_i = TRAINING_SHAPES[i]; shape_j = TRAINING_SHAPES[j]
+        third_verb = rng.choice(_verb_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {shape_j} {SPATIAL_TO_PHRASE[SPATIALS[i]]}."
+        if args.run_type == "spatial":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_i, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(shape_j, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_j, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(shape_i, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'spatial' or 'shape' for spatial_object, got {args.run_type!r}")
+
+    else:
+        raise ValueError(f"Unknown experiment {args.experiment!r}")
+
+    env = gym.make("VerbObjectColor-v1", **make_kw)
+    return env, instruction
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# GR00T policy boundary
+# ──────────────────────────────────────────────────────────────────────────────
+
+def _query_groot(client: GrootClient, img_base: np.ndarray, img_wrist: np.ndarray,
+                  state8: np.ndarray, instruction: str) -> np.ndarray:
+    """Build the nested GR00T observation, query the server, return an
+    (action_horizon, 8) float32 chunk = [7 joint-pos targets, 1 gripper]."""
+    obs = {
+        "video": {
+            "image":       img_base[None, None, ...],            # (1,1,H,W,3) uint8
+            "wrist_image": img_wrist[None, None, ...],            # (1,1,H,W,3) uint8
+        },
+        "state": {
+            "arm":     state8[:7][None, None, :].astype(np.float32),   # (1,1,7)
+            "gripper": state8[7:8][None, None, :].astype(np.float32),  # (1,1,1)
+        },
+        "language": {
+            "annotation.human.task_description": [[instruction]],      # (B=1, T=1)
+        },
+    }
+    action, _info = client.get_action(obs)
+    arm = np.asarray(action["arm"], dtype=np.float32)       # (1, Th, 7)
+    grip = np.asarray(action["gripper"], dtype=np.float32)  # (1, Th, 1)
+    chunk = np.concatenate([arm[0], grip[0]], axis=-1)      # (Th, 8)
+    return chunk
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Batch eval (mirrors genie_envisioner/main.py :: batch_eval_conflict)
+# ──────────────────────────────────────────────────────────────────────────────
+
+def batch_eval_conflict(args: Args) -> None:
+    import json
+
+    logging.basicConfig(level=logging.INFO, force=True)
+
+    if not args.batch_jobs_file:
+        raise ValueError("--batch-jobs-file is required for batch mode")
+    if not args.batch_results_txt:
+        raise ValueError("--batch-results-txt is required for batch mode")
+
+    all_jobs = json.loads(pathlib.Path(args.batch_jobs_file).read_text())
+    if args.batch_skip_to > 0:
+        jobs = all_jobs[args.batch_skip_to:]
+        logging.info("Resuming from job %d (skipping first %d)",
+                     args.batch_skip_to + 1, args.batch_skip_to)
+    else:
+        jobs = all_jobs
+    results_path = pathlib.Path(args.batch_results_txt)
+    total = len(all_jobs)
+    logging.info("Batch mode: %d/%d jobs remaining, experiment=%s, server=%s:%d",
+                 len(jobs), total, args.experiment, args.host, args.port)
+
+    # ── Connect to the GR00T inference server (loaded once, reused for all jobs) ──
+    client = GrootClient(host=args.host, port=args.port)
+    for _ in range(120):
+        if client.ping():
+            break
+        import time
+        time.sleep(2.0)
+    else:
+        raise RuntimeError(f"GR00T server not reachable at {args.host}:{args.port}")
+    logging.info("GR00T server ready at %s:%d", args.host, args.port)
+
+    try:
+        import imageio.v2 as imageio
+    except ImportError:
+        import imageio  # type: ignore
+
+    _run_type_to_label = {
+        "verb": "verb_success", "color": "color_success",
+        "shape": "shape_success", "size": "size_success", "spatial": "spatial_success",
+    }
+    _SPATIAL_EXPS = {"verb_spatial", "color_spatial", "spatial_size", "spatial_object"}
+    _FIRST_RUN_MAP = {
+        "verb_spatial": "verb", "color_spatial": "color",
+        "spatial_size": "spatial", "spatial_object": "spatial",
+    }
+    _run_type_pairs = {
+        "verb_color": ("verb", "color"),   "verb_object": ("verb", "shape"),
+        "verb_size":  ("verb", "size"),    "verb_spatial": ("verb", "spatial"),
+        "color_object": ("color", "shape"), "size_object": ("size", "shape"),
+        "color_size": ("color", "size"),   "color_spatial": ("color", "spatial"),
+        "spatial_size": ("spatial", "size"), "spatial_object": ("spatial", "shape"),
+    }
+    _f1_label_map = {
+        "verb_color": "verb_success",    "verb_object": "verb_success",
+        "verb_size":  "verb_success",    "verb_spatial": "verb_success",
+        "color_object": "color_success", "size_object": "size_success",
+        "color_size": "color_success",   "color_spatial": "color_success",
+        "spatial_size": "spatial_success", "spatial_object": "spatial_success",
+    }
+    _f2_label_map = {
+        "verb_color": "color_success",   "verb_object": "shape_success",
+        "verb_size":  "size_success",    "verb_spatial": "spatial_success",
+        "color_object": "shape_success", "size_object": "shape_success",
+        "color_size": "size_success",    "color_spatial": "spatial_success",
+        "spatial_size": "size_success",  "spatial_object": "shape_success",
+    }
+    first_type = _run_type_pairs.get(args.experiment, ("", ""))[0]
+
+    first_ok = 0;  first_total = 0
+    second_ok = 0; second_total = 0
+    if args.batch_skip_to > 0 and results_path.exists():
+        import re as _re
+        for line in results_path.read_text().splitlines():
+            parts = line.split()
+            if not parts or not parts[0].isdigit():
+                continue
+            if len(parts) >= 5:
+                rt = parts[3]
+                m = _re.match(r"(\d+)/(\d+)", parts[4])
+                if m:
+                    ok, den = int(m.group(1)), int(m.group(2))
+                    if rt == first_type:
+                        first_ok += ok; first_total += den
+                    else:
+                        second_ok += ok; second_total += den
+
+    # ── Loop over jobs ───────────────────────────────────────────────────────
+    for job in jobs:
+        idx = int(job["index"])
+
+        job_args = dataclasses.replace(
+            args,
+            pair_i=int(job["pair_i"]),
+            pair_j=int(job["pair_j"]),
+            run_type=str(job["run_type"]),
+            third_seed=int(job["third_seed"]),
+            num_episodes=int(job["num_episodes"]),
+            seed=int(job["seed"]),
+            experiment_name=str(job["experiment_name"]),
+        )
+
+        logging.info("[%d/%d] pair=(%d,%d) run_type=%s  experiment=%s",
+                     idx, total, job_args.pair_i, job_args.pair_j,
+                     job_args.run_type, job_args.experiment)
+
+        env, instruction = _build_env_and_instruction(job_args)
+        logging.info("OOD instruction: %r", instruction)
+
+        timestamp = _dt.datetime.now().strftime("%Y%m%d_%H%M%S")
+        run_name = (f"{job_args.experiment_name.strip()}_{timestamp}"
+                    if job_args.experiment_name.strip()
+                    else f"{job_args.experiment}_{job_args.pair_i}_{job_args.pair_j}_{job_args.run_type}_{timestamp}")
+        exp_dir   = pathlib.Path(job_args.experiment_root) / run_name
+        video_dir = exp_dir / "video"
+        video_dir.mkdir(parents=True, exist_ok=True)
+
+        if job_args.experiment in _SPATIAL_EXPS:
+            _anchor_i = list(SPATIAL_ANCHORS[SPATIALS[job_args.pair_i]])
+            _anchor_j = list(SPATIAL_ANCHORS[SPATIALS[job_args.pair_j]])
+            if job_args.run_type == _FIRST_RUN_MAP[job_args.experiment]:
+                _reset_opts: dict = {"num_distractors": 1, "obj_xy": _anchor_i, "distractor_xy": [_anchor_j]}
+            else:
+                _reset_opts = {"num_distractors": 1, "obj_xy": _anchor_j, "distractor_xy": [_anchor_i]}
+        elif job_args.experiment in (
+            "color_object", "verb_object", "verb_color",
+            "verb_size", "size_object", "color_size",
+        ):
+            _reset_opts = {"num_distractors": 1}
+        else:
+            _reset_opts = {}
+
+        verb_successes = 0
+        factor2_successes = 0
+
+        for ep in tqdm.tqdm(range(job_args.num_episodes), desc=f"[{idx}/{total}]"):
+            obs, _ = env.reset(seed=job_args.seed + ep, options=_reset_opts)
+            client.reset()
+
+            action_plan: collections.deque = collections.deque()
+            base_writer  = imageio.get_writer(str(video_dir / f"ep{ep:03d}.mp4"), fps=30)
+            wrist_writer = (
+                imageio.get_writer(str(video_dir / f"ep{ep:03d}_wrist.mp4"), fps=30)
+                if job_args.save_wrist_video else None
+            )
+            ep_verb_ok    = False
+            ep_factor2_ok = False
+            done = False
+
+            try:
+                while not done:
+                    img_base  = _to_hwc_uint8(obs["sensor_data"]["base_camera"]["rgb"])
+                    img_wrist = _to_hwc_uint8(obs["sensor_data"]["hand_camera"]["rgb"])
+
+                    base_writer.append_data(img_base)
+                    if wrist_writer is not None:
+                        wrist_writer.append_data(img_wrist)
+
+                    if not action_plan:
+                        state = _state8(env)
+                        chunk = _query_groot(client, img_base, img_wrist, state, instruction)
+                        n = min(job_args.replan_steps, len(chunk))
+                        if n < 1:
+                            break
+                        action_plan.extend(chunk[:n])
+
+                    action = np.asarray(action_plan.popleft(), dtype=np.float32).ravel()[:8]
+                    obs, _reward, term, trunc, info = env.step(action)
+
+                    if _bool_info(info, "success_first_axis"):
+                        ep_verb_ok = True
+                    elif _bool_info(info, "success"):
+                        ep_verb_ok = True
+                    if _bool_info(info, "success_second_axis"):
+                        ep_factor2_ok = True
+
+                    done = bool(term or trunc) or (ep_verb_ok and ep_factor2_ok)
+
+            finally:
+                base_writer.close()
+                if wrist_writer is not None:
+                    wrist_writer.close()
+
+            if ep_verb_ok:
+                verb_successes += 1
+            if ep_factor2_ok:
+                factor2_successes += 1
+
+            logging.info("ep=%d  verb_ok=%s  factor2_ok=%s", ep, ep_verb_ok, ep_factor2_ok)
+
+        env.close()
+        import gc
+        gc.collect()
+
+        n = max(job_args.num_episodes, 1)
+        label1 = _run_type_to_label.get(job_args.run_type, f"{job_args.run_type}_success")
+        print(f"Success rate ({label1}): {verb_successes} / {n} ({100.0*verb_successes/n:.1f}%)")
+        sys.stdout.flush()
+
+        run_name_base = job_args.experiment_name.strip() or (
+            f"{job_args.experiment}_{job_args.pair_i}_{job_args.pair_j}_{job_args.run_type}")
+        with open(results_path, "a") as f:
+            f.write(f"{idx} {job_args.pair_i} {job_args.pair_j} {job_args.run_type} "
+                    f"{verb_successes}/{n} {run_name_base}\n")
+
+        summary_lines = [
+            f"experiment={job_args.experiment}",
+            f"pair=({job_args.pair_i},{job_args.pair_j})",
+            f"run_type={job_args.run_type}",
+            f"instruction={instruction!r}",
+            f"num_episodes={job_args.num_episodes}",
+            f"{label1}={verb_successes}/{n} ({100.0*verb_successes/n:.1f}%)",
+            f"server={job_args.host}:{job_args.port}",
+        ]
+        (exp_dir / "success_rate.txt").write_text("\n".join(summary_lines) + "\n", encoding="utf-8")
+        logging.info("Saved results to %s", str(exp_dir))
+
+        if job_args.run_type == first_type:
+            first_ok += verb_successes;  first_total += n
+        else:
+            second_ok += verb_successes; second_total += n
+
+    def _rate(s, n):
+        return f"{100.0*s/n:.1f}" if n > 0 else "0.0"
+
+    f1l = _f1_label_map.get(args.experiment, "first_success")
+    f2l = _f2_label_map.get(args.experiment, "second_success")
+    with open(results_path, "a") as f:
+        f.write(f"\noverall_{f1l}={first_ok}/{first_total} ({_rate(first_ok, first_total)}%)\n")
+        f.write(f"overall_{f2l}={second_ok}/{second_total} ({_rate(second_ok, second_total)}%)\n")
+
+    print(f"\noverall_{f1l}={first_ok}/{first_total} ({_rate(first_ok, first_total)}%)")
+    print(f"overall_{f2l}={second_ok}/{second_total} ({_rate(second_ok, second_total)}%)")
+    print(f"\nSaved summary to {results_path}")
+    print(f"Done: {total} runs for {args.experiment}")
+
+
+def eval_conflict(args: Args) -> None:
+    """Single-pair debug mode (mirrors genie eval_conflict, GR00T policy)."""
+    import json
+    import tempfile
+    job = [{
+        "index": 1, "pair_i": args.pair_i, "pair_j": args.pair_j,
+        "run_type": args.run_type, "seed": args.seed,
+        "third_seed": args.third_seed, "num_episodes": args.num_episodes,
+        "experiment_name": args.experiment_name or
+        f"{args.experiment}_{args.pair_i}_{args.pair_j}_{args.run_type}",
+    }]
+    jf = tempfile.NamedTemporaryFile("w", suffix=".json", delete=False)
+    json.dump(job, jf); jf.close()
+    rt = tempfile.NamedTemporaryFile("w", suffix=".txt", delete=False); rt.close()
+    args.batch_jobs_file = jf.name
+    args.batch_results_txt = rt.name
+    batch_eval_conflict(args)
+    print("\n--- results file ---")
+    print(pathlib.Path(rt.name).read_text())
+
+
+def main() -> None:
+    args = tyro.cli(Args)
+    if args.batch_jobs_file:
+        batch_eval_conflict(args)
+    else:
+        eval_conflict(args)
+
+
+if __name__ == "__main__":
+    main()

code/pi0_grid_eval.py

@@ -0,0 +1,293 @@
+#!/usr/bin/env python3
+"""
+pi0 PAIRWISE-GRID eval — mirrors yqi19/Maniskill_gen_new data collection
+exactly, with the MP solver replaced by a pi0 openpi-websocket policy.
+
+Implemented: color_size  (collect_pairwise_attribute.py :: experiment=color_size)
+  • sweep color(6) × size(6) full grid (small,large,smaller,larger,smallest,largest)
+  • FIXED verb=lift, shape=cube  (per collection definition)
+  • size→scene preset (verbatim from collect_pairwise_attribute._size_controls /
+    _sample_small_large_distractors):
+      small  0.72 / no distractor       large  1.34 / no distractor
+      smaller 0.82 d[1.08] n1           larger 1.18 d[0.92] n1
+      smallest 0.78 d[1.00,1.24] n2     largest 1.26 d[1.00,0.80] n2
+  • env built via the repo's own get_env_id_and_color(); size scales →
+    make_kw, num_distractors → reset options (exactly as the MP runner).
+  • instruction: "Lift the {size} {color} cube."  (color AND size in language)
+  • success = env "success"; task_difficulty configurable (harder).
+"""
+from __future__ import annotations
+
+import argparse
+import os
+import pathlib
+import random
+import sys
+
+import numpy as np
+
+MGEN_ROOT = os.environ.get("MGEN_ROOT", "/workspace/Maniskill_gen_new")
+SIM_ROOT = os.environ.get("SIM_ROOT", "/workspace/eval_simulation/simulation")
+for _p in (SIM_ROOT, MGEN_ROOT):
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
+
+import gymnasium as gym  # noqa: E402
+import mani_skill.envs  # noqa: E402,F401  (registers VerbObjectColor-v1)
+from openpi_client import image_tools  # noqa: E402
+from openpi_client import websocket_client_policy as _wcp  # noqa: E402
+# Repo's own canonical env-id/color resolver (handles legacy routing exactly).
+from scripts.run_verb_color_shape_motion_planning import get_env_id_and_color  # noqa: E402
+
+COLORS = ("red", "yellow", "blue", "orange", "green", "black")
+SIZES = ("small", "large", "smaller", "larger", "smallest", "largest")
+SPATIALS = ("left", "right", "middle", "front", "behind")
+VERB_POOL = ("lift", "grasp", "push")  # training_vocab THIRD_VERBS (first 3)
+VERB_CAP = {"lift": "Lift", "grasp": "Grasp", "push": "Push",
+            "pull": "Pull", "rotate": "Rotate", "slide": "Slide"}
+SPATIAL_PHRASE = {"left": "on the left", "right": "on the right",
+                  "middle": "in the middle", "front": "in front",
+                  "behind": "at the back"}
+SPATIAL_ANCHOR = {"left": (-0.10, 0.0), "right": (0.10, 0.0),
+                  "middle": (0.0, 0.0), "front": (0.0, 0.10),
+                  "behind": (0.0, -0.10)}
+
+
+def _spatial_xy(spatial, rng):
+    ax, ay = SPATIAL_ANCHOR[spatial]
+    return [ax + rng.uniform(-0.012, 0.012), ay + rng.uniform(-0.012, 0.012)]
+
+
+# ── size presets: VERBATIM from collect_pairwise_attribute.py ───────────────
+def _sample_small_large(size_label):
+    if size_label == "small":
+        return 0.72, None, 0
+    if size_label == "large":
+        return 1.34, None, 0
+    raise ValueError(size_label)
+
+
+def _size_controls(size_label):
+    if size_label == "smaller":
+        return 0.82, [1.08], 1
+    if size_label == "larger":
+        return 1.18, [0.92], 1
+    if size_label == "smallest":
+        return 0.78, [1.00, 1.24], 2
+    if size_label == "largest":
+        return 1.26, [1.00, 0.80], 2
+    raise ValueError(size_label)
+
+
+def _to_hwc_uint8(x):
+    import torch
+    if torch.is_tensor(x):
+        x = x.detach().float().cpu().numpy()
+    x = np.asarray(x)
+    if x.ndim == 4:
+        x = x[0]
+    if x.ndim == 3 and x.shape[0] in (1, 3) and x.shape[-1] != 3:
+        x = np.transpose(x, (1, 2, 0))
+    if np.issubdtype(x.dtype, np.floating) and x.max() <= 1.0 + 1e-6:
+        x = (np.clip(x, 0, 1) * 255).astype(np.uint8)
+    else:
+        x = x.astype(np.uint8)
+    return np.ascontiguousarray(x)
+
+
+def _state8(env):
+    import torch
+    q = env.unwrapped.agent.robot.get_qpos()
+    if torch.is_tensor(q):
+        q = q[0].detach().cpu().numpy()
+    q = np.asarray(q, dtype=np.float32).ravel()
+    out = np.zeros(8, dtype=np.float32)
+    out[: min(8, q.size)] = q[: min(8, q.size)]
+    return out
+
+
+def _success(info):
+    import torch
+    s = info.get("success", False)
+    if torch.is_tensor(s):
+        return bool(s.squeeze().item())
+    return bool(np.asarray(s).squeeze())
+
+
+# HARD color_size: EVERY cell has same-color same-shape(cube) distractor(s)
+# differing ONLY in size → model MUST use the size word to disambiguate.
+# (target_scale, [distractor_scales], n_d).  small/large now also forced a
+# contrasting same-color cube distractor (no more single-object gift cells).
+HARD_SIZE = {
+    "small":    (0.72, [1.20], 1),
+    "large":    (1.34, [0.80], 1),
+    "smaller":  (0.82, [1.08], 1),
+    "larger":   (1.18, [0.92], 1),
+    "smallest": (0.78, [1.00, 1.24], 2),
+    "largest":  (1.26, [1.00, 0.80], 2),
+}
+
+
+def build_color_size_cell(color, size_label, *, distractor_max_arg, task_difficulty):
+    """HARD: ≥1 same-color cube distractor, size-only difference."""
+    verb, shape = "lift", "cube"
+    t_scale, d_scales, n_d = HARD_SIZE[size_label]
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=max(int(distractor_max_arg), n_d),
+        task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        # distractor_specs is an env CONSTRUCTOR arg: force cube + target color
+        make_kw["distractor_specs"] = [("cube", int(color_id))] * n_d + \
+            [None] * (3 - n_d)
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": int(n_d),
+                  "target_size_scale": float(t_scale),
+                  "distractor_size_scales": [float(x) for x in d_scales]}
+    instruction = f"Lift the {size_label} {color} cube."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def build_color_spatial_cell(color, spatial, *, rng, distractor_max_arg, task_difficulty):
+    """HARD: same-color same-shape(cube) distractor at a DIFFERENT spatial
+    anchor → model MUST use the spatial phrase to disambiguate."""
+    verb = rng.choice(VERB_POOL)
+    shape = "cube"
+    others = [s for s in SPATIALS if s != spatial]
+    d_spatial = rng.choice(others)
+    env_id, color_id, extra = get_env_id_and_color(
+        verb, color, shape, distractor_max=1, task_difficulty=task_difficulty)
+    make_kw = dict(obs_mode="rgb", control_mode="pd_joint_pos",
+                   render_mode="rgb_array")
+    if env_id == "VerbObjectColor-v1":
+        make_kw.update(extra)
+        make_kw["object_size_jiggle"] = 0.0
+        make_kw["distractor_specs"] = [("cube", int(color_id)), None, None]
+    else:
+        make_kw["object_color_id"] = color_id
+    reset_opts = {"num_distractors": 1, "target_size_scale": 1.0,
+                  "obj_xy": _spatial_xy(spatial, rng),
+                  "distractor_xy": [_spatial_xy(d_spatial, rng)],
+                  "distractor_size_scales": [1.0]}
+    instruction = f"{VERB_CAP[verb]} the {color} cube {SPATIAL_PHRASE[spatial]}."
+    return env_id, make_kw, reset_opts, instruction
+
+
+def run_cell(client, env_id, make_kw, reset_opts, instruction, *,
+             seed, sim_backend, max_steps, replan, resize=224):
+    mk = dict(make_kw)
+    mk["sim_backend"] = sim_backend
+    mk["render_backend"] = sim_backend
+    env = gym.make(env_id, **mk)
+    obs, _ = env.reset(seed=seed, options=reset_opts)
+    plan = []
+    done = ok = False
+    try:
+        while not done:
+            b = _to_hwc_uint8(obs["sensor_data"]["base_camera"]["rgb"])
+            h = _to_hwc_uint8(obs["sensor_data"]["hand_camera"]["rgb"])
+            if not plan:
+                img = image_tools.convert_to_uint8(image_tools.resize_with_pad(b, resize, resize))
+                wri = image_tools.convert_to_uint8(image_tools.resize_with_pad(h, resize, resize))
+                chunk = client.infer({
+                    "observation/image": img,
+                    "observation/wrist_image": wri,
+                    "observation/state": _state8(env).astype(np.float32),
+                    "prompt": instruction,
+                })["actions"]
+                chunk = np.asarray(chunk, dtype=np.float32)
+                n = min(replan, len(chunk))
+                if n < 1:
+                    break
+                plan = list(chunk[:n])
+            act = np.asarray(plan.pop(0), dtype=np.float32).ravel()[:8]
+            obs, _r, term, trunc, info = env.step(act)
+            if _success(info):
+                ok = True
+            done = bool(term or trunc) or ok
+    finally:
+        env.close()
+    return ok
+
+
+def main():
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--experiment", required=True,
+                    choices=["color_size", "color_spatial"])
+    ap.add_argument("--host", default="127.0.0.1")
+    ap.add_argument("--port", type=int, default=8000)
+    ap.add_argument("--seed", type=int, default=42)
+    ap.add_argument("--results-txt", required=True)
+    ap.add_argument("--sim-backend", default="cpu")
+    ap.add_argument("--task-difficulty", type=float, default=1.3)
+    ap.add_argument("--max-episode-steps", type=int, default=150)
+    ap.add_argument("--replan-steps", type=int, default=10)
+    ap.add_argument("--distractor-max", type=int, default=2)
+    ap.add_argument("--max-cells", type=int, default=0)
+    ap.add_argument("--target-episodes", type=int, default=0,
+                    help=">0: run ceil(target/ncells) reps per cell (~target total)")
+    a = ap.parse_args()
+
+    if a.experiment == "color_size":
+        f2vals, f2name = SIZES, "size"
+        meta = "fixed verb=lift shape=cube"
+    else:  # color_spatial
+        f2vals, f2name = SPATIALS, "spatial"
+        meta = "fixed shape=cube; verb~{lift,grasp,push}; no distractor"
+    cells = [(c, x) for c in COLORS for x in f2vals]  # full grid
+    if a.max_cells > 0:
+        cells = cells[: a.max_cells]
+
+    client = _wcp.WebsocketClientPolicy(a.host, a.port)
+    rt = pathlib.Path(a.results_txt)
+    rt.parent.mkdir(parents=True, exist_ok=True)
+    with rt.open("w") as f:
+        f.write(f"# pi0 {a.experiment} grid  {meta}  "
+                f"seed={a.seed} task_difficulty={a.task_difficulty} "
+                f"sim={a.sim_backend} cells={len(cells)}\n")
+        f.write(f"idx color {f2name} success prompt\n")
+
+    import math as _m
+    ncells = len(cells)
+    reps = max(1, _m.ceil(a.target_episodes / ncells)) if a.target_episodes > 0 else 1
+    print(f"cells={ncells} reps/cell={reps} → {ncells*reps} episodes/seed", flush=True)
+
+    succ = tot = 0
+    for idx, (color, f2) in enumerate(cells, 1):
+        for r in range(reps):
+            rng = random.Random(a.seed * 100003 + idx * 131 + r)
+            try:
+                if a.experiment == "color_size":
+                    env_id, mk, ro, instr = build_color_size_cell(
+                        color, f2, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                else:
+                    env_id, mk, ro, instr = build_color_spatial_cell(
+                        color, f2, rng=rng, distractor_max_arg=a.distractor_max,
+                        task_difficulty=a.task_difficulty)
+                ok = run_cell(client, env_id, mk, ro, instr,
+                              seed=a.seed + idx * 1000 + r,
+                              sim_backend=a.sim_backend,
+                              max_steps=a.max_episode_steps, replan=a.replan_steps)
+            except Exception as e:  # noqa: BLE001
+                print(f"[{idx}/{ncells} r{r}] FAIL {color},{f2}: {e}", flush=True)
+                ok, instr = False, f"ERROR:{e}"
+            succ += int(ok)
+            tot += 1
+            with rt.open("a") as f:
+                f.write(f'{idx} {color} {f2} rep{r} {int(ok)} "{instr}"\n')
+        print(f"[{idx}/{ncells}] {color} {f2} {reps}reps → cum {succ}/{tot}",
+              flush=True)
+
+    rate = 100.0 * succ / tot if tot else 0.0
+    with rt.open("a") as f:
+        f.write(f"\noverall_success={succ}/{tot} ({rate:.1f}%)\n")
+    print(f"\nDone {a.experiment}: overall_success={succ}/{tot} ({rate:.1f}%)")
+
+
+if __name__ == "__main__":
+    main()

code/pi0_pairwise_main.py

@@ -0,0 +1,747 @@
+#!/usr/bin/env python3
+"""
+Roll out a fine-tuned GR00T N1.7 policy on OOD pairwise conflict experiments.
+
+This is the GR00T counterpart to genie-inference-maniskill's
+``genie_envisioner/main.py``.  The environment-construction logic
+(`_build_env_and_instruction`, all 10 experiment types incl. size / spatial),
+the rollout loop, dual-success metrics, video saving and results-file format
+are kept *verbatim* from the Genie-Envisioner version so results are directly
+comparable.  The only difference is the policy: instead of an in-process
+MVActor we query an out-of-process GR00T inference server
+(`gr00t/eval/run_gr00t_server.py`) over zmq, which keeps GR00T's heavy
+dependency set isolated from ManiSkill's.
+
+Supports all 10 experiment types:
+  verb_color | verb_object | color_object
+  verb_size  | verb_spatial
+  size_object | color_size | color_spatial | spatial_size | spatial_object
+
+Batch mode (used by run_ood_groot_inference.sh): a single process, GR00T server
+loaded once, all jobs from a JSON file executed sequentially.
+"""
+from __future__ import annotations
+
+import collections
+import dataclasses
+import datetime as _dt
+import logging
+import os
+import pathlib
+import sys
+
+import gymnasium as gym
+import numpy as np
+import tqdm
+import tyro
+
+# ── repo roots ─────────────────────────────────────────────────────────────────
+# maniskill_conflict provides both the `mani_skill` package (pip-installed) and
+# the top-level `collection_strategy` package (NOT installed; needs sys.path).
+_MANISKILL_CONFLICT_ROOT = pathlib.Path(
+    os.environ.get(
+        "MANISKILL_CONFLICT_ROOT",
+        "/workspace/groot_eval/genie_repo/maniskill_conflict",
+    )
+).resolve()
+
+# Same meta_path redirect trick as openpi / genie (harmless if no such finder).
+for _f in sys.meta_path:
+    _fmod = sys.modules.get(getattr(type(_f), "__module__", ""), None)
+    if _fmod is not None and "mani_skill" in getattr(_fmod, "MAPPING", {}):
+        _fmod.MAPPING["mani_skill"] = str(_MANISKILL_CONFLICT_ROOT / "mani_skill")
+        break
+
+if _MANISKILL_CONFLICT_ROOT.exists():
+    _s = str(_MANISKILL_CONFLICT_ROOT)
+    if _s in sys.path:
+        sys.path.remove(_s)
+    sys.path.insert(0, _s)
+
+# groot_client.py lives next to this file
+sys.path.insert(0, str(pathlib.Path(__file__).resolve().parent))
+
+import mani_skill.envs  # noqa: F401 — registers VerbObjectColor-v1
+
+from collection_strategy.lib.pairwise_task_language import VERB_TO_EN
+from collection_strategy.lib.training_vocab import (
+    THIRD_COLORS_FOR_VERB_OBJECT,
+    THIRD_OBJECTS_FOR_VERB_COLOR,
+    THIRD_VERBS_FOR_COLOR_OBJECT,
+    TRAINING_COLORS,
+    TRAINING_SHAPES,
+    TRAINING_VERBS,
+)
+from openpi_client import image_tools
+from openpi_client import websocket_client_policy as _websocket_client_policy
+
+COLOR_TO_ID = {c: i for i, c in enumerate(TRAINING_COLORS)}
+
+# ── size / spatial vocabularies (verbatim from genie main.py) ──────────────────
+SIZES: tuple[str, ...] = ("small", "large", "smaller", "larger", "smallest", "largest")
+SIZE_SCALES: dict[str, float] = {
+    "small": 0.72, "large": 1.34,
+    "smaller": 0.82, "larger": 1.18,
+    "smallest": 0.78, "largest": 1.26,
+}
+SPATIALS: tuple[str, ...] = ("left", "right", "middle", "front", "behind")
+SPATIAL_ANCHORS: dict[str, tuple[float, float]] = {
+    "left": (-0.10, 0.00), "right": (0.10, 0.00),
+    "middle": (0.00, 0.00), "front": (0.00, 0.10), "behind": (0.00, -0.10),
+}
+SPATIAL_TO_PHRASE: dict[str, str] = {
+    "left": "on the left", "right": "on the right",
+    "middle": "in the middle", "front": "in front", "behind": "at the back",
+}
+_VERB_CAPS: dict[str, str] = {
+    "lift": "Lift", "grasp": "Grasp", "push": "Push",
+    "pull": "Pull", "rotate": "Rotate", "slide": "Slide",
+}
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Args
+# ──────────────────────────────────────────────────────────────────────────────
+
+@dataclasses.dataclass
+class Args:
+    # ── GR00T inference server ──
+    host: str = "127.0.0.1"
+    port: int = 5555
+    replan_steps: int = 5
+    """Execute this many env steps before querying the model again."""
+
+    # ── OOD pair spec ──
+    experiment: str = "verb_color"
+    """verb_color | verb_object | color_object | verb_size | verb_spatial |
+    size_object | color_size | color_spatial | spatial_size | spatial_object."""
+    pair_i: int = 0
+    pair_j: int = 1
+    run_type: str = "verb"
+    third_seed: int = 0
+    third_indices: str = "0,1"
+    """Comma-separated indices into THIRD_* list (default '0,1' matches conflict training)."""
+
+    # ── Episode settings ──
+    num_episodes: int = 20
+    max_episode_steps: int = 300
+    sim_backend: str = "gpu"
+    seed: int = 0
+    task_difficulty: float = 1.0
+    """VerbObjectColor-v1 difficulty (default 1.0; >1 harder, clamped [0.5,3.0])."""
+
+    # ── Output ──
+    experiment_root: str = "data/conflict_groot/experiments"
+    experiment_name: str = ""
+    save_wrist_video: bool = True
+
+    # ── Batch mode (server loaded once, all runs executed in-process) ──
+    batch_jobs_file: str = ""
+    batch_results_txt: str = ""
+    batch_skip_to: int = 0
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Helpers (verbatim from genie main.py)
+# ──────────────────────────────────────────────────────────────────────────────
+
+def _to_hwc_uint8(x) -> np.ndarray:
+    try:
+        import torch
+        if torch.is_tensor(x):
+            x = x.detach().float().cpu().numpy()
+    except Exception:
+        pass
+    x = np.asarray(x)
+    if x.ndim == 4:
+        x = x[0]
+    if x.ndim == 3 and x.shape[0] in (1, 3) and x.shape[-1] != 3:
+        x = np.transpose(x, (1, 2, 0))
+    if np.issubdtype(x.dtype, np.floating) and x.max() <= 1.0 + 1e-6:
+        x = (np.clip(x, 0.0, 1.0) * 255).astype(np.uint8)
+    else:
+        x = x.astype(np.uint8)
+    return np.ascontiguousarray(x)
+
+
+def _state8(env: gym.Env) -> np.ndarray:
+    qpos = env.unwrapped.agent.robot.get_qpos()
+    try:
+        import torch
+        if torch.is_tensor(qpos):
+            qpos = qpos[0].detach().cpu().numpy()
+    except Exception:
+        pass
+    qpos = np.asarray(qpos, dtype=np.float32).ravel()
+    out = np.zeros(8, dtype=np.float32)
+    out[: min(8, len(qpos))] = qpos[: min(8, len(qpos))]
+    return out
+
+
+def _bool_info(info: dict, key: str) -> bool:
+    v = info.get(key, False)
+    try:
+        import torch
+        if torch.is_tensor(v):
+            return bool(v.squeeze().item())
+    except Exception:
+        pass
+    return bool(np.asarray(v).squeeze())
+
+
+def _parse_third_pool(full: tuple, spec: str) -> tuple:
+    idxs = [int(x.strip()) for x in spec.split(",") if x.strip()]
+    return tuple(full[i] for i in idxs)
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Environment factory  (VERBATIM from genie_envisioner/main.py)
+# ──────────────────────────────────────────────────────────────────────────────
+
+def _build_env_and_instruction(args: Args) -> tuple[gym.Env, str]:
+    """Create VerbObjectColor-v1 for the given OOD pair. Returns (env, instruction)."""
+    import random as _random
+    rng = _random.Random(args.third_seed)
+
+    i, j = args.pair_i, args.pair_j
+    assert i != j, f"pair_i must differ from pair_j, got ({i}, {j})"
+
+    verb_i = TRAINING_VERBS[i]
+    verb_j = TRAINING_VERBS[j]
+
+    _shape_pool = _parse_third_pool(THIRD_OBJECTS_FOR_VERB_COLOR, args.third_indices)
+    _color_pool = _parse_third_pool(THIRD_COLORS_FOR_VERB_OBJECT, args.third_indices)
+    _verb_pool  = _parse_third_pool(THIRD_VERBS_FOR_COLOR_OBJECT, args.third_indices)
+
+    make_kw: dict = dict(
+        obs_mode="rgb",
+        control_mode="pd_joint_pos",
+        sim_backend=args.sim_backend,
+        render_backend=args.sim_backend,
+        max_episode_steps=args.max_episode_steps,
+        task_difficulty=args.task_difficulty,
+    )
+
+    if args.experiment == "verb_color":
+        shape = rng.choice(_shape_pool)
+        color_i = TRAINING_COLORS[i];  color_j = TRAINING_COLORS[j]
+        instruction = VERB_TO_EN[verb_i].format(color=color_j, shape=shape)
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=shape, object_color_id=COLOR_TO_ID[color_i],
+                distractor_max=3, distractor_specs=[(shape, COLOR_TO_ID[color_j]), None, None],
+            )
+        elif args.run_type == "color":
+            make_kw.update(
+                verb=verb_j, object_shape=shape, object_color_id=COLOR_TO_ID[color_j],
+                distractor_max=3, distractor_specs=[(shape, COLOR_TO_ID[color_i]), None, None],
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'color' for verb_color, got {args.run_type!r}")
+
+    elif args.experiment == "verb_object":
+        color = rng.choice(_color_pool)
+        shape_i = TRAINING_SHAPES[i];  shape_j = TRAINING_SHAPES[j]
+        instruction = VERB_TO_EN[verb_i].format(color=color, shape=shape_j)
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=shape_i, object_color_id=COLOR_TO_ID[color],
+                distractor_max=3, distractor_specs=[(shape_j, COLOR_TO_ID[color]), None, None],
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=verb_j, object_shape=shape_j, object_color_id=COLOR_TO_ID[color],
+                distractor_max=3, distractor_specs=[(shape_i, COLOR_TO_ID[color]), None, None],
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'shape' for verb_object, got {args.run_type!r}")
+
+    elif args.experiment == "color_object":
+        third_verb = rng.choice(_verb_pool)
+        color_i, shape_i = TRAINING_COLORS[i], TRAINING_SHAPES[i]
+        color_j, shape_j = TRAINING_COLORS[j], TRAINING_SHAPES[j]
+        instruction = VERB_TO_EN[third_verb].format(color=color_i, shape=shape_j)
+        if args.run_type == "color":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_i, object_color_id=COLOR_TO_ID[color_i],
+                distractor_max=3, distractor_specs=[(shape_j, COLOR_TO_ID[color_j]), None, None],
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_j, object_color_id=COLOR_TO_ID[color_j],
+                distractor_max=3, distractor_specs=[(shape_i, COLOR_TO_ID[color_i]), None, None],
+            )
+        else:
+            raise ValueError(f"run_type must be 'color' or 'shape' for color_object, got {args.run_type!r}")
+
+    elif args.experiment == "verb_size":
+        size_i = SIZES[i];   size_j = SIZES[j]
+        scale_i = SIZE_SCALES[size_i]; scale_j = SIZE_SCALES[size_j]
+        _superlative = (i // 2 == 2)
+        third_shape = rng.choice(_shape_pool)
+        instruction = f"{_VERB_CAPS[verb_i]} the {size_j} {third_shape}."
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=third_shape, object_color_id=0,
+                target_size_scale=scale_i,
+                distractor_specs=[(third_shape, 0), (third_shape, 0) if _superlative else None, None],
+                distractor_size_scales=[scale_j, 1.00] if _superlative else [scale_j],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "size":
+            make_kw.update(
+                verb=verb_j, object_shape=third_shape, object_color_id=0,
+                target_size_scale=scale_j,
+                distractor_specs=[(third_shape, 0), (third_shape, 0) if _superlative else None, None],
+                distractor_size_scales=[scale_i, 1.00] if _superlative else [scale_i],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'size' for verb_size, got {args.run_type!r}")
+
+    elif args.experiment == "verb_spatial":
+        spatial_i = SPATIALS[i]; spatial_j = SPATIALS[j]
+        third_shape = rng.choice(_shape_pool)
+        instruction = f"{_VERB_CAPS[verb_i]} the {third_shape} {SPATIAL_TO_PHRASE[spatial_j]}."
+        if args.run_type == "verb":
+            make_kw.update(
+                verb=verb_i, object_shape=third_shape, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "spatial":
+            make_kw.update(
+                verb=verb_j, object_shape=third_shape, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'verb' or 'spatial' for verb_spatial, got {args.run_type!r}")
+
+    elif args.experiment == "size_object":
+        size_i = SIZES[i];   size_j = SIZES[j]
+        shape_i = TRAINING_SHAPES[i]; shape_j = TRAINING_SHAPES[j]
+        _superlative = (i // 2 == 2)
+        third_verb = rng.choice(_verb_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {size_i} {shape_j}."
+        if args.run_type == "size":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_i, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_i],
+                distractor_specs=[(shape_j, 0), (shape_i, 0) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_j], 1.00] if _superlative else [SIZE_SCALES[size_j]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_j, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_j],
+                distractor_specs=[(shape_i, 0), (shape_i, 0) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_i], 1.00] if _superlative else [SIZE_SCALES[size_i]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'size' or 'shape' for size_object, got {args.run_type!r}")
+
+    elif args.experiment == "color_size":
+        color_i = TRAINING_COLORS[i]; color_j = TRAINING_COLORS[j]
+        size_i = SIZES[i];            size_j = SIZES[j]
+        _superlative = (i // 2 == 2)
+        third_verb = rng.choice(_verb_pool)
+        if _superlative:
+            _neutral_color_pool = [c for c in TRAINING_COLORS if c not in (color_i, color_j)]
+            _neutral_color = rng.choice(_neutral_color_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {color_i} {size_j} cube."
+        if args.run_type == "color":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_i],
+                target_size_scale=SIZE_SCALES[size_i],
+                distractor_specs=[("cube", COLOR_TO_ID[color_j]),
+                                   ("cube", COLOR_TO_ID[_neutral_color]) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_j], 1.00] if _superlative else [SIZE_SCALES[size_j]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "size":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_j],
+                target_size_scale=SIZE_SCALES[size_j],
+                distractor_specs=[("cube", COLOR_TO_ID[color_i]),
+                                   ("cube", COLOR_TO_ID[_neutral_color]) if _superlative else None, None],
+                distractor_size_scales=[SIZE_SCALES[size_i], 1.00] if _superlative else [SIZE_SCALES[size_i]],
+                distractor_max=2 if _superlative else 1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'color' or 'size' for color_size, got {args.run_type!r}")
+
+    elif args.experiment == "color_spatial":
+        color_i = TRAINING_COLORS[i]; color_j = TRAINING_COLORS[j]
+        third_verb = rng.choice(_verb_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {color_i} cube {SPATIAL_TO_PHRASE[SPATIALS[j]]}."
+        if args.run_type == "color":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_i],
+                target_size_scale=1.0,
+                distractor_specs=[("cube", COLOR_TO_ID[color_j]), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "spatial":
+            make_kw.update(
+                verb=third_verb, object_shape="cube", object_color_id=COLOR_TO_ID[color_j],
+                target_size_scale=1.0,
+                distractor_specs=[("cube", COLOR_TO_ID[color_i]), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'color' or 'spatial' for color_spatial, got {args.run_type!r}")
+
+    elif args.experiment == "spatial_size":
+        size_i = SIZES[i]; size_j = SIZES[j]
+        third_shape = rng.choice(_shape_pool)
+        instruction = f"Lift the {size_j} {third_shape} {SPATIAL_TO_PHRASE[SPATIALS[i]]}."
+        if args.run_type == "spatial":
+            make_kw.update(
+                verb="lift", object_shape=third_shape, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_i],
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_size_scales=[SIZE_SCALES[size_j]],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "size":
+            make_kw.update(
+                verb="lift", object_shape=third_shape, object_color_id=0,
+                target_size_scale=SIZE_SCALES[size_j],
+                distractor_specs=[(third_shape, 0), None, None],
+                distractor_size_scales=[SIZE_SCALES[size_i]],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'spatial' or 'size' for spatial_size, got {args.run_type!r}")
+
+    elif args.experiment == "spatial_object":
+        shape_i = TRAINING_SHAPES[i]; shape_j = TRAINING_SHAPES[j]
+        third_verb = rng.choice(_verb_pool)
+        instruction = f"{_VERB_CAPS[third_verb]} the {shape_j} {SPATIAL_TO_PHRASE[SPATIALS[i]]}."
+        if args.run_type == "spatial":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_i, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(shape_j, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        elif args.run_type == "shape":
+            make_kw.update(
+                verb=third_verb, object_shape=shape_j, object_color_id=0,
+                target_size_scale=1.0,
+                distractor_specs=[(shape_i, 0), None, None],
+                distractor_max=1, object_size_jiggle=0.0,
+            )
+        else:
+            raise ValueError(f"run_type must be 'spatial' or 'shape' for spatial_object, got {args.run_type!r}")
+
+    else:
+        raise ValueError(f"Unknown experiment {args.experiment!r}")
+
+    env = gym.make("VerbObjectColor-v1", **make_kw)
+    return env, instruction
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# pi0 policy boundary (openpi websocket) — mirrors
+# eval_pi0/examples/maniskill_full_factor/main.py
+# ──────────────────────────────────────────────────────────────────────────────
+
+def _query_pi0(client, img_base: np.ndarray, img_wrist: np.ndarray,
+                state8: np.ndarray, instruction: str,
+                resize_size: int = 224) -> np.ndarray:
+    """Resize cameras, query the openpi pi0 server, return an action chunk
+    (T, action_dim); caller slices [:8] for the env."""
+    img = image_tools.convert_to_uint8(
+        image_tools.resize_with_pad(img_base, resize_size, resize_size))
+    wrist = image_tools.convert_to_uint8(
+        image_tools.resize_with_pad(img_wrist, resize_size, resize_size))
+    chunk = client.infer({
+        "observation/image":       img,
+        "observation/wrist_image": wrist,
+        "observation/state":       state8.astype(np.float32),
+        "prompt":                  instruction,
+    })["actions"]
+    return np.asarray(chunk, dtype=np.float32)
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Batch eval (mirrors genie_envisioner/main.py :: batch_eval_conflict)
+# ──────────────────────────────────────────────────────────────────────────────
+
+def batch_eval_conflict(args: Args) -> None:
+    import json
+
+    logging.basicConfig(level=logging.INFO, force=True)
+
+    if not args.batch_jobs_file:
+        raise ValueError("--batch-jobs-file is required for batch mode")
+    if not args.batch_results_txt:
+        raise ValueError("--batch-results-txt is required for batch mode")
+
+    all_jobs = json.loads(pathlib.Path(args.batch_jobs_file).read_text())
+    if args.batch_skip_to > 0:
+        jobs = all_jobs[args.batch_skip_to:]
+        logging.info("Resuming from job %d (skipping first %d)",
+                     args.batch_skip_to + 1, args.batch_skip_to)
+    else:
+        jobs = all_jobs
+    results_path = pathlib.Path(args.batch_results_txt)
+    total = len(all_jobs)
+    logging.info("Batch mode: %d/%d jobs remaining, experiment=%s, server=%s:%d",
+                 len(jobs), total, args.experiment, args.host, args.port)
+
+    # ── Connect to the pi0 openpi websocket policy server (loaded once) ──
+    import time as _t
+    client = None
+    for _ in range(120):
+        try:
+            client = _websocket_client_policy.WebsocketClientPolicy(args.host, args.port)
+            break
+        except Exception:
+            _t.sleep(2.0)
+    if client is None:
+        raise RuntimeError(f"pi0 server not reachable at {args.host}:{args.port}")
+    logging.info("pi0 server ready at %s:%d", args.host, args.port)
+
+    try:
+        import imageio.v2 as imageio
+    except ImportError:
+        import imageio  # type: ignore
+
+    _run_type_to_label = {
+        "verb": "verb_success", "color": "color_success",
+        "shape": "shape_success", "size": "size_success", "spatial": "spatial_success",
+    }
+    _SPATIAL_EXPS = {"verb_spatial", "color_spatial", "spatial_size", "spatial_object"}
+    _FIRST_RUN_MAP = {
+        "verb_spatial": "verb", "color_spatial": "color",
+        "spatial_size": "spatial", "spatial_object": "spatial",
+    }
+    _run_type_pairs = {
+        "verb_color": ("verb", "color"),   "verb_object": ("verb", "shape"),
+        "verb_size":  ("verb", "size"),    "verb_spatial": ("verb", "spatial"),
+        "color_object": ("color", "shape"), "size_object": ("size", "shape"),
+        "color_size": ("color", "size"),   "color_spatial": ("color", "spatial"),
+        "spatial_size": ("spatial", "size"), "spatial_object": ("spatial", "shape"),
+    }
+    _f1_label_map = {
+        "verb_color": "verb_success",    "verb_object": "verb_success",
+        "verb_size":  "verb_success",    "verb_spatial": "verb_success",
+        "color_object": "color_success", "size_object": "size_success",
+        "color_size": "color_success",   "color_spatial": "color_success",
+        "spatial_size": "spatial_success", "spatial_object": "spatial_success",
+    }
+    _f2_label_map = {
+        "verb_color": "color_success",   "verb_object": "shape_success",
+        "verb_size":  "size_success",    "verb_spatial": "spatial_success",
+        "color_object": "shape_success", "size_object": "shape_success",
+        "color_size": "size_success",    "color_spatial": "spatial_success",
+        "spatial_size": "size_success",  "spatial_object": "shape_success",
+    }
+    first_type = _run_type_pairs.get(args.experiment, ("", ""))[0]
+
+    first_ok = 0;  first_total = 0
+    second_ok = 0; second_total = 0
+    if args.batch_skip_to > 0 and results_path.exists():
+        import re as _re
+        for line in results_path.read_text().splitlines():
+            parts = line.split()
+            if not parts or not parts[0].isdigit():
+                continue
+            if len(parts) >= 5:
+                rt = parts[3]
+                m = _re.match(r"(\d+)/(\d+)", parts[4])
+                if m:
+                    ok, den = int(m.group(1)), int(m.group(2))
+                    if rt == first_type:
+                        first_ok += ok; first_total += den
+                    else:
+                        second_ok += ok; second_total += den
+
+    # ── Loop over jobs ───────────────────────────────────────────────────────
+    for job in jobs:
+        idx = int(job["index"])
+
+        job_args = dataclasses.replace(
+            args,
+            pair_i=int(job["pair_i"]),
+            pair_j=int(job["pair_j"]),
+            run_type=str(job["run_type"]),
+            third_seed=int(job["third_seed"]),
+            num_episodes=int(job["num_episodes"]),
+            seed=int(job["seed"]),
+            experiment_name=str(job["experiment_name"]),
+        )
+
+        logging.info("[%d/%d] pair=(%d,%d) run_type=%s  experiment=%s",
+                     idx, total, job_args.pair_i, job_args.pair_j,
+                     job_args.run_type, job_args.experiment)
+
+        env, instruction = _build_env_and_instruction(job_args)
+        logging.info("OOD instruction: %r", instruction)
+
+        timestamp = _dt.datetime.now().strftime("%Y%m%d_%H%M%S")
+        run_name = (f"{job_args.experiment_name.strip()}_{timestamp}"
+                    if job_args.experiment_name.strip()
+                    else f"{job_args.experiment}_{job_args.pair_i}_{job_args.pair_j}_{job_args.run_type}_{timestamp}")
+        exp_dir   = pathlib.Path(job_args.experiment_root) / run_name
+        video_dir = exp_dir / "video"
+        video_dir.mkdir(parents=True, exist_ok=True)
+
+        if job_args.experiment in _SPATIAL_EXPS:
+            _anchor_i = list(SPATIAL_ANCHORS[SPATIALS[job_args.pair_i]])
+            _anchor_j = list(SPATIAL_ANCHORS[SPATIALS[job_args.pair_j]])
+            if job_args.run_type == _FIRST_RUN_MAP[job_args.experiment]:
+                _reset_opts: dict = {"num_distractors": 1, "obj_xy": _anchor_i, "distractor_xy": [_anchor_j]}
+            else:
+                _reset_opts = {"num_distractors": 1, "obj_xy": _anchor_j, "distractor_xy": [_anchor_i]}
+        elif job_args.experiment in (
+            "color_object", "verb_object", "verb_color",
+            "verb_size", "size_object", "color_size",
+        ):
+            _reset_opts = {"num_distractors": 1}
+        else:
+            _reset_opts = {}
+
+        verb_successes = 0
+        factor2_successes = 0
+
+        for ep in tqdm.tqdm(range(job_args.num_episodes), desc=f"[{idx}/{total}]"):
+            obs, _ = env.reset(seed=job_args.seed + ep, options=_reset_opts)
+            client.reset()
+
+            action_plan: collections.deque = collections.deque()
+            base_writer  = imageio.get_writer(str(video_dir / f"ep{ep:03d}.mp4"), fps=30)
+            wrist_writer = (
+                imageio.get_writer(str(video_dir / f"ep{ep:03d}_wrist.mp4"), fps=30)
+                if job_args.save_wrist_video else None
+            )
+            ep_verb_ok    = False
+            ep_factor2_ok = False
+            done = False
+
+            try:
+                while not done:
+                    img_base  = _to_hwc_uint8(obs["sensor_data"]["base_camera"]["rgb"])
+                    img_wrist = _to_hwc_uint8(obs["sensor_data"]["hand_camera"]["rgb"])
+
+                    base_writer.append_data(img_base)
+                    if wrist_writer is not None:
+                        wrist_writer.append_data(img_wrist)
+
+                    if not action_plan:
+                        state = _state8(env)
+                        chunk = _query_pi0(client, img_base, img_wrist, state, instruction)
+                        n = min(job_args.replan_steps, len(chunk))
+                        if n < 1:
+                            break
+                        action_plan.extend(chunk[:n])
+
+                    action = np.asarray(action_plan.popleft(), dtype=np.float32).ravel()[:8]
+                    obs, _reward, term, trunc, info = env.step(action)
+
+                    if _bool_info(info, "success_first_axis"):
+                        ep_verb_ok = True
+                    elif _bool_info(info, "success"):
+                        ep_verb_ok = True
+                    if _bool_info(info, "success_second_axis"):
+                        ep_factor2_ok = True
+
+                    done = bool(term or trunc) or (ep_verb_ok and ep_factor2_ok)
+
+            finally:
+                base_writer.close()
+                if wrist_writer is not None:
+                    wrist_writer.close()
+
+            if ep_verb_ok:
+                verb_successes += 1
+            if ep_factor2_ok:
+                factor2_successes += 1
+
+            logging.info("ep=%d  verb_ok=%s  factor2_ok=%s", ep, ep_verb_ok, ep_factor2_ok)
+
+        env.close()
+        import gc
+        gc.collect()
+
+        n = max(job_args.num_episodes, 1)
+        label1 = _run_type_to_label.get(job_args.run_type, f"{job_args.run_type}_success")
+        print(f"Success rate ({label1}): {verb_successes} / {n} ({100.0*verb_successes/n:.1f}%)")
+        sys.stdout.flush()
+
+        run_name_base = job_args.experiment_name.strip() or (
+            f"{job_args.experiment}_{job_args.pair_i}_{job_args.pair_j}_{job_args.run_type}")
+        with open(results_path, "a") as f:
+            f.write(f"{idx} {job_args.pair_i} {job_args.pair_j} {job_args.run_type} "
+                    f"{verb_successes}/{n} {run_name_base}\n")
+
+        summary_lines = [
+            f"experiment={job_args.experiment}",
+            f"pair=({job_args.pair_i},{job_args.pair_j})",
+            f"run_type={job_args.run_type}",
+            f"instruction={instruction!r}",
+            f"num_episodes={job_args.num_episodes}",
+            f"{label1}={verb_successes}/{n} ({100.0*verb_successes/n:.1f}%)",
+            f"server={job_args.host}:{job_args.port}",
+        ]
+        (exp_dir / "success_rate.txt").write_text("\n".join(summary_lines) + "\n", encoding="utf-8")
+        logging.info("Saved results to %s", str(exp_dir))
+
+        if job_args.run_type == first_type:
+            first_ok += verb_successes;  first_total += n
+        else:
+            second_ok += verb_successes; second_total += n
+
+    def _rate(s, n):
+        return f"{100.0*s/n:.1f}" if n > 0 else "0.0"
+
+    f1l = _f1_label_map.get(args.experiment, "first_success")
+    f2l = _f2_label_map.get(args.experiment, "second_success")
+    with open(results_path, "a") as f:
+        f.write(f"\noverall_{f1l}={first_ok}/{first_total} ({_rate(first_ok, first_total)}%)\n")
+        f.write(f"overall_{f2l}={second_ok}/{second_total} ({_rate(second_ok, second_total)}%)\n")
+
+    print(f"\noverall_{f1l}={first_ok}/{first_total} ({_rate(first_ok, first_total)}%)")
+    print(f"overall_{f2l}={second_ok}/{second_total} ({_rate(second_ok, second_total)}%)")
+    print(f"\nSaved summary to {results_path}")
+    print(f"Done: {total} runs for {args.experiment}")
+
+
+def eval_conflict(args: Args) -> None:
+    """Single-pair debug mode (mirrors genie eval_conflict, GR00T policy)."""
+    import json
+    import tempfile
+    job = [{
+        "index": 1, "pair_i": args.pair_i, "pair_j": args.pair_j,
+        "run_type": args.run_type, "seed": args.seed,
+        "third_seed": args.third_seed, "num_episodes": args.num_episodes,
+        "experiment_name": args.experiment_name or
+        f"{args.experiment}_{args.pair_i}_{args.pair_j}_{args.run_type}",
+    }]
+    jf = tempfile.NamedTemporaryFile("w", suffix=".json", delete=False)
+    json.dump(job, jf); jf.close()
+    rt = tempfile.NamedTemporaryFile("w", suffix=".txt", delete=False); rt.close()
+    args.batch_jobs_file = jf.name
+    args.batch_results_txt = rt.name
+    batch_eval_conflict(args)
+    print("\n--- results file ---")
+    print(pathlib.Path(rt.name).read_text())
+
+
+def main() -> None:
+    args = tyro.cli(Args)
+    if args.batch_jobs_file:
+        batch_eval_conflict(args)
+    else:
+        eval_conflict(args)
+
+
+if __name__ == "__main__":
+    main()

code/resume_genie.sh

@@ -0,0 +1,106 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# Resume an OOM-truncated genie experiment WITHOUT redoing completed runs.
+# Job index is deterministic (seed 42) so a run with index K is the same job
+# regardless of which attempt produced it.  We pool every completed ood_<idx>
+# dir across all experiments_partial_*/ + current experiments/, then run ONLY
+# the missing indices in a FRESH low-memory process (won't hit the ~step280 OOM
+# ceiling because it only does the ~70-150 missing runs).
+#
+# Usage: resume_genie.sh <experiment> <gpu>
+ROOT=/workspace/groot_eval
+GENIE="${ROOT}/genie_repo/genie_envisioner"
+CONDA=/opt/miniforge3/condabin/conda
+ENV=genie_envisioner
+exp="${1:?experiment}"; gpu="${2:?gpu}"
+SEED=42; TOTAL=200; SEED_BASE=0; THIRD_SEED=42
+EXPDIR="${ROOT}/results_genie/${exp}/experiments"
+RESULTS_TXT="${ROOT}/results_genie/${exp}/genie_${exp}_ood_seed${SEED}.txt"
+LOG="${ROOT}/logs/genie/resume_${exp}.log"
+WEIGHT="${ROOT}/genie_ckpts/${exp}"; LTX="${ROOT}/LTX-Video"
+
+case "${EXPDIR}" in "${ROOT}/results_genie/"*) : ;; *) echo REFUSING; exit 2;; esac
+mkdir -p "${EXPDIR}"
+
+# 1) Consolidate: bring one dir per index from every partial backup into EXPDIR
+for bk in "${ROOT}/results_genie/${exp}"/experiments_partial_*/; do
+  [ -d "$bk" ] || continue
+  for d in "$bk"ood_*/; do
+    [ -d "$d" ] || continue
+    idx=$(basename "$d" | grep -oE '^ood_[0-9]+')
+    ls -d "${EXPDIR}/${idx}_"*/ >/dev/null 2>&1 || cp -r "$d" "${EXPDIR}/"
+  done
+done
+
+# 2) Full deterministic 400-job list
+python - "$exp" "$SEED" "$TOTAL" "$SEED_BASE" "$THIRD_SEED" > /tmp/jobs_${exp}_full.json <<'PY'
+import random, sys, json, math
+experiment=sys.argv[1]; seed=int(sys.argv[2]); n_episodes=int(sys.argv[3])
+seed_base=int(sys.argv[4]); third_seed=int(sys.argv[5])
+rng=random.Random(seed)
+def _ss(n):
+    p=[]
+    for a,b in ((0,1),(2,3),(4,5)):
+        if a<n and b<n: p+=[(a,b),(b,a)]
+    return p
+_SZ={"verb_size","size_object","color_size"}
+_SP={"verb_spatial","color_spatial","spatial_size","spatial_object"}
+if experiment in _SZ: all_pairs=_ss(6)
+elif experiment=="spatial_size": all_pairs=_ss(5)
+elif experiment in _SP: n=5; all_pairs=[(i,j) for i in range(n) for j in range(n) if i!=j]
+else: n=6; all_pairs=[(i,j) for i in range(n) for j in range(n) if i!=j]
+_rt={"verb_color":("verb","color"),"verb_object":("verb","shape"),"verb_size":("verb","size"),
+"verb_spatial":("verb","spatial"),"color_object":("color","shape"),"size_object":("size","shape"),
+"color_size":("color","size"),"color_spatial":("color","spatial"),"spatial_size":("spatial","size"),
+"spatial_object":("spatial","shape")}
+first,second=_rt[experiment]
+raw=[]
+for ep in range(n_episodes):
+    i,j=rng.choice(all_pairs); raw.append((i,j,first,ep)); raw.append((i,j,second,ep))
+total=len(raw); num_ep=math.ceil(n_episodes/total)
+jobs=[]
+for k,(i,j,rt,ep) in enumerate(raw):
+    idx=k+1; rn=f"ood_{idx:03d}_{experiment}_{i}_{j}_{rt}"
+    if experiment=="verb_object": rs=seed_base+ep; ets=ep
+    else: rs=seed_base+idx; ets=third_seed
+    jobs.append({"index":idx,"pair_i":i,"pair_j":j,"run_type":rt,"seed":rs,
+                 "third_seed":ets,"num_episodes":num_ep,"experiment_name":rn})
+print(json.dumps(jobs))
+PY
+
+# 3) Filter to MISSING indices only
+python - "$exp" "$EXPDIR" > /tmp/jobs_${exp}_missing.json <<'PY'
+import sys, json, os, glob, re
+exp=sys.argv[1]; expdir=sys.argv[2]
+full=json.load(open(f"/tmp/jobs_{exp}_full.json"))
+done=set()
+for d in glob.glob(os.path.join(expdir,"ood_*/")):
+    m=re.match(r"ood_(\d+)", os.path.basename(d.rstrip("/")))
+    if m: done.add(int(m.group(1)))
+missing=[j for j in full if j["index"] not in done]
+json.dump(missing, open(f"/tmp/jobs_{exp}_missing.json","w"))
+print(f"done={len(done)} missing={len(missing)} total={len(full)}", file=sys.stderr)
+PY
+nmiss=$(python -c "import json;print(len(json.load(open('/tmp/jobs_${exp}_missing.json'))))")
+echo "[$(date +%H:%M:%S)] ${exp}: consolidated done; MISSING=${nmiss}/400 -> resuming on gpu=${gpu}"
+if [ "${nmiss}" -eq 0 ]; then echo "${exp}: already complete (400/400)"; exit 0; fi
+
+# 4) Run ONLY missing jobs in a fresh process (hardened cpu config)
+CUDA_VISIBLE_DEVICES="${gpu}" \
+PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True,garbage_collection_threshold:0.6,max_split_size_mb:64 \
+HF_HOME="${ROOT}/.hf_cache" HF_HUB_OFFLINE=1 TRANSFORMERS_OFFLINE=1 \
+TOKENIZERS_PARALLELISM=false NO_ALBUMENTATIONS_UPDATE=1 \
+"${CONDA}" run -n "${ENV}" --no-capture-output \
+  python "${GENIE}/main.py" \
+  --experiment "${exp}" --weight "${WEIGHT}" \
+  --pretrained-model-name-or-path "${LTX}" \
+  --domain-name conflict --num-inference-steps 5 --replan-steps 5 \
+  --max-episode-steps 300 --sim-backend cpu \
+  --experiment-root "${EXPDIR}" \
+  --batch-jobs-file /tmp/jobs_${exp}_missing.json \
+  --batch-results-txt "${RESULTS_TXT}" \
+  >> "${LOG}" 2>&1
+rc=$?
+fin=$(ls -d "${EXPDIR}"/ood_*/ 2>/dev/null | grep -oE 'ood_[0-9]+' | sort -u | wc -l)
+echo "[$(date +%H:%M:%S)] ${exp}: resume rc=${rc}  total unique indices now=${fin}/400"
+exit ${rc}

code/run_af_one_ckpt.sh

@@ -0,0 +1,108 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_af_one_ckpt.sh — evaluate ONE GR00T all-factor checkpoint, 3 seeds,
+# aligned 1:1 with the pi0.5 full-factor protocol, on ONE dedicated GPU.
+#
+# Starts a dedicated GR00T zmq server pinned to <gpu> for the given checkpoint,
+# runs run_full_factor_groot.sh for seed_base ∈ {40,41,42} (sample_n=200,
+# sample_seed=42, 200 episodes, max_episode_steps=500, no_distractor_prob=0.70,
+# cpu sim/render, default difficulty), then stops the server and writes a
+# 3-seed-averaged SUMMARY.txt.
+#
+# Output is written ONLY under results_af/<ckpt>/ — it never touches results/
+# (GR00T conflict) or results_genie/.
+#
+# Usage: run_af_one_ckpt.sh <ckpt_name> <gpu> <port>
+#   ckpt_name e.g. all_factor_Lrandom_f50_n400
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+CKPT="${1:?ckpt_name (dir under gr00t_af_ckpts)}"
+GPU="${2:?gpu}"
+PORT="${3:?port}"
+SEEDS=(${SEEDS_OVERRIDE:-40 41 42})
+SAMPLE_N="${SAMPLE_N:-200}"
+TOTAL_EPISODES="${TOTAL_EPISODES:-200}"
+
+MODEL_PATH="${ROOT}/gr00t_af_ckpts/${CKPT}/checkpoint-10000"
+OUT_DIR="${ROOT}/results_af/${CKPT}"
+LOG_DIR="${ROOT}/logs/gr00t_af"
+mkdir -p "${OUT_DIR}" "${LOG_DIR}"
+slog="${LOG_DIR}/server_${CKPT}.log"
+
+if [[ ! -d "${MODEL_PATH}" ]]; then
+  echo "[${CKPT}] MODEL_PATH not found: ${MODEL_PATH}" ; exit 1
+fi
+
+echo "[$(date +%H:%M:%S)] ${CKPT}: starting GR00T server gpu=${GPU} port=${PORT}"
+( cd "${ROOT}/gr00t_repo/codebase" && CUDA_VISIBLE_DEVICES="${GPU}" \
+    HF_HOME="${ROOT}/.hf_cache" HF_TOKEN="$(cat ${ROOT}/.hf_token)" \
+    NO_ALBUMENTATIONS_UPDATE=1 TOKENIZERS_PARALLELISM=false \
+    "${ROOT}/.venv_groot/bin/python" -m gr00t.eval.run_gr00t_server \
+      --model-path "${MODEL_PATH}" \
+      --embodiment-tag new_embodiment --device cuda:0 \
+      --host 127.0.0.1 --port "${PORT}" ) > "${slog}" 2>&1 &
+spid=$!
+
+ok=0
+for _ in $(seq 1 300); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${CKPT}] SERVER DIED during load"; break; }
+  grep -q "Server ready\|Server is ready and listening" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+if [ "${ok}" != "1" ]; then
+  echo "[${CKPT}] server not ready; tail server log:"; tail -n 30 "${slog}"
+  kill "${spid}" 2>/dev/null; wait "${spid}" 2>/dev/null
+  exit 1
+fi
+echo "[$(date +%H:%M:%S)] ${CKPT}: server ready (pid ${spid})"
+
+rc_all=0
+for sb in "${SEEDS[@]}"; do
+  rt="${OUT_DIR}/full_factor_${CKPT}_seed${sb}.txt"
+  echo "[$(date +%H:%M:%S)] ${CKPT}: seed_base=${sb} → ${rt}"
+  HOST=127.0.0.1 PORT="${PORT}" SEED_BASE="${sb}" SAMPLE_SEED=42 \
+  SIM_BACKEND="${SIM_BACKEND:-gpu}" RENDER_BACKEND="${RENDER_BACKEND:-gpu}" MAX_EPISODE_STEPS=500 \
+  NO_DISTRACTOR_PROB=0.70 REPLAN_STEPS=5 \
+  CUDA_VISIBLE_DEVICES="${GPU}" \
+  VIDEO_ROOT="${OUT_DIR}/videos_seed${sb}" \
+  MS_PY="${ROOT}/.venv_ms/bin/python" \
+  bash "${HARNESS}/run_full_factor_groot.sh" "${TOTAL_EPISODES}" "${rt}" "${SAMPLE_N}" \
+    > "${LOG_DIR}/client_${CKPT}_seed${sb}.log" 2>&1
+  src=$?
+  [[ "${src}" -ne 0 ]] && rc_all=1
+  grep -E "^overall_success" "${rt}" 2>/dev/null || echo "[${CKPT} seed${sb}] no overall_ line"
+done
+
+kill "${spid}" 2>/dev/null; wait "${spid}" 2>/dev/null
+
+# ── 3-seed average ──
+python3 - "${OUT_DIR}" "${CKPT}" "${SEEDS[@]}" > "${OUT_DIR}/SUMMARY.txt" <<'PY'
+import re, sys
+from pathlib import Path
+out_dir, ckpt, *seeds = sys.argv[1:]
+rates, line = [], []
+for sb in seeds:
+    p = Path(out_dir) / f"full_factor_{ckpt}_seed{sb}.txt"
+    if not p.exists():
+        line.append(f"seed{sb}: MISSING"); continue
+    m = re.search(r"overall_success=(\d+)/(\d+) \(([\d.]+)%\)", p.read_text())
+    if m:
+        s, n, r = int(m.group(1)), int(m.group(2)), float(m.group(3))
+        rates.append(r)
+        line.append(f"seed{sb}: {s}/{n} ({r:.1f}%)")
+    else:
+        line.append(f"seed{sb}: PARSE_FAILED")
+avg = sum(rates) / len(rates) if rates else 0.0
+print(f"# {ckpt} — full-factor, TASK-aligned (sample_n=200 seed=42, 200 eps, "
+      f"max_steps=500, no_distractor=0.70, default difficulty); "
+      f"sim_backend=gpu → 任务/采样/prompt/success 与 pi0.5 一致,但数值非 1:1 可比")
+for l in line:
+    print(l)
+print(f"AVG over {len(rates)} seed(s): {avg:.1f}%")
+PY
+
+echo "[$(date +%H:%M:%S)] ${CKPT}: DONE rc=${rc_all}"
+cat "${OUT_DIR}/SUMMARY.txt"
+exit ${rc_all}

code/run_af_one_ckpt_fast.sh

@@ -0,0 +1,104 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_af_one_ckpt_fast.sh — method-A (single-process batch) version of
+# run_af_one_ckpt.sh.  Starts ONE GR00T server for the ckpt, then runs
+# groot_full_factor_batch.py once per seed (200 cells in one process → no
+# per-cell cold start).  Output identical layout, only faster.
+#
+# Usage: run_af_one_ckpt_fast.sh <ckpt_name> <gpu> <port>
+# Env: SEEDS_OVERRIDE ("40" / "40 41 42"), SAMPLE_N (200), TOTAL_EPISODES (200),
+#      SIM_BACKEND (gpu), RENDER_BACKEND (gpu)
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+CKPT="${1:?ckpt_name}"
+GPU="${2:?gpu}"
+PORT="${3:?port}"
+SEEDS=(${SEEDS_OVERRIDE:-40 41 42})
+SAMPLE_N="${SAMPLE_N:-200}"
+TOTAL_EPISODES="${TOTAL_EPISODES:-200}"
+SIM_BACKEND="${SIM_BACKEND:-gpu}"
+RENDER_BACKEND="${RENDER_BACKEND:-gpu}"
+NO_DISTRACTOR_PROB="${NO_DISTRACTOR_PROB:-0.70}"
+
+# Auto-detect ckpt layout: either <ckpt>/checkpoint-10000/ or <ckpt>/ directly (HF layout)
+MODEL_PATH="${ROOT}/gr00t_af_ckpts/${CKPT}/checkpoint-10000"
+[[ -d "${MODEL_PATH}" ]] || MODEL_PATH="${ROOT}/gr00t_af_ckpts/${CKPT}"
+OUT_DIR="${RESULTS_ROOT:-${ROOT}/results_af}/${CKPT}"
+LOG_DIR="${ROOT}/logs/gr00t_af"
+mkdir -p "${OUT_DIR}" "${LOG_DIR}"
+slog="${LOG_DIR}/server_fast_${CKPT}_gpu${GPU}.log"
+[[ -d "${MODEL_PATH}" ]] || { echo "[${CKPT}] MODEL_PATH not found: ${MODEL_PATH}"; exit 1; }
+ls "${MODEL_PATH}"/*.safetensors >/dev/null 2>&1 || { echo "[${CKPT}] no safetensors in ${MODEL_PATH}"; exit 1; }
+
+echo "[$(date +%H:%M:%S)] ${CKPT}: start GR00T server gpu=${GPU} port=${PORT}"
+( cd "${ROOT}/gr00t_repo/codebase" && CUDA_VISIBLE_DEVICES="${GPU}" \
+    HF_HOME="${ROOT}/.hf_cache" HF_TOKEN="$(cat ${ROOT}/.hf_token)" \
+    NO_ALBUMENTATIONS_UPDATE=1 TOKENIZERS_PARALLELISM=false \
+    "${ROOT}/.venv_groot/bin/python" -m gr00t.eval.run_gr00t_server \
+      --model-path "${MODEL_PATH}" \
+      --embodiment-tag new_embodiment --device cuda:0 \
+      --host 127.0.0.1 --port "${PORT}" ) > "${slog}" 2>&1 &
+spid=$!
+
+ok=0
+for _ in $(seq 1 300); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${CKPT}] SERVER DIED during load"; break; }
+  grep -q "Server ready\|Server is ready" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+[[ "${ok}" == "1" ]] || { echo "[${CKPT}] server not ready"; tail -n 25 "${slog}"; kill "${spid}" 2>/dev/null; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: server ready (pid ${spid})"
+
+_MS_TORCH_LIB="$("${ROOT}/.venv_ms/bin/python" -c 'import torch,os;print(os.path.join(os.path.dirname(torch.__file__),"lib"))' 2>/dev/null || true)"
+export LD_LIBRARY_PATH="${_MS_TORCH_LIB}:${LD_LIBRARY_PATH:-}"
+export MANISKILL_CONFLICT_ROOT="${MANISKILL_CONFLICT_ROOT:-${ROOT}/genie_repo/maniskill_conflict}"
+
+rc_all=0
+for sb in "${SEEDS[@]}"; do
+  rt="${OUT_DIR}/full_factor_${CKPT}_seed${sb}.txt"
+  echo "[$(date +%H:%M:%S)] ${CKPT}: seed_base=${sb} → ${rt}"
+  CUDA_VISIBLE_DEVICES="${GPU}" \
+  "${ROOT}/.venv_ms/bin/python" "${HARNESS}/groot_full_factor_batch.py" \
+    --host 127.0.0.1 --port "${PORT}" \
+    --results-txt "${rt}" \
+    --video-root "${OUT_DIR}/videos_seed${sb}" \
+    --sample-n "${SAMPLE_N}" --sample-seed 42 --seed-base "${sb}" \
+    --total-episodes "${TOTAL_EPISODES}" --max-episode-steps 500 \
+    --no-distractor-prob "${NO_DISTRACTOR_PROB}" --replan-steps 5 \
+    --sim-backend "${SIM_BACKEND}" --render-backend "${RENDER_BACKEND}" \
+    > "${LOG_DIR}/client_fast_${CKPT}_seed${sb}.log" 2>&1
+  src=$?
+  [[ "${src}" -ne 0 ]] && rc_all=1
+  grep -E "^overall_success" "${rt}" 2>/dev/null || echo "[${CKPT} seed${sb}] no overall_ line"
+done
+
+kill "${spid}" 2>/dev/null; wait "${spid}" 2>/dev/null
+
+python3 - "${OUT_DIR}" "${CKPT}" "${SEEDS[@]}" > "${OUT_DIR}/SUMMARY.txt" <<'PY'
+import re, sys
+from pathlib import Path
+out_dir, ckpt, *seeds = sys.argv[1:]
+rates, line = [], []
+for sb in seeds:
+    p = Path(out_dir) / f"full_factor_{ckpt}_seed{sb}.txt"
+    if not p.exists():
+        line.append(f"seed{sb}: MISSING"); continue
+    m = re.search(r"overall_success=(\d+)/(\d+) \(([\d.]+)%\)", p.read_text())
+    if m:
+        s, n, r = int(m.group(1)), int(m.group(2)), float(m.group(3))
+        rates.append(r); line.append(f"seed{sb}: {s}/{n} ({r:.1f}%)")
+    else:
+        line.append(f"seed{sb}: PARSE_FAILED")
+avg = sum(rates)/len(rates) if rates else 0.0
+print(f"# {ckpt} — full-factor TASK-aligned, single-process batch "
+      f"(sample_n=200 seed=42, 200 eps, max_steps=500, no_distractor=0.70, "
+      f"default difficulty, sim=gpu)")
+for l in line: print(l)
+print(f"AVG over {len(rates)} seed(s): {avg:.1f}%")
+PY
+
+echo "[$(date +%H:%M:%S)] ${CKPT}: DONE rc=${rc_all}"
+cat "${OUT_DIR}/SUMMARY.txt"
+exit ${rc_all}

code/run_all_groot.sh

@@ -0,0 +1,135 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_all_groot.sh — evaluate ALL 10 GR00T conflict checkpoints.
+#
+# One checkpoint per GPU: for each category we start a dedicated GR00T zmq
+# inference server (gr00t venv) pinned to one GPU, then run the ManiSkill OOD
+# sweep (ms venv) on the SAME GPU for GPU physics+render, then stop the server.
+# Up to NUM_GPUS checkpoints run concurrently (processed in waves).
+#
+# Each checkpoint is evaluated on its OWN conflict experiment (category name).
+# Every episode video is saved under
+#   ${OUT_ROOT}/<category>/experiments/<run_name>/video/ep000{,_wrist}.mp4
+#
+# Usage:
+#   bash run_all_groot.sh [seed] [total_episodes]
+# Defaults: seed=42  total_episodes=200  (→ 400 runs per checkpoint)
+# Override episode count: bash run_all_groot.sh 42 20
+# ─────────────────────────────────────────────────────────────────────────────
+
+SEED="${1:-42}"
+TOTAL_EPISODES="${2:-200}"
+
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+CKPT_ROOT="${ROOT}/checkpoints"
+OUT_ROOT="${OUT_ROOT:-${ROOT}/results}"
+LOG_DIR="${ROOT}/logs/run_$(date +%Y%m%d_%H%M%S)"
+GROOT_VENV_PY="${ROOT}/gr00t_repo/codebase/.venv/bin/python"
+GROOT_CODE="${ROOT}/gr00t_repo/codebase"
+MS_PY="${ROOT}/.venv_ms/bin/python"
+NUM_GPUS="${NUM_GPUS:-8}"
+SIM_BACKEND="${SIM_BACKEND:-gpu}"
+BASE_PORT="${BASE_PORT:-5600}"
+
+CATEGORIES=(
+  color_object color_size color_spatial size_object spatial_object
+  spatial_size verb_color verb_object verb_size verb_spatial
+)
+
+mkdir -p "${LOG_DIR}" "${OUT_ROOT}"
+echo "seed=${SEED} total_episodes=${TOTAL_EPISODES} sim_backend=${SIM_BACKEND}"
+echo "logs: ${LOG_DIR}"
+echo "results/videos: ${OUT_ROOT}"
+
+run_one() {
+  local cat="$1" gpu="$2" port="$3"
+  local ckpt="${CKPT_ROOT}/${cat}"
+  local slog="${LOG_DIR}/server_${cat}.log"
+  local clog="${LOG_DIR}/client_${cat}.log"
+  local exp_root="${OUT_ROOT}/${cat}/experiments"
+  local results_txt="${OUT_ROOT}/${cat}/${cat}_ood_seed${SEED}.txt"
+  mkdir -p "${exp_root}" "$(dirname "${results_txt}")"
+
+  echo "[$(date +%H:%M:%S)] START ${cat}  gpu=${gpu} port=${port}"
+
+  # ── Start GR00T inference server on this GPU ──
+  ( cd "${GROOT_CODE}" && CUDA_VISIBLE_DEVICES="${gpu}" \
+      HF_HOME="${ROOT}/.hf_cache" HF_TOKEN="$(cat ${ROOT}/.hf_token)" \
+      NO_ALBUMENTATIONS_UPDATE=1 TOKENIZERS_PARALLELISM=false \
+      "${GROOT_VENV_PY}" -m gr00t.eval.run_gr00t_server \
+        --model-path "${ckpt}" \
+        --embodiment-tag new_embodiment \
+        --device cuda:0 \
+        --host 127.0.0.1 \
+        --port "${port}" ) > "${slog}" 2>&1 &
+  local server_pid=$!
+
+  # ── Wait until the server is listening (model load can take ~1-2 min) ──
+  local ready=0
+  for _ in $(seq 1 180); do
+    if ! kill -0 "${server_pid}" 2>/dev/null; then
+      echo "[${cat}] SERVER DIED during startup — see ${slog}"
+      break
+    fi
+    if grep -q "Server ready\|Server is ready and listening" "${slog}" 2>/dev/null; then
+      ready=1; break
+    fi
+    sleep 3
+  done
+  if [[ "${ready}" -ne 1 ]]; then
+    echo "[${cat}] server not ready; killing. tail server log:"
+    tail -n 20 "${slog}" || true
+    kill "${server_pid}" 2>/dev/null || true
+    wait "${server_pid}" 2>/dev/null || true
+    return 1
+  fi
+  echo "[$(date +%H:%M:%S)] [${cat}] server ready (pid ${server_pid}), starting sweep"
+
+  # ── Run the OOD sweep on the SAME GPU (ManiSkill GPU physics+render) ──
+  CUDA_VISIBLE_DEVICES="${gpu}" \
+  HOST=127.0.0.1 PORT="${port}" SIM_BACKEND="${SIM_BACKEND}" \
+  EXPERIMENT_ROOT="${exp_root}" GROOT_MAIN="${HARNESS}/groot_main.py" MS_PY="${MS_PY}" \
+  bash "${HARNESS}/run_ood_groot_inference.sh" \
+      "${cat}" "${SEED}" "${TOTAL_EPISODES}" "${results_txt}" \
+      > "${clog}" 2>&1
+  local rc=$?
+
+  # ── Stop the server ──
+  kill "${server_pid}" 2>/dev/null || true
+  wait "${server_pid}" 2>/dev/null || true
+  echo "[$(date +%H:%M:%S)] DONE ${cat} rc=${rc}  results=${results_txt}"
+  return ${rc}
+}
+
+# ── Process in waves of NUM_GPUS (one checkpoint per GPU) ──
+launched=0
+pids=()
+for idx in "${!CATEGORIES[@]}"; do
+  cat="${CATEGORIES[$idx]}"
+  gpu=$(( launched % NUM_GPUS ))
+  port=$(( BASE_PORT + gpu ))
+  run_one "${cat}" "${gpu}" "${port}" &
+  pids+=($!)
+  launched=$(( launched + 1 ))
+  if (( launched % NUM_GPUS == 0 )); then
+    echo "[$(date +%H:%M:%S)] --- waiting for wave to finish ---"
+    for p in "${pids[@]}"; do wait "${p}" || true; done
+    pids=()
+  fi
+done
+for p in "${pids[@]}"; do wait "${p}" || true; done
+
+echo
+echo "================ ALL DONE ================"
+for cat in "${CATEGORIES[@]}"; do
+  rt="${OUT_ROOT}/${cat}/${cat}_ood_seed${SEED}.txt"
+  if [[ -f "${rt}" ]]; then
+    echo "### ${cat}"
+    grep -E "^overall_" "${rt}" 2>/dev/null || echo "  (no overall_ line — check logs)"
+  else
+    echo "### ${cat}: MISSING ${rt}"
+  fi
+done
+echo "Videos under: ${OUT_ROOT}/<category>/experiments/<run_name>/video/"

code/run_full_factor_groot.sh

@@ -0,0 +1,179 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_full_factor_groot.sh — GR00T N1.7 full-factor sweep.
+#
+# Cell sampling, instruction format, header and the per-cell / overall result
+# format are COPIED VERBATIM from
+#   eval_pi0_5/examples/maniskill_full_factor/run_full_factor_inference.sh
+# so the produced full_factor_<ckpt>_seed<NN>.txt files are directly comparable
+# to the pi0.5 ones.  Only the per-cell python call drives a GR00T zmq server
+# (groot_full_factor_main.py) instead of the openpi websocket policy.
+#
+# A GR00T inference server must already be reachable at ${HOST}:${PORT}.
+#
+# Usage:
+#   bash run_full_factor_groot.sh [total_episodes] [results_txt_path] [sample_n]
+#
+# Env overrides: HOST PORT SIM_BACKEND RENDER_BACKEND MAX_EPISODE_STEPS
+#                SEED_BASE NO_DISTRACTOR_PROB SAMPLE_SEED REPLAN_STEPS
+#                MS_PY GROOT_FF_MAIN VIDEO_ROOT MANISKILL_CONFLICT_ROOT
+# pi0.5 protocol values: sample_n=200 sample_seed=42 total_episodes=200
+#                        max_episode_steps=500 no_distractor_prob=0.70
+#                        sim/render=cpu  seed_base ∈ {40,41,42}
+# ─────────────────────────────────────────────────────────────────────────────
+
+TOTAL_EPISODES_TARGET="${1:-200}"
+RESULTS_TXT_PATH="${2:-}"
+SAMPLE_N="${3:-200}"
+
+HOST="${HOST:-127.0.0.1}"
+PORT="${PORT:-5555}"
+SIM_BACKEND="${SIM_BACKEND:-cpu}"
+RENDER_BACKEND="${RENDER_BACKEND:-cpu}"
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS:-500}"
+SEED_BASE="${SEED_BASE:-40}"
+NO_DISTRACTOR_PROB="${NO_DISTRACTOR_PROB:-0.70}"
+SAMPLE_SEED="${SAMPLE_SEED:-42}"
+REPLAN_STEPS="${REPLAN_STEPS:-5}"
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+MS_PY="${MS_PY:-/workspace/groot_eval/.venv_ms/bin/python}"
+GROOT_FF_MAIN="${GROOT_FF_MAIN:-${SCRIPT_DIR}/groot_full_factor_main.py}"
+
+if [[ -z "${RESULTS_TXT_PATH}" ]]; then
+  ts="$(date +%Y%m%d_%H%M%S)"
+  RESULTS_TXT_PATH="/workspace/groot_eval/results_af/full_factor_ep${TOTAL_EPISODES_TARGET}_sample${SAMPLE_N}_${ts}.txt"
+fi
+mkdir -p "$(dirname "${RESULTS_TXT_PATH}")"
+VIDEO_ROOT="${VIDEO_ROOT:-$(dirname "${RESULTS_TXT_PATH}")/videos_seed${SEED_BASE}}"
+
+# ── Generate task list: all 4320 or a random subset of sample_n ──
+# (VERBATIM from pi0.5 run_full_factor_inference.sh)
+mapfile -t TASK_LINES < <(python3 - "${SAMPLE_N}" "${SAMPLE_SEED}" <<'PY'
+import itertools, random, sys
+
+VERBS    = ["lift","grasp","push","pull","rotate","slide"]
+COLORS   = ["red","yellow","blue","orange","green","black"]
+SHAPES   = ["cube","sphere","cup","car","pyramid","star"]
+SPATIALS = ["left","right","middle","front","behind"]
+SIZES    = ["small","large","smaller","larger"]
+
+all_tasks = list(itertools.product(VERBS, COLORS, SHAPES, SPATIALS, SIZES))
+n    = int(sys.argv[1])
+seed = int(sys.argv[2])
+if n > 0:
+    rng = random.Random(seed)
+    rng.shuffle(all_tasks)
+    all_tasks = all_tasks[:n]
+for t in all_tasks:
+    print(" ".join(t))
+PY
+)
+
+TOTAL_CELLS="${#TASK_LINES[@]}"
+NUM_EPISODES="$(python3 -c "import math; print(math.ceil(${TOTAL_EPISODES_TARGET} / ${TOTAL_CELLS}))")"
+echo "sample_n=${SAMPLE_N} → ${TOTAL_CELLS} cells, ${NUM_EPISODES} episodes/cell (target ${TOTAL_EPISODES_TARGET} total)"
+
+{
+  echo "# Full-factor inference (GR00T N1.7)"
+  echo "sample_n=${SAMPLE_N}  sample_seed=${SAMPLE_SEED}  total_cells=${TOTAL_CELLS}"
+  echo "total_episodes_target=${TOTAL_EPISODES_TARGET}  num_episodes_per_cell=${NUM_EPISODES}"
+  echo "total_episodes_actual=$((TOTAL_CELLS * NUM_EPISODES))"
+  echo "host=${HOST}  port=${PORT}"
+  echo "sim_backend=${SIM_BACKEND}  render_backend=${RENDER_BACKEND}"
+  echo "max_episode_steps=${MAX_EPISODE_STEPS}  seed_base=${SEED_BASE}"
+  echo "no_distractor_prob=${NO_DISTRACTOR_PROB}  replan_steps=${REPLAN_STEPS}"
+  echo
+  echo "index verb color shape spatial size prompt successes/total"
+} > "${RESULTS_TXT_PATH}"
+
+# ManiSkill C-extensions link libtorch.so — make ms-venv torch libs findable.
+_MS_TORCH_LIB="$("${MS_PY}" -c 'import torch,os;print(os.path.join(os.path.dirname(torch.__file__),"lib"))' 2>/dev/null || true)"
+export LD_LIBRARY_PATH="${_MS_TORCH_LIB}:${LD_LIBRARY_PATH:-}"
+export MANISKILL_CONFLICT_ROOT="${MANISKILL_CONFLICT_ROOT:-/workspace/groot_eval/genie_repo/maniskill_conflict}"
+
+total_success=0
+total_episodes_done=0
+i=0
+
+for line in "${TASK_LINES[@]}"; do
+  read -r verb color shape spatial size <<<"${line}"
+  i=$((i + 1))
+
+  verb_cap="${verb^}"
+  case "${spatial}" in
+    left)   phrase="on the left"   ;;
+    right)  phrase="on the right"  ;;
+    middle) phrase="in the middle" ;;
+    front)  phrase="in front"      ;;
+    behind) phrase="at the back"   ;;
+  esac
+  prompt="${verb_cap} the ${size} ${color} ${shape} ${phrase}."
+  seed=$((SEED_BASE + i))
+
+  echo "[${i}/${TOTAL_CELLS}] ${prompt}"
+
+  run_log="$(mktemp)"
+  "${MS_PY}" "${GROOT_FF_MAIN}" \
+    --host "${HOST}" \
+    --port "${PORT}" \
+    --verb "${verb}" \
+    --color "${color}" \
+    --shape "${shape}" \
+    --spatial "${spatial}" \
+    --size "${size}" \
+    --num-episodes "${NUM_EPISODES}" \
+    --max-episode-steps "${MAX_EPISODE_STEPS}" \
+    --sim-backend "${SIM_BACKEND}" \
+    --render-backend "${RENDER_BACKEND}" \
+    --replan-steps "${REPLAN_STEPS}" \
+    --no-distractor-prob "${NO_DISTRACTOR_PROB}" \
+    --seed "${seed}" \
+    --video-out-path "${VIDEO_ROOT}/${verb}_${size}_${color}_${shape}_${spatial}" \
+    2>&1 | tee "${run_log}"
+  py_status="${PIPESTATUS[0]}"
+
+  if [[ "${py_status}" -ne 0 ]]; then
+    echo "${i} ${verb} ${color} ${shape} ${spatial} ${size} ERROR" >> "${RESULTS_TXT_PATH}"
+    rm -f "${run_log}"
+    echo "Task ${i}/${TOTAL_CELLS} failed. Partial results: ${RESULTS_TXT_PATH}"
+    exit "${py_status}"
+  fi
+
+  success_info="$(python3 - "${run_log}" <<'PY'
+import re, sys
+from pathlib import Path
+txt = Path(sys.argv[1]).read_text(encoding="utf-8", errors="ignore")
+m = re.search(r"Success rate:\s*(\d+)\s*/\s*(\d+)", txt)
+print(f"{m.group(1)} {m.group(2)}" if m else "NA NA")
+PY
+)"
+  rm -f "${run_log}"
+
+  read -r ep_succ ep_total <<<"${success_info}"
+  if [[ "${ep_succ}" == "NA" ]]; then
+    cell_result="NA"
+  else
+    total_success=$((total_success + ep_succ))
+    total_episodes_done=$((total_episodes_done + ep_total))
+    cell_result="${ep_succ}/${ep_total}"
+  fi
+
+  echo "${i} ${verb} ${color} ${shape} ${spatial} ${size} \"${prompt}\" ${cell_result}" >> "${RESULTS_TXT_PATH}"
+done
+
+overall_rate="$(python3 -c "
+s, n = ${total_success}, ${total_episodes_done}
+print(f'{100.0*s/n:.1f}' if n > 0 else '0.0')
+")"
+
+{
+  echo
+  echo "overall_success=${total_success}/${total_episodes_done} (${overall_rate}%)"
+} >> "${RESULTS_TXT_PATH}"
+
+echo ""
+echo "Done: ${total_episodes_done} episodes across ${TOTAL_CELLS} cells"
+echo "Overall: ${total_success}/${total_episodes_done} (${overall_rate}%)"
+echo "Results: ${RESULTS_TXT_PATH}"

code/run_groot_one_ckpt.sh

@@ -0,0 +1,127 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_groot_one_ckpt.sh — evaluate ONE GR00T ckpt on the HARD pairwise grid
+# (color_size or color_spatial), ONE GPU, ONE or more seeds.  Mirrors
+# run_pi0_one_ckpt.sh structure; policy=GR00T zmq server (gr00t venv).
+#
+# Speed-matched to pi0:  MAX_EPISODE_STEPS=150, REPLAN_STEPS=10, gpu-sim.
+# Output ONLY under results_gr00t_pair_color_size/<ckpt>/ (never touches
+# canonical results/ or genie's results_af/).
+#
+# Usage:  run_groot_one_ckpt.sh <ckpt_name> <gpu> <port>
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+SIM_ROOT=/workspace/eval_simulation/simulation
+CKPT="${1:?ckpt_name (dir under /workspace/gr00t_pair_ckpt)}"
+GPU="${2:?gpu}"
+PORT="${3:?port}"
+SEEDS=(${SEEDS_OVERRIDE:-42 40 41})
+TASK_DIFFICULTY="${TASK_DIFFICULTY:-1.5}"
+SIM_BACKEND="${SIM_BACKEND:-gpu}"
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS:-150}"
+REPLAN_STEPS="${REPLAN_STEPS:-10}"
+TARGET_EPISODES="${TARGET_EPISODES:-200}"
+COLOR_SIZE_MODE="${COLOR_SIZE_MODE:-hard}"
+
+MODEL_DIR="/workspace/gr00t_pair_ckpt/${CKPT}"
+# OUT_DIR depends on parsed EXPERIMENT (set just below);  we re-set it after parsing.
+LOG_DIR="${ROOT}/logs/gr00t_pair"
+[[ -d "${MODEL_DIR}" ]] || { echo "[${CKPT}] MODEL_DIR not found: ${MODEL_DIR}"; exit 1; }
+
+# experiment from ckpt name (longest matching pair prefix)
+EXPERIMENT="$(python3 - "${CKPT}" <<'PY'
+import sys
+name = sys.argv[1]
+exps = ["color_spatial","color_size","color_object","spatial_size","spatial_object",
+        "size_object","verb_spatial","verb_object","verb_color","verb_size"]
+hit = [e for e in exps if name.startswith(e)]
+print(max(hit, key=len) if hit else "")
+PY
+)"
+[[ -n "${EXPERIMENT}" ]] || { echo "[${CKPT}] cannot parse experiment from name"; exit 1; }
+# OUT_DIR: experiment-aware; color_size_easy goes to a separate dir so HARD/EASY don't collide
+if [[ "${EXPERIMENT}" == "color_size" && "${COLOR_SIZE_MODE}" == "easy" ]]; then
+  OUT_DIR="${ROOT}/results_gr00t_pair_color_size_easy/${CKPT}"
+else
+  OUT_DIR="${ROOT}/results_gr00t_pair_${EXPERIMENT}/${CKPT}"
+fi
+mkdir -p "${OUT_DIR}" "${LOG_DIR}"
+slog="${LOG_DIR}/server_${CKPT}_gpu${GPU}.log"
+echo "[$(date +%H:%M:%S)] ${CKPT}: experiment=${EXPERIMENT} gpu=${GPU} port=${PORT} difficulty=${TASK_DIFFICULTY} sim=${SIM_BACKEND}"
+
+# start GR00T zmq inference server (uses GPU)
+GROOT_VENV_PY="${ROOT}/.venv_groot/bin/python"
+GROOT_CODE="${ROOT}/gr00t_repo/codebase"
+HF_TOKEN_FILE="${ROOT}/.hf_token"
+HF_TOKEN_ARG=""
+[[ -f "${HF_TOKEN_FILE}" ]] && HF_TOKEN_ARG="$(cat ${HF_TOKEN_FILE})"
+( cd "${GROOT_CODE}" && CUDA_VISIBLE_DEVICES="${GPU}" \
+    HF_HOME="${ROOT}/.hf_cache" HF_TOKEN="${HF_TOKEN_ARG}" \
+    NO_ALBUMENTATIONS_UPDATE=1 TOKENIZERS_PARALLELISM=false \
+    "${GROOT_VENV_PY}" -m gr00t.eval.run_gr00t_server \
+      --model-path "${MODEL_DIR}" \
+      --embodiment-tag new_embodiment \
+      --device cuda:0 \
+      --host 127.0.0.1 \
+      --port "${PORT}" ) > "${slog}" 2>&1 &
+spid=$!
+ok=0
+for _ in $(seq 1 200); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${CKPT}] SERVER DIED during load"; break; }
+  grep -qE "Server ready|Server is ready and listening|listening on" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+[[ "${ok}" == "1" ]] || { echo "[${CKPT}] server not ready"; tail -n 25 "${slog}"; kill -9 "${spid}" 2>/dev/null; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: GR00T server ready (pid ${spid})"
+
+export MANISKILL_CONFLICT_ROOT="${SIM_ROOT}"
+MS_PY="${ROOT}/.venv_ms/bin/python"
+[[ -x "${MS_PY}" ]] || MS_PY="/venv/pi05_ms/bin/python"
+_MS_TORCH_LIB="$(${MS_PY} -c 'import torch,os;print(os.path.join(os.path.dirname(torch.__file__),"lib"))' 2>/dev/null || true)"
+export LD_LIBRARY_PATH="${_MS_TORCH_LIB}:${LD_LIBRARY_PATH:-}"
+
+rc_all=0
+for sb in "${SEEDS[@]}"; do
+  rt="${OUT_DIR}/gr00t_${EXPERIMENT}_${CKPT}_seed${sb}.txt"
+  echo "[$(date +%H:%M:%S)] ${CKPT}: seed${sb} experiment=${EXPERIMENT} → ${rt}"
+  CUDA_VISIBLE_DEVICES="${GPU}" SIM_ROOT="${SIM_ROOT}" \
+  "${MS_PY}" "${HARNESS}/groot_grid_eval.py" \
+    --experiment "${EXPERIMENT}" --host 127.0.0.1 --port "${PORT}" \
+    --seed "${sb}" --sim-backend "${SIM_BACKEND}" \
+    --task-difficulty "${TASK_DIFFICULTY}" \
+    --max-episode-steps "${MAX_EPISODE_STEPS}" --replan-steps "${REPLAN_STEPS}" \
+    --target-episodes "${TARGET_EPISODES}" \
+    ${MAX_CELLS:+--max-cells "${MAX_CELLS}"} \
+    --results-txt "${rt}" \
+    > "${LOG_DIR}/client_${CKPT}_seed${sb}.log" 2>&1
+  src=$?
+  [[ "${src}" -ne 0 ]] && rc_all=1
+  grep -E "^overall_" "${rt}" 2>/dev/null || echo "[${CKPT} seed${sb}] no overall_ line (see client log)"
+done
+
+kill -9 "${spid}" 2>/dev/null; pkill -9 -P "${spid}" 2>/dev/null
+
+python3 - "${OUT_DIR}" "${EXPERIMENT}" "${CKPT}" "${SEEDS[@]}" > "${OUT_DIR}/SUMMARY.txt" <<'PY'
+import re, sys
+from pathlib import Path
+out_dir, exp, ckpt, *seeds = sys.argv[1:]
+acc = {}
+for sb in seeds:
+    p = Path(out_dir) / f"gr00t_{exp}_{ckpt}_seed{sb}.txt"
+    if not p.exists():
+        print(f"seed{sb}: MISSING"); continue
+    txt = p.read_text()
+    line = []
+    for m in re.finditer(r"^(overall_\w+)=(\d+)/(\d+) \(([\d.]+)%\)", txt, re.M):
+        k, _s, _n, r = m.group(1), m.group(2), m.group(3), float(m.group(4))
+        acc.setdefault(k, []).append(r); line.append(f"{k}={r:.1f}%")
+    print(f"seed{sb}: " + "  ".join(line) if line else f"seed{sb}: PARSE_FAILED")
+print(f"\n# {ckpt}  experiment={exp}  (gr00t HARD-def, {len(seeds)} seeds)")
+for k, v in acc.items():
+    print(f"AVG {k} = {sum(v)/len(v):.1f}%  (n={len(v)} seeds)")
+PY
+echo "[$(date +%H:%M:%S)] ${CKPT}: DONE rc=${rc_all}"
+cat "${OUT_DIR}/SUMMARY.txt"
+exit ${rc_all}

code/run_one_cat.sh

@@ -0,0 +1,59 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# Evaluate ONE GR00T checkpoint on its own conflict experiment.
+# Starts a dedicated GR00T zmq server on the given GPU, runs the OOD sweep on
+# the same GPU (ManiSkill GPU physics+render), then stops the server.
+#
+# Usage: run_one_cat.sh <category> <gpu> <port> [seed=42] [total_episodes=200]
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+cat="${1:?category}"
+gpu="${2:?gpu}"
+port="${3:?port}"
+SEED="${4:-42}"
+TOTAL="${5:-200}"
+OUT_ROOT="${OUT_ROOT:-${ROOT}/results}"
+LOG_DIR="${ROOT}/logs/tmux"
+mkdir -p "${LOG_DIR}"
+slog="${LOG_DIR}/server_${cat}.log"
+clog="${LOG_DIR}/client_${cat}.log"
+exp_root="${OUT_ROOT}/${cat}/experiments"
+results_txt="${OUT_ROOT}/${cat}/${cat}_ood_seed${SEED}.txt"
+mkdir -p "${exp_root}" "$(dirname "${results_txt}")"
+
+echo "[$(date +%H:%M:%S)] ${cat}: starting GR00T server gpu=${gpu} port=${port}"
+( cd "${ROOT}/gr00t_repo/codebase" && CUDA_VISIBLE_DEVICES="${gpu}" \
+    HF_HOME="${ROOT}/.hf_cache" HF_TOKEN="$(cat ${ROOT}/.hf_token)" \
+    NO_ALBUMENTATIONS_UPDATE=1 TOKENIZERS_PARALLELISM=false \
+    "${ROOT}/.venv_groot/bin/python" -m gr00t.eval.run_gr00t_server \
+      --model-path "${ROOT}/checkpoints/${cat}" \
+      --embodiment-tag new_embodiment --device cuda:0 \
+      --host 127.0.0.1 --port "${port}" ) > "${slog}" 2>&1 &
+spid=$!
+
+ok=0
+for _ in $(seq 1 300); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${cat}] SERVER DIED during load"; break; }
+  grep -q "Server ready\|Server is ready and listening" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+if [ "${ok}" != "1" ]; then
+  echo "[${cat}] server not ready; tail server log:"; tail -n 30 "${slog}"
+  kill "${spid}" 2>/dev/null; wait "${spid}" 2>/dev/null
+  exit 1
+fi
+echo "[$(date +%H:%M:%S)] ${cat}: server ready (pid ${spid}); running OOD sweep (seed=${SEED} total=${TOTAL})"
+
+CUDA_VISIBLE_DEVICES="${gpu}" \
+HOST=127.0.0.1 PORT="${port}" SIM_BACKEND=gpu \
+EXPERIMENT_ROOT="${exp_root}" GROOT_MAIN="${HARNESS}/groot_main.py" \
+MS_PY="${ROOT}/.venv_ms/bin/python" \
+MANISKILL_CONFLICT_ROOT="${ROOT}/genie_repo/maniskill_conflict" \
+bash "${HARNESS}/run_ood_groot_inference.sh" "${cat}" "${SEED}" "${TOTAL}" "${results_txt}" \
+  > "${clog}" 2>&1
+rc=$?
+
+kill "${spid}" 2>/dev/null; wait "${spid}" 2>/dev/null
+echo "[$(date +%H:%M:%S)] ${cat}: DONE rc=${rc}  results=${results_txt}"
+grep -E "^overall_" "${results_txt}" 2>/dev/null || echo "[${cat}] (no overall_ line — check ${clog})"
+exit ${rc}

code/run_one_genie.sh

@@ -0,0 +1,57 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# Evaluate ONE Genie-Envisioner (GE-Act) checkpoint on its conflict experiment.
+# genie main.py loads MVActor in-process (NO separate server) and runs the
+# full OOD sweep in batch mode on a single dedicated GPU.
+#
+# Usage: run_one_genie.sh <experiment> <gpu> [seed=42] [total=200] [sim_backend=cpu]
+#
+# Output isolation (MUST NOT touch GR00T's results/):
+#   results_genie/<exp>/experiments/<run>/video/ep###.mp4 (+ _wrist) + success_rate.txt
+ROOT=/workspace/groot_eval
+GENIE="${ROOT}/genie_repo/genie_envisioner"
+CONDA=/opt/miniforge3/condabin/conda
+ENV_NAME=genie_envisioner
+
+exp="${1:?experiment}"
+gpu="${2:?gpu}"
+SEED="${3:-42}"
+TOTAL="${4:-200}"
+SIM_BACKEND="${5:-cpu}"
+
+WEIGHT="${ROOT}/genie_ckpts/${exp}"
+LTX_MODEL="${ROOT}/LTX-Video"
+EXP_ROOT="${ROOT}/results_genie/${exp}/experiments"
+RESULTS_TXT="${ROOT}/results_genie/${exp}/genie_${exp}_ood_seed${SEED}.txt"
+LOG_DIR="${ROOT}/logs/genie"
+clog="${LOG_DIR}/client_${exp}.log"
+
+# ── Safety guard: never write into GR00T's results/ ──────────────────────────
+case "${EXP_ROOT}" in
+  "${ROOT}/results_genie/"*) : ;;
+  *) echo "REFUSING: EXPERIMENT_ROOT not under results_genie/ (${EXP_ROOT})"; exit 2 ;;
+esac
+[ -d "${WEIGHT}" ] || { echo "missing checkpoint dir ${WEIGHT}"; exit 1; }
+[ -f "${WEIGHT}/diffusion_pytorch_model.safetensors" ] || { echo "missing safetensors in ${WEIGHT}"; exit 1; }
+[ -d "${LTX_MODEL}/text_encoder" ] || { echo "missing LTX-Video at ${LTX_MODEL}"; exit 1; }
+
+mkdir -p "${EXP_ROOT}" "$(dirname "${RESULTS_TXT}")" "${LOG_DIR}"
+
+echo "[$(date +%H:%M:%S)] ${exp}: genie sweep gpu=${gpu} sim=${SIM_BACKEND} seed=${SEED} total=${TOTAL}"
+echo "[$(date +%H:%M:%S)] ${exp}: weight=${WEIGHT}  out=${EXP_ROOT}"
+
+CUDA_VISIBLE_DEVICES="${gpu}" \
+PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True,garbage_collection_threshold:0.6,max_split_size_mb:64 \
+HF_HOME="${ROOT}/.hf_cache" HF_HUB_OFFLINE=1 TRANSFORMERS_OFFLINE=1 \
+TOKENIZERS_PARALLELISM=false NO_ALBUMENTATIONS_UPDATE=1 \
+WEIGHT="${WEIGHT}" LTX_MODEL="${LTX_MODEL}" \
+SIM_BACKEND="${SIM_BACKEND}" EXPERIMENT_ROOT="${EXP_ROOT}" \
+"${CONDA}" run -n "${ENV_NAME}" --no-capture-output \
+  bash "${GENIE}/run_ood_experiment_inference.sh" \
+  "${exp}" "${SEED}" "${TOTAL}" "${RESULTS_TXT}" \
+  > "${clog}" 2>&1
+rc=$?
+
+echo "[$(date +%H:%M:%S)] ${exp}: DONE rc=${rc}  results=${RESULTS_TXT}"
+grep -E "^overall_|FDR" "${RESULTS_TXT}" 2>/dev/null | tail -3 || echo "[${exp}] (检查 ${clog})"
+exit ${rc}

code/run_ood_groot_inference.sh

@@ -0,0 +1,179 @@
+#!/usr/bin/env bash
+set -euo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_ood_groot_inference.sh   (GR00T N1.7 version)
+#
+# Mirrors genie_envisioner/run_ood_experiment_inference.sh EXACTLY (same job
+# generation / pair+seed sampling / results format) but the policy is a
+# fine-tuned GR00T N1.7 checkpoint served over zmq by run_gr00t_server.py.
+#
+# The GR00T inference server must already be running and reachable at
+# ${HOST}:${PORT} (run_all_groot.sh starts/stops one per checkpoint).
+#
+# Usage:
+#   bash run_ood_groot_inference.sh <experiment> [seed] [total_episodes] [results_txt_path]
+#
+# experiment:
+#   verb_color | verb_object | color_object | verb_size | verb_spatial |
+#   size_object | color_size | color_spatial | spatial_size | spatial_object
+#
+# seed (default: 42)            RNG seed for pair / third-factor / episode seeds
+# total_episodes (default: 200) #(i,j) pairs sampled; total_runs = 2 × this
+# results_txt_path              summary log (default: auto-timestamped)
+#
+# Env vars:
+#   HOST 127.0.0.1   PORT 5555            GR00T server address
+#   SIM_BACKEND gpu                       ManiSkill physics+render backend
+#   MAX_EPISODE_STEPS 300   REPLAN_STEPS 5
+#   SEED_BASE 0
+#   EXPERIMENT_ROOT data/conflict_groot/experiments   (videos saved here)
+#   GROOT_MAIN  path to groot_main.py
+#   MS_PY       python interpreter of the ManiSkill venv
+# ─────────────────────────────────────────────────────────────────────────────
+
+EXPERIMENT="${1:-}"
+SEED="${2:-42}"
+TOTAL_EPISODES_TARGET="${3:-200}"
+RESULTS_TXT_PATH="${4:-}"
+THIRD_SEED="${SEED}"
+
+if [[ -z "${EXPERIMENT}" ]]; then
+  echo "Usage: $0 <experiment> [seed] [total_episodes] [results_txt_path]"
+  exit 1
+fi
+case "${EXPERIMENT}" in
+  verb_color|verb_object|color_object|verb_size|verb_spatial|\
+  size_object|color_size|color_spatial|spatial_size|spatial_object) ;;
+  *) echo "Unsupported experiment: ${EXPERIMENT}"; exit 1 ;;
+esac
+if ! [[ "${TOTAL_EPISODES_TARGET}" =~ ^[0-9]+$ ]] || [[ "${TOTAL_EPISODES_TARGET}" -le 0 ]]; then
+  echo "total_episodes must be a positive integer"; exit 1
+fi
+
+HOST="${HOST:-127.0.0.1}"
+PORT="${PORT:-5555}"
+SIM_BACKEND="${SIM_BACKEND:-gpu}"
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS:-300}"
+REPLAN_STEPS="${REPLAN_STEPS:-5}"
+SEED_BASE="${SEED_BASE:-0}"
+EXPERIMENT_ROOT="${EXPERIMENT_ROOT:-data/conflict_groot/experiments}"
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+GROOT_MAIN="${GROOT_MAIN:-${SCRIPT_DIR}/groot_main.py}"
+MS_PY="${MS_PY:-/workspace/groot_eval/.venv_ms/bin/python}"
+
+if [[ -z "${RESULTS_TXT_PATH}" ]]; then
+  ts="$(date +%Y%m%d_%H%M%S)"
+  RESULTS_TXT_PATH="${EXPERIMENT_ROOT}/${EXPERIMENT}_ood_${ts}.txt"
+fi
+mkdir -p "$(dirname "${RESULTS_TXT_PATH}")"
+
+# ── Generate jobs JSON — VERBATIM logic from genie run_ood_experiment_inference.sh ──
+JOBS_JSON="$(mktemp --suffix=.json)"
+python3 - "${EXPERIMENT}" "${SEED}" "${TOTAL_EPISODES_TARGET}" "${SEED_BASE}" "${THIRD_SEED}" <<'PY' > "${JOBS_JSON}"
+import random, sys, json, math
+experiment  = sys.argv[1]
+seed        = int(sys.argv[2])
+n_episodes  = int(sys.argv[3])
+seed_base   = int(sys.argv[4])
+third_seed  = int(sys.argv[5])
+rng = random.Random(seed)
+def _size_swap(n_size):
+    p = []
+    for a, b in ((0,1),(2,3),(4,5)):
+        if a < n_size and b < n_size:
+            p += [(a,b),(b,a)]
+    return p
+_SIZE_EXPS    = {"verb_size", "size_object", "color_size"}
+_SPATIAL_EXPS = {"verb_spatial", "color_spatial", "spatial_size", "spatial_object"}
+if experiment in _SIZE_EXPS:
+    all_pairs = _size_swap(6)
+elif experiment == "spatial_size":
+    all_pairs = _size_swap(5)
+elif experiment in _SPATIAL_EXPS:
+    n = 5
+    all_pairs = [(i,j) for i in range(n) for j in range(n) if i != j]
+else:
+    n = 6
+    all_pairs = [(i,j) for i in range(n) for j in range(n) if i != j]
+_run_types = {
+    "verb_color":("verb","color"), "verb_object":("verb","shape"),
+    "verb_size":("verb","size"),   "verb_spatial":("verb","spatial"),
+    "color_object":("color","shape"), "size_object":("size","shape"),
+    "color_size":("color","size"), "color_spatial":("color","spatial"),
+    "spatial_size":("spatial","size"), "spatial_object":("spatial","shape"),
+}
+first, second = _run_types[experiment]
+raw_jobs = []
+for ep_idx in range(n_episodes):
+    i, j = rng.choice(all_pairs)
+    raw_jobs.append((i, j, first, ep_idx))
+    raw_jobs.append((i, j, second, ep_idx))
+total = len(raw_jobs)
+num_episodes = math.ceil(n_episodes / total)
+jobs = []
+for k, (i, j, run_type, ep_idx) in enumerate(raw_jobs):
+    idx = k + 1
+    run_name = f"ood_{idx:03d}_{experiment}_{i}_{j}_{run_type}"
+    if experiment == "verb_object":
+        run_seed = seed_base + ep_idx
+        ep_third_seed = ep_idx
+    else:
+        run_seed = seed_base + idx
+        ep_third_seed = third_seed
+    jobs.append({
+        "index": idx, "pair_i": i, "pair_j": j, "run_type": run_type,
+        "seed": run_seed, "third_seed": ep_third_seed,
+        "num_episodes": num_episodes, "experiment_name": run_name,
+    })
+print(json.dumps(jobs))
+PY
+
+read -r TOTAL NUM_EPISODES < <(python3 - "${JOBS_JSON}" <<'PY'
+import json, sys
+jobs = json.loads(open(sys.argv[1]).read())
+print(len(jobs), jobs[0]["num_episodes"] if jobs else 1)
+PY
+)
+echo "Requested ${TOTAL_EPISODES_TARGET} total episodes across ${TOTAL} runs → ${NUM_EPISODES} eps/run (actual: $((TOTAL * NUM_EPISODES)))"
+
+{
+  echo "# OOD pairwise inference summary (GR00T N1.7)"
+  echo "experiment=${EXPERIMENT}"
+  echo "seed=${SEED}"
+  echo "total_episodes_target=${TOTAL_EPISODES_TARGET}"
+  echo "num_episodes_per_run=${NUM_EPISODES}"
+  echo "total_runs=${TOTAL}"
+  echo "total_episodes_actual=$((TOTAL * NUM_EPISODES))"
+  echo "third_seed=${THIRD_SEED}"
+  echo "groot_server=${HOST}:${PORT}"
+  echo "sim_backend=${SIM_BACKEND}"
+  echo "max_episode_steps=${MAX_EPISODE_STEPS}"
+  echo "replan_steps=${REPLAN_STEPS}"
+  echo "seed_base=${SEED_BASE}"
+  echo
+  echo "index pair_i pair_j run_type success run_name"
+} > "${RESULTS_TXT_PATH}"
+
+# ManiSkill C-extensions (fast_kinematics/mplib) link libtorch.so — make the
+# ms-venv torch libs discoverable regardless of caller.
+_MS_TORCH_LIB="$("${MS_PY}" -c 'import torch,os;print(os.path.join(os.path.dirname(torch.__file__),"lib"))' 2>/dev/null || true)"
+export LD_LIBRARY_PATH="${_MS_TORCH_LIB}:${LD_LIBRARY_PATH:-}"
+export MANISKILL_CONFLICT_ROOT="${MANISKILL_CONFLICT_ROOT:-/workspace/groot_eval/genie_repo/maniskill_conflict}"
+
+"${MS_PY}" "${GROOT_MAIN}" \
+  --experiment          "${EXPERIMENT}" \
+  --host                "${HOST}" \
+  --port                "${PORT}" \
+  --replan-steps        "${REPLAN_STEPS}" \
+  --max-episode-steps   "${MAX_EPISODE_STEPS}" \
+  --sim-backend         "${SIM_BACKEND}" \
+  --experiment-root     "${EXPERIMENT_ROOT}" \
+  --batch-jobs-file     "${JOBS_JSON}" \
+  --batch-results-txt   "${RESULTS_TXT_PATH}"
+py_status=$?
+rm -f "${JOBS_JSON}"
+if [[ "${py_status}" -ne 0 ]]; then
+  echo "Batch eval failed (exit ${py_status}); partial results in ${RESULTS_TXT_PATH}"
+  exit "${py_status}"
+fi
+echo "Saved summary to ${RESULTS_TXT_PATH}"

code/run_pi0_ckpt_3seed.sh

@@ -0,0 +1,96 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_pi0_ckpt_3seed.sh — ONE pi0 ckpt, ONE GPU, GR00T conflict definition,
+# 200 episodes × seeds {40,41,42}, server loaded ONCE, 3-seed average.
+#
+#   GR00T def = run_ood_groot_inference.sh + pi0_pairwise_main.py
+#   (= groot_main.py mirror, policy=pi0 openpi websocket).  Always-distractor,
+#   dual-axis (overall_color_success / overall_size|spatial_success).
+#   sim_backend=cpu (pi0+GPU-sim → CUDA illegal access).
+#
+# Usage: run_pi0_ckpt_3seed.sh <ckpt> <gpu> [port]
+# Env: SEEDS_OVERRIDE ("42" for smoke), TOTAL_EPISODES (200)
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+SIM_ROOT=/workspace/eval_simulation/simulation
+CKPT="${1:?ckpt}"
+GPU="${2:?gpu}"
+PORT="${3:-$((8300 + GPU))}"
+STEP="${STEP:-17999}"
+SEEDS=(${SEEDS_OVERRIDE:-40 41 42})
+TOTAL_EPISODES="${TOTAL_EPISODES:-200}"
+SIM_BACKEND="${SIM_BACKEND:-cpu}"
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS:-150}"
+REPLAN_STEPS="${REPLAN_STEPS:-10}"
+
+MODEL_DIR="/workspace/pi0_ckpt/${CKPT}/${STEP}"
+OUT_DIR="${ROOT}/results_pi0_gr00tdef/${CKPT}"
+LOG_DIR="${ROOT}/logs/pi0_pairwise"
+mkdir -p "${OUT_DIR}" "${LOG_DIR}"
+slog="${LOG_DIR}/srv_${CKPT}_g${GPU}.log"
+[[ -d "${MODEL_DIR}" ]] || { echo "[${CKPT}] MODEL_DIR not found"; exit 1; }
+
+EXPERIMENT="$(python3 - "${CKPT}" <<'PY'
+import sys
+n=sys.argv[1]
+e=["spatial_object","spatial_size","color_spatial","color_object","color_size",
+   "verb_spatial","verb_object","verb_color","verb_size","size_object"]
+h=[x for x in e if n.startswith(x)]
+print(max(h,key=len) if h else "")
+PY
+)"
+[[ -n "${EXPERIMENT}" ]] || { echo "[${CKPT}] cannot parse experiment"; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: exp=${EXPERIMENT} gpu=${GPU} port=${PORT} seeds=${SEEDS[*]} ${TOTAL_EPISODES}ep cpu-sim"
+
+( cd /workspace/eval_pi0 && CUDA_VISIBLE_DEVICES="${GPU}" \
+    UV_PROJECT_ENVIRONMENT=/venv/pi0_eval uv run --no-sync scripts/serve_policy.py \
+      --port "${PORT}" policy:checkpoint \
+      --policy.config pi0_maniskill --policy.dir "${MODEL_DIR}" ) > "${slog}" 2>&1 &
+spid=$!
+ok=0
+for _ in $(seq 1 200); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${CKPT}] SERVER DIED"; break; }
+  grep -qiE "Creating server|websocket_policy_server|server listening" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+[[ "${ok}" == "1" ]] || { echo "[${CKPT}] server not ready"; tail -n 25 "${slog}"; kill -9 "${spid}" 2>/dev/null; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: pi0 server ready (pid ${spid})"
+
+rc_all=0
+for sb in "${SEEDS[@]}"; do
+  rt="${OUT_DIR}/pi0_${EXPERIMENT}_${CKPT}_seed${sb}.txt"
+  echo "[$(date +%H:%M:%S)] ${CKPT}: seed${sb} → ${rt}"
+  HOST=127.0.0.1 PORT="${PORT}" SIM_BACKEND="${SIM_BACKEND}" \
+  MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS}" REPLAN_STEPS="${REPLAN_STEPS}" SEED_BASE=0 \
+  EXPERIMENT_ROOT="${OUT_DIR}/video_seed${sb}" \
+  GROOT_MAIN="${HARNESS}/pi0_pairwise_main.py" \
+  MS_PY="/venv/pi05_ms/bin/python" \
+  MANISKILL_CONFLICT_ROOT="${SIM_ROOT}" \
+  bash "${HARNESS}/run_ood_groot_inference.sh" "${EXPERIMENT}" "${sb}" "${TOTAL_EPISODES}" "${rt}" \
+    > "${LOG_DIR}/cli_${CKPT}_s${sb}.log" 2>&1
+  [[ $? -ne 0 ]] && rc_all=1
+  grep -E "^overall_" "${rt}" 2>/dev/null || echo "[${CKPT} s${sb}] no overall_ (see cli log)"
+done
+
+kill -9 "${spid}" 2>/dev/null; pkill -9 -P "${spid}" 2>/dev/null
+
+python3 - "${OUT_DIR}" "${EXPERIMENT}" "${CKPT}" "${SEEDS[@]}" > "${OUT_DIR}/SUMMARY.txt" <<'PY'
+import re,sys
+from pathlib import Path
+out,exp,ck,*seeds=sys.argv[1:]
+acc={}
+for sb in seeds:
+    p=Path(out)/f"pi0_{exp}_{ck}_seed{sb}.txt"
+    if not p.exists(): print(f"seed{sb}: MISSING"); continue
+    t=p.read_text(); line=[]
+    for m in re.finditer(r"^(overall_\w+)=(\d+)/(\d+) \(([\d.]+)%\)",t,re.M):
+        k,r=m.group(1),float(m.group(4)); acc.setdefault(k,[]).append(r); line.append(f"{k}={r:.1f}%")
+    print(f"seed{sb}: "+("  ".join(line) if line else "PARSE_FAILED"))
+print(f"\n# {ck}  exp={exp}  GR00T-def {len(seeds)}seed×{__import__('os').environ.get('TE','200')}ep cpu-sim")
+for k,v in acc.items(): print(f"AVG {k} = {sum(v)/len(v):.1f}%  (n={len(v)})")
+PY
+echo "[$(date +%H:%M:%S)] ${CKPT}: DONE rc=${rc_all}"
+cat "${OUT_DIR}/SUMMARY.txt"
+exit ${rc_all}

code/run_pi0_one_ckpt.sh

@@ -0,0 +1,112 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_pi0_one_ckpt.sh — evaluate ONE pi0 checkpoint on the clean pairwise-OOD
+# grid, ONE dedicated GPU, 3 seeds (42/40/41) averaged.
+#
+#  • experiment is parsed from the ckpt name (longest matching pair prefix).
+#  • full grid: ALL (i,j) incl. diagonal over the experiment's two factors,
+#    third factor sampled from training_vocab pool (size 3), num_traj=1/cell.
+#  • per-pair env / instruction / dual-axis success + distractor: reused from
+#    the validated GR00T conflict logic (pi0_pairwise_main.py).
+#  • policy = pi0 openpi websocket (serve_policy.py, config pi0_maniskill).
+#  • task_difficulty default 1.3 (harder, per request).
+#  • output ONLY under results_pi0_pairwise/<ckpt>/ (never touches GR00T/genie).
+#
+# Usage: run_pi0_one_ckpt.sh <ckpt_name> <gpu> <port>
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+SIM_ROOT=/workspace/eval_simulation/simulation
+CKPT="${1:?ckpt_name (dir under /workspace/pi0_ckpt)}"
+GPU="${2:?gpu}"
+PORT="${3:?port}"
+STEP="${STEP:-17999}"
+SEEDS=(${SEEDS_OVERRIDE:-42 40 41})
+TASK_DIFFICULTY="${TASK_DIFFICULTY:-1.5}"
+SIM_BACKEND="${SIM_BACKEND:-cpu}"
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS:-150}"
+REPLAN_STEPS="${REPLAN_STEPS:-10}"
+
+MODEL_DIR="/workspace/pi0_ckpt/${CKPT}/${STEP}"
+OUT_DIR="${ROOT}/results_pi0_pairwise/${CKPT}"
+LOG_DIR="${ROOT}/logs/pi0_pairwise"
+mkdir -p "${OUT_DIR}" "${LOG_DIR}"
+slog="${LOG_DIR}/server_${CKPT}_gpu${GPU}.log"
+[[ -d "${MODEL_DIR}" ]] || { echo "[${CKPT}] MODEL_DIR not found: ${MODEL_DIR}"; exit 1; }
+
+# ── experiment from ckpt name (longest matching pair prefix) ──
+EXPERIMENT="$(python3 - "${CKPT}" <<'PY'
+import sys
+name = sys.argv[1]
+exps = ["spatial_object","spatial_size","color_spatial","color_object","color_size",
+        "verb_spatial","verb_object","verb_color","verb_size","size_object"]
+hit = [e for e in exps if name.startswith(e)]
+print(max(hit, key=len) if hit else "")
+PY
+)"
+[[ -n "${EXPERIMENT}" ]] || { echo "[${CKPT}] cannot parse experiment from name"; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: experiment=${EXPERIMENT} gpu=${GPU} port=${PORT} difficulty=${TASK_DIFFICULTY}"
+
+# ── start pi0 openpi websocket policy server ──
+( cd /workspace/eval_pi0 && CUDA_VISIBLE_DEVICES="${GPU}" \
+    UV_PROJECT_ENVIRONMENT=/venv/pi0_eval uv run --no-sync scripts/serve_policy.py \
+      --port "${PORT}" policy:checkpoint \
+      --policy.config pi0_maniskill --policy.dir "${MODEL_DIR}" ) > "${slog}" 2>&1 &
+spid=$!
+ok=0
+for _ in $(seq 1 200); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${CKPT}] SERVER DIED during load"; break; }
+  grep -qiE "Creating server|websocket_policy_server|server listening" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+[[ "${ok}" == "1" ]] || { echo "[${CKPT}] server not ready"; tail -n 25 "${slog}"; kill -9 "${spid}" 2>/dev/null; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: pi0 server ready (pid ${spid})"
+
+export MANISKILL_CONFLICT_ROOT="${SIM_ROOT}"
+_MS_TORCH_LIB="$(/venv/pi05_ms/bin/python -c 'import torch,os;print(os.path.join(os.path.dirname(torch.__file__),"lib"))' 2>/dev/null || true)"
+export LD_LIBRARY_PATH="${_MS_TORCH_LIB}:${LD_LIBRARY_PATH:-}"
+
+rc_all=0
+for sb in "${SEEDS[@]}"; do
+  rt="${OUT_DIR}/pi0_${EXPERIMENT}_${CKPT}_seed${sb}.txt"
+  echo "[$(date +%H:%M:%S)] ${CKPT}: seed${sb} experiment=${EXPERIMENT} → ${rt}"
+  CUDA_VISIBLE_DEVICES="${GPU}" SIM_ROOT="${SIM_ROOT}" \
+  /venv/pi05_ms/bin/python "${HARNESS}/pi0_grid_eval.py" \
+    --experiment "${EXPERIMENT}" --host 127.0.0.1 --port "${PORT}" \
+    --seed "${sb}" --sim-backend "${SIM_BACKEND}" \
+    --task-difficulty "${TASK_DIFFICULTY}" \
+    --max-episode-steps "${MAX_EPISODE_STEPS}" --replan-steps "${REPLAN_STEPS}" \
+    --target-episodes "${TARGET_EPISODES:-200}" \
+    ${MAX_CELLS:+--max-cells "${MAX_CELLS}"} \
+    --results-txt "${rt}" \
+    > "${LOG_DIR}/client_${CKPT}_seed${sb}.log" 2>&1
+  src=$?
+  [[ "${src}" -ne 0 ]] && rc_all=1
+  grep -E "^overall_" "${rt}" 2>/dev/null || echo "[${CKPT} seed${sb}] no overall_ line (see client log)"
+done
+
+kill -9 "${spid}" 2>/dev/null; pkill -9 -P "${spid}" 2>/dev/null
+
+python3 - "${OUT_DIR}" "${EXPERIMENT}" "${CKPT}" "${SEEDS[@]}" > "${OUT_DIR}/SUMMARY.txt" <<'PY'
+import re, sys
+from pathlib import Path
+out_dir, exp, ckpt, *seeds = sys.argv[1:]
+acc = {}
+for sb in seeds:
+    p = Path(out_dir) / f"pi0_{exp}_{ckpt}_seed{sb}.txt"
+    if not p.exists():
+        print(f"seed{sb}: MISSING"); continue
+    txt = p.read_text()
+    line = []
+    for m in re.finditer(r"^(overall_\w+)=(\d+)/(\d+) \(([\d.]+)%\)", txt, re.M):
+        k, _s, _n, r = m.group(1), m.group(2), m.group(3), float(m.group(4))
+        acc.setdefault(k, []).append(r); line.append(f"{k}={r:.1f}%")
+    print(f"seed{sb}: " + "  ".join(line) if line else f"seed{sb}: PARSE_FAILED")
+print(f"\n# {ckpt}  experiment={exp}  (pi0, pairwise-OOD full grid, task_difficulty harder, {len(seeds)} seeds)")
+for k, v in acc.items():
+    print(f"AVG {k} = {sum(v)/len(v):.1f}%  (n={len(v)} seeds)")
+PY
+echo "[$(date +%H:%M:%S)] ${CKPT}: DONE rc=${rc_all}"
+cat "${OUT_DIR}/SUMMARY.txt"
+exit ${rc_all}

code/run_pi0_queue.sh

@@ -0,0 +1,43 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# Queue ALL fully-transferred pi0 ckpts through the proven 36-cell grid eval
+# (run_pi0_one_ckpt.sh + pi0_grid_eval.py, cpu-sim, 3 seeds 40/41/42, avg).
+# ONE model per GPU; when 8 GPUs busy, the rest wait. Re-scans transfer state
+# so newly-finished ckpts get queued too. Skips ckpts already done (SUMMARY).
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+RES="${ROOT}/results_pi0_pairwise"
+LOG="${ROOT}/logs/pi0_pairwise"
+CK=/workspace/pi0_ckpt
+GPUS=(0 1 2 3 4 5 6 7)
+mkdir -p "${RES}" "${LOG}"
+log(){ echo "[$(date +%H:%M:%S)] $*" >> "${LOG}/queue.log"; }
+
+complete(){ s="$1/17999"; [[ -f "$s/model.safetensors" && -f "$s/metadata.pt" ]] || return 1
+  ls "$s"/assets/*/norm_stats.json >/dev/null 2>&1 || return 1
+  mb=$(du -m "$s/model.safetensors" 2>/dev/null|cut -f1); [[ "${mb:-0}" -ge 3000 ]] || return 1
+  [[ $(( $(date +%s) - $(stat -c %Y "$s/model.safetensors" 2>/dev/null||echo 0) )) -ge 90 ]]; }
+free_gpu(){ for g in "${GPUS[@]}"; do tmux has-session -t "pi0_g${g}" 2>/dev/null || { echo "$g"; return; }; done; echo ""; }
+
+log "queue start"
+while true; do
+  pending=0
+  for d in "${CK}"/*/; do
+    n=$(basename "$d"); case "$n" in ckpt_batch*|ckpt_may_*) continue;; esac
+    [[ -f "${RES}/${n}/SUMMARY.txt" ]] && continue
+    complete "$d" || continue
+    pending=$((pending+1))
+    tmux has-session -t "__claim_${n}" 2>/dev/null && continue   # already dispatched
+    g="$(free_gpu)"; [[ -z "$g" ]] && { log "all GPUs busy; ${n} waits"; break; }
+    p=$((8400+g))
+    log "DISPATCH ${n} → GPU${g} (pi0_g${g})"
+    tmux new-session -d -s "__claim_${n}" "sleep 999999"          # claim marker
+    tmux new-session -d -s "pi0_g${g}" \
+      "cd ${HARNESS} && bash run_pi0_one_ckpt.sh '${n}' ${g} ${p} > ${LOG}/q_${n}.log 2>&1; \
+       tmux kill-session -t __claim_${n} 2>/dev/null"
+  done
+  # done when nothing pending and no pi0_g* running
+  run=$(tmux ls 2>/dev/null | grep -c '^pi0_g[0-7]:' || true)
+  if [[ "${pending}" -eq 0 && "${run}" -eq 0 ]]; then log "ALL DONE"; break; fi
+  sleep 90
+done

code/run_pi0_seed.sh

@@ -0,0 +1,76 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# run_pi0_seed.sh — evaluate ONE pi0 ckpt for ONE seed on ONE GPU, using the
+# EXACT GR00T conflict definition (200 episodes, always-distractor, dual-axis
+# color/size or color/spatial success).  Reuses GR00T's own job-gen+batch:
+#   run_ood_groot_inference.sh  +  pi0_pairwise_main.py (= groot_main.py with
+#   the policy boundary swapped to the pi0 openpi websocket server).
+#
+# Usage: run_pi0_seed.sh <ckpt_name> <gpu> <seed> [port]
+# One (ckpt, seed) per GPU.  sim_backend=cpu (pi0 server holds the GPU; pi0 +
+# ManiSkill-GPU-sim on one card → CUDA illegal access, so physics on CPU).
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+SIM_ROOT=/workspace/eval_simulation/simulation
+CKPT="${1:?ckpt_name}"
+GPU="${2:?gpu}"
+SEED="${3:?seed}"
+PORT="${4:-$((8200 + GPU))}"
+STEP="${STEP:-17999}"
+TOTAL_EPISODES="${TOTAL_EPISODES:-200}"
+SIM_BACKEND="${SIM_BACKEND:-cpu}"
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS:-150}"
+REPLAN_STEPS="${REPLAN_STEPS:-10}"
+
+MODEL_DIR="/workspace/pi0_ckpt/${CKPT}/${STEP}"
+OUT_DIR="${ROOT}/results_pi0_pairwise/${CKPT}"
+LOG_DIR="${ROOT}/logs/pi0_pairwise"
+mkdir -p "${OUT_DIR}" "${LOG_DIR}"
+slog="${LOG_DIR}/server_${CKPT}_gpu${GPU}_s${SEED}.log"
+[[ -d "${MODEL_DIR}" ]] || { echo "[${CKPT}] MODEL_DIR not found: ${MODEL_DIR}"; exit 1; }
+
+EXPERIMENT="$(python3 - "${CKPT}" <<'PY'
+import sys
+name=sys.argv[1]
+exps=["spatial_object","spatial_size","color_spatial","color_object","color_size",
+      "verb_spatial","verb_object","verb_color","verb_size","size_object"]
+hit=[e for e in exps if name.startswith(e)]
+print(max(hit,key=len) if hit else "")
+PY
+)"
+[[ -n "${EXPERIMENT}" ]] || { echo "[${CKPT}] cannot parse experiment"; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: experiment=${EXPERIMENT} gpu=${GPU} seed=${SEED} port=${PORT} (GR00T-def, ${TOTAL_EPISODES}ep, cpu-sim)"
+
+# ── start pi0 openpi websocket server (uses the GPU) ──
+( cd /workspace/eval_pi0 && CUDA_VISIBLE_DEVICES="${GPU}" \
+    UV_PROJECT_ENVIRONMENT=/venv/pi0_eval uv run --no-sync scripts/serve_policy.py \
+      --port "${PORT}" policy:checkpoint \
+      --policy.config pi0_maniskill --policy.dir "${MODEL_DIR}" ) > "${slog}" 2>&1 &
+spid=$!
+ok=0
+for _ in $(seq 1 200); do
+  kill -0 "${spid}" 2>/dev/null || { echo "[${CKPT}] SERVER DIED"; break; }
+  grep -qiE "Creating server|websocket_policy_server|server listening" "${slog}" 2>/dev/null && { ok=1; break; }
+  sleep 3
+done
+[[ "${ok}" == "1" ]] || { echo "[${CKPT}] server not ready"; tail -n 25 "${slog}"; kill -9 "${spid}" 2>/dev/null; exit 1; }
+echo "[$(date +%H:%M:%S)] ${CKPT}: pi0 server ready (pid ${spid})"
+
+rt="${OUT_DIR}/pi0_${EXPERIMENT}_${CKPT}_seed${SEED}.txt"
+HOST=127.0.0.1 PORT="${PORT}" SIM_BACKEND="${SIM_BACKEND}" \
+MAX_EPISODE_STEPS="${MAX_EPISODE_STEPS}" REPLAN_STEPS="${REPLAN_STEPS}" \
+SEED_BASE=0 \
+EXPERIMENT_ROOT="${OUT_DIR}/video_seed${SEED}" \
+GROOT_MAIN="${HARNESS}/pi0_pairwise_main.py" \
+MS_PY="/venv/pi05_ms/bin/python" \
+MANISKILL_CONFLICT_ROOT="${SIM_ROOT}" \
+bash "${HARNESS}/run_ood_groot_inference.sh" "${EXPERIMENT}" "${SEED}" "${TOTAL_EPISODES}" "${rt}" \
+  > "${LOG_DIR}/client_${CKPT}_seed${SEED}.log" 2>&1
+rc=$?
+
+kill -9 "${spid}" 2>/dev/null; pkill -9 -P "${spid}" 2>/dev/null
+echo "[$(date +%H:%M:%S)] ${CKPT} seed${SEED}: DONE rc=${rc}"
+grep -E "^overall_" "${rt}" 2>/dev/null || echo "[${CKPT} s${SEED}] no overall_ (see client log)"
+exit ${rc}

code/run_tmux_all.sh

@@ -0,0 +1,37 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# Launch 10 tmux sessions, one per GR00T conflict checkpoint, all in parallel.
+# GPU assignment: gpu = idx % NUM_GPUS  (with 8 GPUs, gpu0 & gpu1 host 2 each).
+# Each session: GR00T server + ManiSkill OOD sweep on the same GPU.
+#
+# Usage: bash run_tmux_all.sh [seed=42] [total_episodes=200]
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+SEED="${1:-42}"
+TOTAL="${2:-200}"
+NUM_GPUS="${NUM_GPUS:-8}"
+BASE_PORT="${BASE_PORT:-5600}"
+
+CATS=(color_object color_size color_spatial size_object spatial_object
+      spatial_size verb_color verb_object verb_size verb_spatial)
+
+mkdir -p "${ROOT}/logs/tmux"
+for i in "${!CATS[@]}"; do
+  cat="${CATS[$i]}"
+  gpu=$(( i % NUM_GPUS ))
+  port=$(( BASE_PORT + i ))
+  sess="groot_${cat}"
+  tmux kill-session -t "${sess}" 2>/dev/null || true
+  tmux new-session -d -s "${sess}" \
+    "bash ${HARNESS}/run_one_cat.sh ${cat} ${gpu} ${port} ${SEED} ${TOTAL} 2>&1 | tee ${ROOT}/logs/tmux/session_${cat}.log; echo; echo '=== ${cat} session finished (rc=$?) — pane kept for inspection ==='; exec bash"
+  echo "launched tmux ${sess}  cat=${cat} gpu=${gpu} port=${port}"
+  sleep 1
+done
+echo
+echo "tmux sessions:"
+tmux ls
+echo
+echo "Attach:  tmux attach -t groot_<category>"
+echo "Logs:    ${ROOT}/logs/tmux/{server,client,session}_<category>.log"
+echo "Results: ${ROOT}/results/<category>/<category>_ood_seed${SEED}.txt"
+echo "Videos:  ${ROOT}/results/<category>/experiments/<run_name>/video/ep000{,_wrist}.mp4"

code/smoke_e2e.sh

@@ -0,0 +1,48 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# End-to-end smoke: GR00T server + ManiSkill client, tiny color_object run.
+ROOT=/workspace/groot_eval
+CAT="${1:-color_object}"
+GPU="${2:-0}"
+PORT="${3:-5599}"
+TOTAL_EP="${4:-2}"        # -> 4 runs x 1 episode
+MAXSTEPS="${5:-40}"
+OUT="${ROOT}/results_smoke/${CAT}"
+mkdir -p "${OUT}/experiments" "${ROOT}/logs"
+SLOG="${ROOT}/logs/smoke_server_${CAT}.log"
+CLOG="${ROOT}/logs/smoke_client_${CAT}.log"
+
+echo "[smoke] start GR00T server cat=${CAT} gpu=${GPU} port=${PORT}"
+( cd "${ROOT}/gr00t_repo/codebase" && CUDA_VISIBLE_DEVICES="${GPU}" \
+   HF_HOME="${ROOT}/.hf_cache" HF_TOKEN="$(cat ${ROOT}/.hf_token)" \
+   NO_ALBUMENTATIONS_UPDATE=1 TOKENIZERS_PARALLELISM=false \
+   "${ROOT}/.venv_groot/bin/python" -m gr00t.eval.run_gr00t_server \
+     --model-path "${ROOT}/checkpoints/${CAT}" \
+     --embodiment-tag new_embodiment --device cuda:0 \
+     --host 127.0.0.1 --port "${PORT}" ) > "${SLOG}" 2>&1 &
+SPID=$!
+
+ready=0
+for _ in $(seq 1 150); do
+  kill -0 "${SPID}" 2>/dev/null || { echo "[smoke] SERVER DIED"; tail -n 30 "${SLOG}"; exit 1; }
+  grep -q "Server ready\|Server is ready and listening" "${SLOG}" 2>/dev/null && { ready=1; break; }
+  sleep 3
+done
+[ "${ready}" -eq 1 ] || { echo "[smoke] server not ready"; tail -n 30 "${SLOG}"; kill "${SPID}"; exit 1; }
+echo "[smoke] server ready; running client sweep"
+
+TORCH_LIB=$("${ROOT}/.venv_ms/bin/python" -c "import torch,os;print(os.path.join(os.path.dirname(torch.__file__),'lib'))")
+CUDA_VISIBLE_DEVICES="${GPU}" LD_LIBRARY_PATH="${TORCH_LIB}:${LD_LIBRARY_PATH:-}" \
+HOST=127.0.0.1 PORT="${PORT}" SIM_BACKEND=gpu MAX_EPISODE_STEPS="${MAXSTEPS}" \
+EXPERIMENT_ROOT="${OUT}/experiments" GROOT_MAIN="${ROOT}/harness/groot_main.py" \
+MS_PY="${ROOT}/.venv_ms/bin/python" MANISKILL_CONFLICT_ROOT="${ROOT}/genie_repo/maniskill_conflict" \
+bash "${ROOT}/harness/run_ood_groot_inference.sh" "${CAT}" 42 "${TOTAL_EP}" "${OUT}/${CAT}_smoke.txt" \
+  > "${CLOG}" 2>&1
+RC=$?
+
+kill "${SPID}" 2>/dev/null || true; wait "${SPID}" 2>/dev/null || true
+echo "[smoke] client rc=${RC}"
+echo "------ results txt ------"; cat "${OUT}/${CAT}_smoke.txt" 2>/dev/null
+echo "------ videos ------"; find "${OUT}/experiments" -name '*.mp4' | head; echo "count: $(find "${OUT}/experiments" -name '*.mp4' | wc -l)"
+echo "------ client log tail ------"; tail -n 25 "${CLOG}"
+exit ${RC}

code/smoke_env.py

@@ -0,0 +1,37 @@
+"""Env-only smoke test (no GR00T server): build VerbObjectColor-v1 via
+groot_main._build_env_and_instruction for a few experiments, reset+step,
+print obs/info structure. Validates the ManiSkill side + our kwargs + GPU sim.
+"""
+import sys, numpy as np
+sys.path.insert(0, "/workspace/groot_eval/harness")
+from groot_main import _build_env_and_instruction, _state8, _to_hwc_uint8, Args, SPATIAL_ANCHORS, SPATIALS
+
+CASES = [
+    dict(experiment="color_object", pair_i=0, pair_j=1, run_type="color"),
+    dict(experiment="spatial_object", pair_i=0, pair_j=1, run_type="spatial"),
+]
+for c in CASES:
+    a = Args(sim_backend=sys.argv[1] if len(sys.argv) > 1 else "gpu",
+             max_episode_steps=40, third_seed=42, seed=42, **c)
+    print(f"\n==== {c} ====")
+    env, instr = _build_env_and_instruction(a)
+    print("instruction:", repr(instr))
+    if a.experiment in {"verb_spatial","color_spatial","spatial_size","spatial_object"}:
+        ai = list(SPATIAL_ANCHORS[SPATIALS[a.pair_i]]); aj = list(SPATIAL_ANCHORS[SPATIALS[a.pair_j]])
+        ropts = {"num_distractors": 1, "obj_xy": ai, "distractor_xy": [aj]}
+    else:
+        ropts = {"num_distractors": 1}
+    obs, _ = env.reset(seed=42, options=ropts)
+    sd = obs["sensor_data"]
+    print("sensor_data keys:", list(sd.keys()))
+    b = _to_hwc_uint8(sd["base_camera"]["rgb"]); h = _to_hwc_uint8(sd["hand_camera"]["rgb"])
+    print("base img", b.shape, b.dtype, "wrist img", h.shape, h.dtype)
+    s = _state8(env); print("state8", s.shape, s.dtype, np.round(s, 3))
+    act = np.zeros(8, np.float32)
+    for t in range(5):
+        obs, r, term, trunc, info = env.step(act)
+    print("info keys:", sorted(info.keys()))
+    for k in ("success", "success_first_axis", "success_second_axis"):
+        print(f"  {k} = {info.get(k)}")
+    env.close()
+print("\nSMOKE_ENV_OK")

code/vlm_eval.py

@@ -0,0 +1,290 @@
+"""
+VLM evaluation of all 10 conflict experiments using Gemini-2.5-Flash.
+Reads existing videos and success_rate.txt; NEVER modifies existing files.
+
+Adapted for this environment:
+  - model arg: gr00t  -> /workspace/groot_eval/results
+                genie  -> /workspace/groot_eval/results_genie
+  - layout:  <root>/<exp>/experiments/ood_<idx>_<exp>_<pi>_<pj>_<rt>_<ts>/
+  - output:  /workspace/groot_eval/results/vlm_eval/<model>/vlm_eval_<exp>.jsonl
+  - resumable: skips runs already present in the output jsonl
+Usage: python vlm_eval.py <gr00t|genie> [exp ...] [--limit N]
+"""
+
+import base64, json, re, sys, time, threading
+from concurrent.futures import ThreadPoolExecutor
+from pathlib import Path
+import requests
+
+WORKERS = 10  # concurrent VLM requests per model process (overridable via --workers)
+
+API_KEY = "sk-12YA7oNA-9gl9qn2oedejw"
+API_URL = "https://inference-api.nvidia.com/v1/chat/completions"
+MODEL   = "gcp/google/gemini-2.5-flash"
+
+ROOTS = {
+    "gr00t": Path("/workspace/groot_eval/results"),
+    "genie": Path("/workspace/groot_eval/results_genie"),
+}
+OUT_BASE = Path("/workspace/groot_eval/results/vlm_eval")
+
+VERBS   = ["lift", "grasp", "push", "pull", "rotate", "slide"]
+COLORS  = ["red", "yellow", "blue", "orange", "green", "black"]
+SHAPES  = ["cube", "sphere", "cup", "car", "pyramid", "star"]
+SIZES   = ["smallest", "second smallest", "middle-sized", "second largest", "largest", "biggest"]
+SPATIALS= ["on the left", "on the right", "in the middle", "in front", "at the back", "in the center"]
+
+ALL_EXPERIMENTS = [
+    "verb_color", "verb_object", "verb_size", "verb_spatial",
+    "color_object", "size_object", "color_size", "color_spatial",
+    "spatial_size", "spatial_object",
+]
+
+# ── Prompt builders (verbatim from user's script) ────────────────────────────
+
+def prompt_verb_experiment(instruction, pair_i, pair_j, factor2):
+    verb_i = VERBS[pair_i]; verb_j = VERBS[pair_j]
+    if factor2 == "color":
+        factor2_desc = f"color '{COLORS[pair_j]}' (which the robot was trained to interact with using '{verb_j}')"
+    elif factor2 == "shape":
+        factor2_desc = f"shape '{SHAPES[pair_j]}' (which the robot was trained to interact with using '{verb_j}')"
+    elif factor2 == "size":
+        factor2_desc = f"size '{SIZES[pair_j]}' (which the robot was trained to interact with using '{verb_j}')"
+    elif factor2 == "spatial":
+        factor2_desc = f"spatial position '{SPATIALS[pair_j]}' (which the robot was trained to interact with using '{verb_j}')"
+    else:
+        factor2_desc = f"factor {pair_j} (trained verb: '{verb_j}')"
+    return f"""Watch this robot manipulation video carefully.
+
+**Instruction given to the robot:** "{instruction}"
+
+This is a CONFLICT experiment about verb bias. The robot was trained with specific verb-attribute pairings:
+- Verb "{verb_i}" is paired with one attribute set (instruction verb)
+- Verb "{verb_j}" is paired with {factor2_desc}
+
+The robot must choose which action to perform. Observe carefully.
+
+Answer in this EXACT format:
+ROBOT_ACTION: [describe what physical action the robot arm performed in 1 sentence]
+ACTION_TYPE: ["{verb_i}" if robot performed the instructed action, "{verb_j}" if robot performed the competing action, "other" if neither]
+FACTOR_FOLLOWED: ["verb" if robot followed the instructed verb '{verb_i}', "{factor2}" if robot followed the {factor2} factor (doing '{verb_j}'), "neither" if unclear]
+CONFIDENCE: [high / medium / low]
+REASONING: [1-2 sentences explaining your judgment]"""
+
+
+def prompt_two_object_experiment(instruction, pair_i, pair_j, factor1, factor2):
+    def describe_factor(f, k):
+        if f == "color":   return f"color='{COLORS[k]}'"
+        if f == "shape":   return f"shape='{SHAPES[k]}'"
+        if f == "size":    return f"size='{SIZES[k]}'"
+        if f == "spatial": return f"position='{SPATIALS[k]}'"
+        return f"{f}={k}"
+    obj_a_desc = describe_factor(factor1, pair_i)
+    obj_b_desc = describe_factor(factor2, pair_j)
+    return f"""Watch this robot manipulation video carefully.
+
+**Instruction given to the robot:** "{instruction}"
+
+This is a CONFLICT experiment about factor bias. The scene contains two objects:
+- Object A: has {obj_a_desc}  — matches the '{factor1}' factor in the instruction
+- Object B: has {obj_b_desc}  — matches the '{factor2}' factor in the instruction
+
+No object has BOTH attributes simultaneously. The robot must choose one.
+
+Answer in this EXACT format:
+ROBOT_ACTION: [describe what the robot arm did in 1 sentence]
+OBJECT_TOUCHED: [A (matches {factor1}), B (matches {factor2}), or neither]
+FACTOR_FOLLOWED: ["{factor1}" if robot went for Object A, "{factor2}" if robot went for Object B, "neither"]
+CONFIDENCE: [high / medium / low]
+REASONING: [1-2 sentences explaining your judgment]"""
+
+
+VERB_EXPS = {"verb_color": "color", "verb_object": "shape",
+             "verb_size": "size", "verb_spatial": "spatial"}
+TWO_OBJ_EXPS = {
+    "color_object": ("color", "shape"), "size_object": ("size", "shape"),
+    "color_size": ("color", "size"), "color_spatial": ("color", "spatial"),
+    "spatial_size": ("spatial", "size"), "spatial_object": ("spatial", "shape"),
+}
+
+def get_prompt(experiment, instruction, pair_i, pair_j, run_type):
+    if experiment in VERB_EXPS:
+        return prompt_verb_experiment(instruction, pair_i, pair_j, VERB_EXPS[experiment])
+    if experiment in TWO_OBJ_EXPS:
+        f1, f2 = TWO_OBJ_EXPS[experiment]
+        return prompt_two_object_experiment(instruction, pair_i, pair_j, f1, f2)
+    return f'Watch this robot video. Instruction: "{instruction}". What did the robot do?'
+
+# ── API call ─────────────────────────────────────────────────────────────────
+
+def call_vlm(video_path, prompt):
+    with open(video_path, "rb") as f:
+        video_b64 = base64.b64encode(f.read()).decode()
+    payload = {"model": MODEL, "max_tokens": 2000,
+        "messages": [{"role": "user", "content": [
+            {"type": "text", "text": prompt},
+            {"type": "image_url", "image_url": {"url": f"data:video/mp4;base64,{video_b64}"}}]}]}
+    for attempt in range(4):
+        try:
+            resp = requests.post(API_URL,
+                headers={"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"},
+                json=payload, timeout=120)
+            if resp.status_code == 200:
+                content = resp.json()["choices"][0]["message"]["content"] or ""
+                return parse_response(content)
+            elif resp.status_code == 429:
+                print("  rate-limited 30s", flush=True); time.sleep(30)
+            else:
+                print(f"  API {resp.status_code}: {resp.text[:120]}", flush=True); time.sleep(5)
+        except Exception as e:
+            print(f"  exc: {e}", flush=True); time.sleep(5)
+    return {"raw": "FAILED", "factor_followed": "error"}
+
+
+def parse_response(text):
+    result = {"raw": text}
+    for field in ["ROBOT_ACTION", "OBJECT_TOUCHED", "ACTION_TYPE",
+                  "FACTOR_FOLLOWED", "CONFIDENCE", "REASONING"]:
+        m = re.search(rf"{field}:\s*(.+?)(?=\n[A-Z_]+:|$)", text, re.DOTALL)
+        result[field.lower()] = m.group(1).strip() if m else ""
+    return result
+
+# ── Run discovery (one dir per index, latest timestamp) ──────────────────────
+
+def load_runs(root: Path, experiment: str):
+    expdir = root / experiment / "experiments"
+    best = {}
+    for d in sorted(expdir.glob(f"ood_*_{experiment}_*")):
+        if not d.is_dir():
+            continue
+        parts = d.name.split("_")
+        try:
+            exp_len = len(experiment.split("_"))
+            idx = int(parts[1])
+            pair_i = int(parts[2 + exp_len])
+            pair_j = int(parts[3 + exp_len])
+            run_type = parts[4 + exp_len]
+            run_name = "_".join(parts[:5 + exp_len])
+        except (ValueError, IndexError):
+            continue
+        sr = d / "success_rate.txt"
+        existing = "?/1"
+        if sr.exists():
+            m = re.search(r"_success=(\d+/\d+)", sr.read_text())
+            if m: existing = m.group(1)
+        best[idx] = {"idx": idx, "pair_i": pair_i, "pair_j": pair_j,
+                     "run_type": run_type, "existing_success": existing,
+                     "run_name": run_name, "dir": str(d)}  # sorted() => last wins = latest ts
+    return [best[k] for k in sorted(best)]
+
+# ── Per-experiment evaluation ────────────────────────────────────────────────
+
+def evaluate_experiment(model: str, experiment: str, limit: int = 0):
+    root = ROOTS[model]
+    out_dir = OUT_BASE / model
+    out_dir.mkdir(parents=True, exist_ok=True)
+    out_file = out_dir / f"vlm_eval_{experiment}.jsonl"
+    print(f"\n{'='*60}\n[{model}] Evaluating: {experiment}\nOutput: {out_file}", flush=True)
+
+    runs = load_runs(root, experiment)
+    print(f"found {len(runs)} runs", flush=True)
+
+    done_names = set()
+    if out_file.exists():
+        for line in out_file.read_text().splitlines():
+            try:
+                done_names.add(json.loads(line)["run_name"])
+            except Exception:
+                pass
+    if done_names:
+        print(f"resume: {len(done_names)} already done, skipping them", flush=True)
+
+    factor_counts = {}
+    # tally already-done from existing jsonl too (for accurate summary)
+    if out_file.exists():
+        for line in out_file.read_text().splitlines():
+            try:
+                f = json.loads(line).get("vlm_factor_followed", "error")
+                factor_counts[f] = factor_counts.get(f, 0) + 1
+            except Exception:
+                pass
+
+    # Build pending list (skip done / missing), honoring optional limit
+    pending = []
+    for run in runs:
+        if run["run_name"] in done_names:
+            continue
+        run_dir = Path(run["dir"])
+        video_path = run_dir / "video" / "ep000.mp4"
+        if not run_dir.exists() or not video_path.exists():
+            print(f"  [{run['idx']}] MISSING dir/video", flush=True); continue
+        instruction = "unknown"
+        sr_file = run_dir / "success_rate.txt"
+        if sr_file.exists():
+            m = re.search(r"instruction='(.+?)'", sr_file.read_text())
+            if m: instruction = m.group(1)
+        run["_video"] = str(video_path); run["_instruction"] = instruction
+        pending.append(run)
+        if limit and len(pending) >= limit:
+            break
+    print(f"pending={len(pending)}  workers={WORKERS}", flush=True)
+
+    out_f = open(out_file, "a")
+    lock = threading.Lock()
+    state = {"n": 0}
+
+    def work(run):
+        prompt = get_prompt(experiment, run["_instruction"], run["pair_i"],
+                            run["pair_j"], run["run_type"])
+        vlm = call_vlm(Path(run["_video"]), prompt)
+        factor = vlm.get("factor_followed", "error").lower().strip('"\'').strip()
+        record = {"model": model, "experiment": experiment, "run_name": run["run_name"],
+            "idx": run["idx"], "pair_i": run["pair_i"], "pair_j": run["pair_j"],
+            "run_type": run["run_type"], "instruction": run["_instruction"],
+            "existing_success": run["existing_success"],
+            "vlm_factor_followed": factor,
+            "vlm_object_touched": vlm.get("object_touched", ""),
+            "vlm_action_type": vlm.get("action_type", ""),
+            "vlm_robot_action": vlm.get("robot_action", ""),
+            "vlm_confidence": vlm.get("confidence", ""),
+            "vlm_reasoning": vlm.get("reasoning", ""),
+            "vlm_raw": vlm.get("raw", "")}
+        with lock:
+            out_f.write(json.dumps(record) + "\n"); out_f.flush()
+            factor_counts[factor] = factor_counts.get(factor, 0) + 1
+            state["n"] += 1
+            print(f"  [{model}/{experiment} {state['n']}/{len(pending)}] "
+                  f"idx{run['idx']} {factor.upper()} [{vlm.get('confidence','')}]", flush=True)
+
+    if pending:
+        with ThreadPoolExecutor(max_workers=WORKERS) as ex:
+            list(ex.map(work, pending))
+    out_f.close()
+
+    total = sum(factor_counts.values())
+    with open(out_dir / f"vlm_eval_{experiment}_summary.txt", "w") as f:
+        f.write(f"# VLM Eval: {model} {experiment}\nmodel: {MODEL}\ntotal_evaluated: {total}\n\n")
+        for fac, cnt in sorted(factor_counts.items(), key=lambda x: -x[1]):
+            f.write(f"{fac}: {cnt}/{total} ({100*cnt/max(1,total):.1f}%)\n")
+    print(f"  [{model}/{experiment}] +{state['n']} new, totals={factor_counts}", flush=True)
+    return factor_counts
+
+
+if __name__ == "__main__":
+    args = [a for a in sys.argv[1:] if not a.startswith("--")]
+    limit = 0
+    for i, a in enumerate(sys.argv[1:], start=1):
+        if a.startswith("--limit"):
+            limit = int(a.split("=")[1]) if "=" in a else int(sys.argv[i+1])
+        if a.startswith("--workers"):
+            WORKERS = int(a.split("=")[1]) if "=" in a else int(sys.argv[i+1])
+    model = args[0] if args else "gr00t"
+    exps = [a for a in args[1:] if a in ALL_EXPERIMENTS] or ALL_EXPERIMENTS
+    assert model in ROOTS, f"model must be gr00t|genie, got {model}"
+    results = {}
+    for e in exps:
+        results[e] = evaluate_experiment(model, e, limit=limit)
+    print("\n" + "="*60 + f"\n[{model}] ALL DONE")
+    for e, c in results.items():
+        t = sum(c.values()); top = max(c, key=c.get) if c else "NA"
+        print(f"  {e}: dominant={top} ({100*c.get(top,0)/max(1,t):.1f}%)")

code/watchdog_pi0_f18.sh

@@ -0,0 +1,64 @@
+#!/usr/bin/env bash
+set -uo pipefail
+# ─────────────────────────────────────────────────────────────────────────────
+# watchdog_pi0_f18.sh — watch /workspace/pi0_ckpt, auto-evaluate every *f18*
+# checkpoint that has finished transferring.  One model per GPU, each in its
+# own tmux session (pi0_g<gpu>).  task_difficulty harder (run_pi0_one_ckpt.sh
+# default 1.3).  Skips done (SUMMARY.txt) / already-claimed / in-flight.
+#
+# Runs forever (user keeps uploading); rescans every 120s.
+# Launch:  tmux new-session -d -s pi0_watchdog 'bash watchdog_pi0_f18.sh'
+# ─────────────────────────────────────────────────────────────────────────────
+ROOT=/workspace/groot_eval
+HARNESS="${ROOT}/harness"
+CKPT_ROOT=/workspace/pi0_ckpt
+RES_ROOT="${ROOT}/results_pi0_pairwise"
+LOG_DIR="${ROOT}/logs/pi0_pairwise"
+CLAIM_DIR="${LOG_DIR}/.claimed"
+GPUS=(0 1 2 3 4 5 6 7)
+STEP=17999
+mkdir -p "${LOG_DIR}" "${CLAIM_DIR}" "${RES_ROOT}"
+
+log(){ echo "[$(date +%H:%M:%S)] $*"; }
+
+free_gpu(){
+  for g in "${GPUS[@]}"; do
+    tmux has-session -t "pi0_g${g}" 2>/dev/null || { echo "$g"; return 0; }
+  done
+  echo ""
+}
+
+complete(){  # $1=ckpt dir ; true if 17999/ fully transferred & stable
+  local s="$1/${STEP}"
+  [[ -f "$s/model.safetensors" && -f "$s/metadata.pt" ]] || return 1
+  ls "$s"/assets/*/norm_stats.json >/dev/null 2>&1 || return 1
+  local mb age
+  mb=$(du -m "$s/model.safetensors" 2>/dev/null | cut -f1)
+  [[ "${mb:-0}" -ge 3000 ]] || return 1
+  age=$(( $(date +%s) - $(stat -c %Y "$s/model.safetensors" 2>/dev/null || echo 0) ))
+  [[ "${age}" -ge 90 ]] || return 1   # not written to for 90s
+  return 0
+}
+
+log "watchdog start — scanning ${CKPT_ROOT} for *f18* every 120s"
+while true; do
+  for d in "${CKPT_ROOT}"/*f18*/; do
+    [[ -d "$d" ]] || continue
+    name="$(basename "$d")"
+    case "$name" in ckpt_batch*|ckpt_may_*) continue;; esac
+    [[ "$name" == *f18* ]] || continue
+    [[ -f "${RES_ROOT}/${name}/SUMMARY.txt" ]] && continue          # done
+    [[ -f "${CLAIM_DIR}/${name}" ]] && continue                     # claimed/in-flight
+    complete "$d" || continue                                       # not fully transferred
+    g="$(free_gpu)"
+    [[ -z "$g" ]] && { log "all GPUs busy; ${name} waits"; break; }
+    port=$(( 8100 + g ))
+    touch "${CLAIM_DIR}/${name}"
+    log "LAUNCH ${name} → GPU${g} (tmux pi0_g${g}, port ${port})"
+    tmux new-session -d -s "pi0_g${g}" \
+      "cd ${HARNESS} && bash run_pi0_one_ckpt.sh '${name}' ${g} ${port} \
+         > ${LOG_DIR}/run_${name}.log 2>&1; \
+       echo DONE_${name} >> ${LOG_DIR}/watchdog_events.log"
+  done
+  sleep 120
+done

result.md

@@ -0,0 +1,245 @@
+# 仿真评测结果总汇 (GR00T N1.7 / Genie-Envisioner)
+
+生成时间:2026-05-25 19:20:58 UTC
+
+本文件汇总全部实验:① GR00T pair-grid 硬/软口径 ② GR00T all_factor full-factor ③ conflict 双轴 env success(GR00T vs Genie)④ VLM(Gemini-2.5-flash)FDR 对照。
+数值为成功率 %;pair-grid/all_factor 均为 3-seed 平均。
+
+---
+# 一、GR00T pair-grid(硬口径,3-seed)
+
+## color_size (HARD)
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| color_size_Lrandom_f12 | 24.1 | 27.3 | 24.5 | **25.3%** |
+| color_size_Lrandom_f18 | 35.6 | 34.7 | 34.3 | **34.8%** |
+| color_size_Lrandom_f6 | 7.9 | 10.2 | 6.9 | **8.3%** |
+| color_size_random_f12 | 13.4 | 15.3 | 11.6 | **13.4%** |
+| color_size_random_f18 | 27.3 | 32.9 | 28.2 | **29.4%** |
+| color_size_random_f6 | 9.7 | 9.3 | 10.6 | **9.8%** |
+| color_size_stair_f12 | 28.2 | 28.2 | 28.2 | **28.2%** |
+| color_size_stair_f18 | 26.9 | 27.8 | 25.9 | **26.8%** |
+| color_size_stair_f6 | 7.4 | 8.3 | 6.0 | **7.2%** |
+
+## color_size (EASY)
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| color_size_Lrandom_f12 | 23.6 | 24.5 | 22.7 | **23.6%** |
+| color_size_Lrandom_f18 | 36.6 | 38.0 | 37.0 | **37.2%** |
+| color_size_Lrandom_f6 | 5.1 | 10.6 | 9.3 | **8.3%** |
+| color_size_random_f12 | 14.4 | 14.8 | 16.2 | **15.1%** |
+| color_size_random_f18 | 28.7 | 33.3 | 31.0 | **31.0%** |
+| color_size_random_f6 | 6.9 | 11.1 | 6.9 | **8.3%** |
+| color_size_stair_f12 | 29.2 | 27.8 | 29.6 | **28.8%** |
+| color_size_stair_f18 | 26.4 | 25.5 | 27.8 | **26.5%** |
+| color_size_stair_f6 | 9.3 | 8.3 | 6.5 | **8.0%** |
+
+## color_spatial
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| color_spatial_Lrandom_f18 | 52.4 | 42.9 | 46.7 | **47.3%** |
+| color_spatial_random_f12 | 33.3 | 27.6 | 33.8 | **31.5%** |
+| color_spatial_random_f18 | 42.4 | 36.7 | 41.0 | **40.0%** |
+| color_spatial_random_f6 | 23.3 | 26.2 | 24.8 | **24.7%** |
+| color_spatial_stair_f18 | 54.8 | 45.7 | 52.9 | **51.1%** |
+| color_spatial_stair_f6 | 24.3 | 19.5 | 21.4 | **21.7%** |
+
+## verb_spatial
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| verb_spatial_Lrandom_f12 | 31.9 | 29.5 | 32.4 | **31.2%** |
+| verb_spatial_Lrandom_f18 | 32.9 | 30.0 | 34.8 | **32.5%** |
+| verb_spatial_Lrandom_f6 | 22.4 | 31.0 | 27.1 | **26.8%** |
+| verb_spatial_random_f12 | 31.4 | 31.9 | 37.6 | **33.6%** |
+| verb_spatial_random_f18 | 26.2 | 24.3 | 23.3 | **24.6%** |
+| verb_spatial_random_f6 | 16.2 | 16.2 | 14.3 | **15.5%** |
+| verb_spatial_stair_f12 | 26.7 | 28.6 | 25.2 | **26.8%** |
+| verb_spatial_stair_f18 | 34.3 | 37.1 | 35.2 | **35.5%** |
+| verb_spatial_stair_f6 | 23.3 | 27.1 | 22.9 | **24.4%** |
+
+## spatial_size
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| spatial_size_Lrandom_f12 | 38.1 | 30.0 | 33.3 | **33.8%** |
+| spatial_size_Lrandom_f18 | 32.4 | 35.7 | 33.8 | **33.9%** |
+| spatial_size_Lrandom_f6 | 24.3 | 21.0 | 24.8 | **23.3%** |
+| spatial_size_random_f12 | 37.6 | 33.3 | 34.3 | **35.0%** |
+| spatial_size_random_f18 | 27.6 | 30.5 | 28.1 | **28.7%** |
+| spatial_size_random_f6 | 21.4 | 17.6 | 24.3 | **21.1%** |
+| spatial_size_stair_f12 | 36.2 | 37.6 | 40.5 | **38.1%** |
+| spatial_size_stair_f18 | 46.7 | 46.7 | 51.0 | **48.1%** |
+
+## spatial_object
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| spatial_object_Lrandom_f12 | 40.0 | 40.0 | 38.6 | **39.5%** |
+| spatial_object_Lrandom_f18 | 34.8 | 31.4 | 32.9 | **33.0%** |
+| spatial_object_random_f12 | 27.1 | 32.4 | 29.5 | **29.6%** |
+| spatial_object_random_f18 | 34.3 | 36.7 | 32.4 | **34.4%** |
+| spatial_object_random_f6 | 17.6 | 17.6 | 15.7 | **16.9%** |
+| spatial_object_stair_f12 | 30.5 | 39.0 | 34.8 | **34.7%** |
+| spatial_object_stair_f18 | 32.9 | 34.8 | 28.6 | **32.1%** |
+| spatial_object_stair_f6 | 28.6 | 32.9 | 26.2 | **29.2%** |
+
+## verb_size
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| verb_size_Lrandom_f12 | 13.4 | 15.3 | 16.7 | **15.1%** |
+| verb_size_Lrandom_f18 | 22.2 | 23.6 | 23.6 | **23.1%** |
+| verb_size_Lrandom_f6 | 8.3 | 9.3 | 7.4 | **8.3%** |
+| verb_size_random_f12 | 23.1 | 24.5 | 25.0 | **24.2%** |
+| verb_size_random_f18 | 14.4 | 13.9 | 11.6 | **13.3%** |
+| verb_size_random_f6 | 10.2 | 8.8 | 7.4 | **8.8%** |
+| verb_size_stair1_f12 | 20.4 | 17.6 | 17.1 | **18.3%** |
+| verb_size_stair1_f18 | 17.6 | 16.2 | 18.5 | **17.4%** |
+
+---
+# 二、GR00T all_factor (full-factor, pi0.5 对齐, 3-seed)
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| all_factor_Lrandom_f50_n100 | 46.5 | 43.5 | 49.5 | **46.5%** |
+| all_factor_Lrandom_f50_n200 | 44.0 | 43.0 | 36.0 | **41.0%** |
+| all_factor_Lrandom_f50_n400 | 45.5 | 48.0 | 47.5 | **47.0%** |
+| all_factor_Lrandom_n200 | 34.0 | 26.0 | 35.0 | **31.6%** |
+| all_factor_Lrandom_n400 | 31.0 | 33.0 | 37.0 | **33.6%** |
+| all_factor_Lrandom_n800 | 32.0 | 31.5 | 38.0 | **33.8%** |
+| all_factor_stairoriginal2_n200 | 38.5 | 35.0 | 38.5 | **37.3%** |
+| all_factor_stairverbsize_n200 | 36.5 | 32.5 | 38.0 | **35.6%** |
+| all_factor_stairverbsize_n400 | 44.5 | 41.5 | 38.5 | **41.5%** |
+| all_factor_stairverbsize_n800 | 29.0 | 31.5 | 32.0 | **30.8%** |
+| all_factor_stairvss_f50_n200 | 40.0 | 40.5 | 44.5 | **41.6%** |
+| all_factor_stairvss_f50_n400 | 42.0 | 43.0 | 43.5 | **42.8%** |
+| all_factor_stairvss_f50_n800 | 42.5 | 45.5 | 51.5 | **46.5%** |
+| all_factor_true_random_n200 | 36.0 | 31.0 | 34.5 | **33.8%** |
+| all_factor_true_random_n400 | 40.5 | 38.0 | 36.0 | **38.1%** |
+| all_factor_true_random_n800 | 41.5 | 38.5 | 44.0 | **41.3%** |
+
+---
+# 三、Conflict-experiment 双轴 env success (seed42, 200 ep)
+
+每个实验两个因子的 success(模型把任务做对、且作用在正确因子目标上)。
+
+| experiment | GR00T 因子A | GR00T 因子B | Genie 因子A | Genie 因子B |
+|---|---|---|---|---|
+| color_object | 15.0% | 14.0% | 11.7% | 6.7% |
+| color_size | 47.0% | 29.5% | 1.5% | 0.0% |
+| color_spatial | 5.5% | 39.5% | 10.5% | 38.5% |
+| size_object | 18.5% | 29.5% | 0.0% | 0.0% |
+| spatial_object | 9.0% | 22.0% | 1.5% | 16.2% |
+| spatial_size | 1.0% | 25.0% | 30.5% | 18.5% |
+| verb_color | 24.5% | 24.0% | 13.5% | 11.5% |
+| verb_object | 8.0% | 31.5% | 4.9% | 3.2% |
+| verb_size | 17.0% | 9.0% | 19.5% | 22.5% |
+| verb_spatial | 38.5% | 4.0% | 7.0% | 14.5% |
+
+---
+# 四、VLM (Gemini-2.5-flash) FDR 对照
+
+```
+============================================================================================
+Factor Dominance Rate (FDR) 三方对照 — FDR=(S_f1-S_f2)/(S_f1+S_f2), >0偏因子A <0偏因子B
+env  = 基于 ManiSkill 环境 success(每模型 200 episodes)
+vlm  = 基于 Gemini-2.5-flash 看视频判 factor_followed(每模型 400 视频/实验)
+============================================================================================
+experiment     A/B           env_GR00T env_genie  vlm_GR00T  vlm_genie
+--------------------------------------------------------------------------------------------
+color_object   color/shape      +0.034    +0.034     +0.228     +0.222
+color_size     color/size       +0.229    +1.000     +0.524     +0.309
+color_spatial  color/spati      -0.756    -0.571     +0.084     +0.365
+size_object    size/shape       -0.229    +0.602     -0.241     -0.114
+spatial_object spati/shape      -0.419    -0.713     -0.095     -0.209
+spatial_size   spati/size       -0.923    +0.245     +0.110     +0.237
+verb_color     verb/color       +0.010    +0.080     -0.068     -0.220
+verb_object    verb/shape       -0.595    -0.062     -0.253     -0.250
+verb_size      verb/size        +0.308    -0.071     +0.251     +0.326
+verb_spatial   verb/spati       +0.812    -0.349     -0.122     -0.142
+--------------------------------------------------------------------------------------------
+
+VLM 判定明细(N_f1 / N_f2 / neither|error,共400):
+```
+
+---
+# 五、pi0 (openpi pi-zero) pair-grid (硬口径, 3-seed)
+
+> 与 GR00T pair-grid 同口径(同色同形干扰项),但 sim_backend=cpu(pi0 server+GPU-sim 冲突)。注意 pi0 与 GR00T 推理栈不同,数值横向比 GR00T 仅供参考。
+
+## pi0 — color_spatial
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| color_spatial_Lrandom_f12 | 52.4 | 41.4 | 49.5 | **47.7%** |
+| color_spatial_Lrandom_f18 | 53.8 | 41.0 | 48.6 | **47.8%** |
+| color_spatial_Lrandom_f6 | 28.6 | 23.3 | 26.7 | **26.2%** |
+| color_spatial_random_f12 | 38.6 | 32.4 | 37.6 | **36.2%** |
+| color_spatial_random_f18 | 51.0 | 41.9 | 47.6 | **46.8%** |
+| color_spatial_random_f6 | 27.6 | 21.4 | 23.3 | **24.1%** |
+| color_spatial_stair_f12 | 49.0 | 38.1 | 42.4 | **43.1%** |
+| color_spatial_stair_f18 | 60.0 | 49.0 | 55.7 | **54.9%** |
+| color_spatial_stair_f6 | 31.0 | 24.8 | 28.6 | **28.1%** |
+
+## pi0 — color_size
+
+| ckpt | s42 | s40 | s41 | 3-seed 均 |
+|---|---|---|---|---|
+| color_size_Lrandom_f12 | 27.3 | - | - | 1/3(部分) |
+| color_size_Lrandom_f18 | 30.6 | 29.2 | 27.3 | **29.0%** |
+| color_size_Lrandom_f6 | 26.4 | - | - | 1/3(部分) |
+| color_size_random_f12 | - | - | - | 0/3(部分) |
+| color_size_random_f18 | 28.7 | 25.9 | 23.1 | **25.9%** |
+| color_size_random_f6 | - | - | - | 0/3(部分) |
+| color_size_stair_f18 | 26.4 | 28.2 | 24.1 | **26.2%** |
+
+
+---
+# 附:可复现配置(干扰项 / 口径明细)
+
+## 公共设置(pair-grid 全实验)
+- 环境:`VerbObjectColor-v1`(ManiSkill),obs=rgb,control=pd_joint_pos
+- task_difficulty=**1.5**,sim/render_backend=**gpu**,max_episode_steps=**150**,replan_steps=**10**
+- 每 seed:cells × reps,target≈210 episodes;seeds=**{42,40,41}**(3-seed 平均)
+- 干扰项核心:**同色**(且多数同形 cube)干扰项,强制模型靠目标因子区分,而非靠颜色/形状走捷径
+- spatial 锚点(xy,米)+ 抖动 ±0.012:`left(-0.10,0) right(0.10,0) middle(0,0) front(0,0.10) behind(0,-0.10)`
+
+## color_size — HARD 口径
+- verb=lift, shape=cube;指令 `"Lift the {size} {color} cube."`
+- 干扰项 = 同色 cube,仅尺寸不同。(target_scale, [干扰scale...], 干扰数):
+  - small (0.72,[1.20],1) · large (1.34,[0.80],1) · smaller (0.82,[1.08],1) · larger (1.18,[0.92],1) · smallest (0.78,[1.00,1.24],**2**) · largest (1.26,[1.00,0.80],**2**)
+
+## color_size — EASY 口径
+- 同 HARD 但干扰项恒为 1 个、对比拉大 ~2×:
+  - small (0.65,[1.35],1) · large (1.40,[0.65],1) · smaller (0.70,[1.30],1) · larger (1.30,[0.70],1) · smallest (0.65,[1.40],1) · largest (1.40,[0.65],1)
+
+## color_spatial — HARD 口径
+- shape=cube,verb~{lift,grasp,push} 随机;指令 `"{Verb} the {color} cube {spatial_phrase}."`
+- 干扰项 = 1 个**同色 cube**,放在**不同 spatial 锚点**;target_size_scale=1.0
+
+## verb_spatial — HARD 口径
+- shape=cube,color 每 cell 随机;指令 `"{Verb} the cube {spatial_phrase}."`
+- 干扰项 = 1 个**同色 cube**,放在**不同 spatial 锚点**
+
+## spatial_size — HARD 口径
+- shape=cube,color&verb 随机;指令 `"{Verb} the {size} cube {spatial_phrase}."`
+- 干扰项 = 1 个**同色 cube**,**不同 spatial + 不同尺寸**(目标偏小→干扰 scale=1.25;目标偏大→0.75)
+
+## spatial_object — HARD 口径
+- color&verb 随机;指令 `"{Verb} the {shape} {spatial_phrase}."`
+- 干扰项 = 1 个**同色、不同形状**物体,放在**不同 spatial 锚点**
+
+## verb_size — HARD 口径
+- shape=cube,color 随机;指令 `"{Verb} the {size} cube."`
+- 干扰项 = 同色 cube,仅尺寸不同(同 color_size HARD 的 HARD_SIZE 表)
+
+## all_factor — full-factor(对齐 pi0.5)
+- 单指令全因子采样;sample_n=200, sample_seed=42, 200 episodes/seed
+- max_episode_steps=**500**, **no_distractor_prob=0.70**, 默认 difficulty, sim=gpu;seeds={40,41,42}
+
+## pi0 pair-grid
+- 与 GR00T pair-grid 同口径,但 **sim_backend=cpu**(pi0 server 与 GPU-sim 冲突);其余同上

results/genie/conflict_env/genie_color_object_ood_seed42.txt

@@ -0,0 +1,420 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=color_object
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/color_object
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 4 0 color 0/1 ood_001_color_object_4_0_color
+2 4 0 shape 1/1 ood_002_color_object_4_0_shape
+3 0 4 color 0/1 ood_003_color_object_0_4_color
+4 0 4 shape 0/1 ood_004_color_object_0_4_shape
+5 0 1 color 0/1 ood_005_color_object_0_1_color
+6 0 1 shape 0/1 ood_006_color_object_0_1_shape
+7 4 3 color 0/1 ood_007_color_object_4_3_color
+8 4 3 shape 0/1 ood_008_color_object_4_3_shape
+9 1 4 color 0/1 ood_009_color_object_1_4_color
+10 1 4 shape 0/1 ood_010_color_object_1_4_shape
+11 1 3 color 1/1 ood_011_color_object_1_3_color
+12 1 3 shape 0/1 ood_012_color_object_1_3_shape
+13 1 3 color 0/1 ood_013_color_object_1_3_color
+14 1 3 shape 0/1 ood_014_color_object_1_3_shape
+15 0 5 color 0/1 ood_015_color_object_0_5_color
+16 0 5 shape 0/1 ood_016_color_object_0_5_shape
+17 4 3 color 0/1 ood_017_color_object_4_3_color
+18 4 3 shape 1/1 ood_018_color_object_4_3_shape
+19 0 4 color 0/1 ood_019_color_object_0_4_color
+20 0 4 shape 0/1 ood_020_color_object_0_4_shape
+21 4 1 color 0/1 ood_021_color_object_4_1_color
+22 4 1 shape 0/1 ood_022_color_object_4_1_shape
+23 4 3 color 0/1 ood_023_color_object_4_3_color
+24 4 3 shape 0/1 ood_024_color_object_4_3_shape
+25 5 3 color 0/1 ood_025_color_object_5_3_color
+26 5 3 shape 0/1 ood_026_color_object_5_3_shape
+27 3 2 color 0/1 ood_027_color_object_3_2_color
+28 3 2 shape 0/1 ood_028_color_object_3_2_shape
+29 0 3 color 0/1 ood_029_color_object_0_3_color
+30 0 3 shape 0/1 ood_030_color_object_0_3_shape
+31 3 4 color 0/1 ood_031_color_object_3_4_color
+32 3 4 shape 0/1 ood_032_color_object_3_4_shape
+33 2 4 color 0/1 ood_033_color_object_2_4_color
+34 2 4 shape 0/1 ood_034_color_object_2_4_shape
+35 0 2 color 0/1 ood_035_color_object_0_2_color
+36 0 2 shape 0/1 ood_036_color_object_0_2_shape
+37 0 1 color 0/1 ood_037_color_object_0_1_color
+38 0 1 shape 0/1 ood_038_color_object_0_1_shape
+39 0 3 color 0/1 ood_039_color_object_0_3_color
+40 0 3 shape 1/1 ood_040_color_object_0_3_shape
+41 1 2 color 0/1 ood_041_color_object_1_2_color
+42 1 2 shape 0/1 ood_042_color_object_1_2_shape
+43 1 3 color 0/1 ood_043_color_object_1_3_color
+44 1 3 shape 0/1 ood_044_color_object_1_3_shape
+45 3 1 color 0/1 ood_045_color_object_3_1_color
+46 3 1 shape 0/1 ood_046_color_object_3_1_shape
+47 3 5 color 0/1 ood_047_color_object_3_5_color
+48 3 5 shape 0/1 ood_048_color_object_3_5_shape
+49 0 1 color 0/1 ood_049_color_object_0_1_color
+50 0 1 shape 0/1 ood_050_color_object_0_1_shape
+51 3 2 color 0/1 ood_051_color_object_3_2_color
+52 3 2 shape 0/1 ood_052_color_object_3_2_shape
+53 1 2 color 0/1 ood_053_color_object_1_2_color
+54 1 2 shape 0/1 ood_054_color_object_1_2_shape
+55 4 2 color 0/1 ood_055_color_object_4_2_color
+56 4 2 shape 0/1 ood_056_color_object_4_2_shape
+57 4 0 color 0/1 ood_057_color_object_4_0_color
+58 4 0 shape 0/1 ood_058_color_object_4_0_shape
+59 4 2 color 0/1 ood_059_color_object_4_2_color
+60 4 2 shape 0/1 ood_060_color_object_4_2_shape
+61 3 2 color 0/1 ood_061_color_object_3_2_color
+62 3 2 shape 0/1 ood_062_color_object_3_2_shape
+63 2 4 color 0/1 ood_063_color_object_2_4_color
+64 2 4 shape 0/1 ood_064_color_object_2_4_shape
+65 1 3 color 0/1 ood_065_color_object_1_3_color
+66 1 3 shape 0/1 ood_066_color_object_1_3_shape
+67 2 5 color 0/1 ood_067_color_object_2_5_color
+68 2 5 shape 0/1 ood_068_color_object_2_5_shape
+69 3 4 color 0/1 ood_069_color_object_3_4_color
+70 3 4 shape 0/1 ood_070_color_object_3_4_shape
+71 1 4 color 0/1 ood_071_color_object_1_4_color
+72 1 4 shape 0/1 ood_072_color_object_1_4_shape
+73 5 0 color 0/1 ood_073_color_object_5_0_color
+74 5 0 shape 0/1 ood_074_color_object_5_0_shape
+75 5 2 color 0/1 ood_075_color_object_5_2_color
+76 5 2 shape 1/1 ood_076_color_object_5_2_shape
+77 0 1 color 0/1 ood_077_color_object_0_1_color
+78 0 1 shape 0/1 ood_078_color_object_0_1_shape
+79 4 5 color 0/1 ood_079_color_object_4_5_color
+80 4 5 shape 0/1 ood_080_color_object_4_5_shape
+81 5 0 color 0/1 ood_081_color_object_5_0_color
+82 5 0 shape 0/1 ood_082_color_object_5_0_shape
+83 1 0 color 0/1 ood_083_color_object_1_0_color
+84 1 0 shape 0/1 ood_084_color_object_1_0_shape
+85 4 2 color 0/1 ood_085_color_object_4_2_color
+86 4 2 shape 0/1 ood_086_color_object_4_2_shape
+87 2 4 color 1/1 ood_087_color_object_2_4_color
+88 2 4 shape 0/1 ood_088_color_object_2_4_shape
+89 2 0 color 0/1 ood_089_color_object_2_0_color
+90 2 0 shape 0/1 ood_090_color_object_2_0_shape
+91 1 4 color 0/1 ood_091_color_object_1_4_color
+92 1 4 shape 0/1 ood_092_color_object_1_4_shape
+93 0 5 color 0/1 ood_093_color_object_0_5_color
+94 0 5 shape 0/1 ood_094_color_object_0_5_shape
+95 1 2 color 0/1 ood_095_color_object_1_2_color
+96 1 2 shape 0/1 ood_096_color_object_1_2_shape
+97 4 5 color 0/1 ood_097_color_object_4_5_color
+98 4 5 shape 0/1 ood_098_color_object_4_5_shape
+99 2 0 color 0/1 ood_099_color_object_2_0_color
+100 2 0 shape 0/1 ood_100_color_object_2_0_shape
+101 0 4 color 0/1 ood_101_color_object_0_4_color
+102 0 4 shape 0/1 ood_102_color_object_0_4_shape
+103 0 3 color 0/1 ood_103_color_object_0_3_color
+104 0 3 shape 0/1 ood_104_color_object_0_3_shape
+105 2 3 color 0/1 ood_105_color_object_2_3_color
+106 2 3 shape 0/1 ood_106_color_object_2_3_shape
+107 0 4 color 0/1 ood_107_color_object_0_4_color
+108 0 4 shape 0/1 ood_108_color_object_0_4_shape
+109 2 1 color 0/1 ood_109_color_object_2_1_color
+110 2 1 shape 0/1 ood_110_color_object_2_1_shape
+111 5 2 color 0/1 ood_111_color_object_5_2_color
+112 5 2 shape 0/1 ood_112_color_object_5_2_shape
+113 2 1 color 0/1 ood_113_color_object_2_1_color
+114 2 1 shape 0/1 ood_114_color_object_2_1_shape
+115 3 5 color 0/1 ood_115_color_object_3_5_color
+116 3 5 shape 0/1 ood_116_color_object_3_5_shape
+117 1 4 color 0/1 ood_117_color_object_1_4_color
+118 1 4 shape 0/1 ood_118_color_object_1_4_shape
+119 5 0 color 0/1 ood_119_color_object_5_0_color
+120 5 0 shape 0/1 ood_120_color_object_5_0_shape
+121 0 2 color 1/1 ood_121_color_object_0_2_color
+122 0 2 shape 1/1 ood_122_color_object_0_2_shape
+123 4 3 color 0/1 ood_123_color_object_4_3_color
+124 4 3 shape 0/1 ood_124_color_object_4_3_shape
+125 2 5 color 0/1 ood_125_color_object_2_5_color
+126 2 5 shape 0/1 ood_126_color_object_2_5_shape
+127 3 2 color 0/1 ood_127_color_object_3_2_color
+128 3 2 shape 0/1 ood_128_color_object_3_2_shape
+129 0 4 color 0/1 ood_129_color_object_0_4_color
+130 0 4 shape 0/1 ood_130_color_object_0_4_shape
+131 5 4 color 0/1 ood_131_color_object_5_4_color
+132 5 4 shape 0/1 ood_132_color_object_5_4_shape
+133 2 3 color 0/1 ood_133_color_object_2_3_color
+134 2 3 shape 0/1 ood_134_color_object_2_3_shape
+135 0 3 color 1/1 ood_135_color_object_0_3_color
+136 0 3 shape 0/1 ood_136_color_object_0_3_shape
+137 3 2 color 0/1 ood_137_color_object_3_2_color
+138 3 2 shape 0/1 ood_138_color_object_3_2_shape
+139 1 5 color 0/1 ood_139_color_object_1_5_color
+140 1 5 shape 0/1 ood_140_color_object_1_5_shape
+141 5 1 color 0/1 ood_141_color_object_5_1_color
+142 5 1 shape 0/1 ood_142_color_object_5_1_shape
+143 4 0 color 0/1 ood_143_color_object_4_0_color
+144 4 0 shape 0/1 ood_144_color_object_4_0_shape
+145 3 5 color 0/1 ood_145_color_object_3_5_color
+146 3 5 shape 0/1 ood_146_color_object_3_5_shape
+147 5 3 color 0/1 ood_147_color_object_5_3_color
+148 5 3 shape 0/1 ood_148_color_object_5_3_shape
+149 5 2 color 0/1 ood_149_color_object_5_2_color
+150 5 2 shape 0/1 ood_150_color_object_5_2_shape
+151 2 1 color 0/1 ood_151_color_object_2_1_color
+152 2 1 shape 0/1 ood_152_color_object_2_1_shape
+153 3 4 color 0/1 ood_153_color_object_3_4_color
+154 3 4 shape 0/1 ood_154_color_object_3_4_shape
+155 1 2 color 0/1 ood_155_color_object_1_2_color
+156 1 2 shape 0/1 ood_156_color_object_1_2_shape
+157 4 2 color 1/1 ood_157_color_object_4_2_color
+158 4 2 shape 0/1 ood_158_color_object_4_2_shape
+159 0 3 color 0/1 ood_159_color_object_0_3_color
+160 0 3 shape 0/1 ood_160_color_object_0_3_shape
+161 0 2 color 0/1 ood_161_color_object_0_2_color
+162 0 2 shape 0/1 ood_162_color_object_0_2_shape
+163 4 1 color 0/1 ood_163_color_object_4_1_color
+164 4 1 shape 0/1 ood_164_color_object_4_1_shape
+165 1 3 color 0/1 ood_165_color_object_1_3_color
+166 1 3 shape 0/1 ood_166_color_object_1_3_shape
+167 4 5 color 0/1 ood_167_color_object_4_5_color
+168 4 5 shape 0/1 ood_168_color_object_4_5_shape
+169 1 5 color 0/1 ood_169_color_object_1_5_color
+170 1 5 shape 0/1 ood_170_color_object_1_5_shape
+171 0 3 color 0/1 ood_171_color_object_0_3_color
+172 0 3 shape 0/1 ood_172_color_object_0_3_shape
+173 5 2 color 0/1 ood_173_color_object_5_2_color
+174 5 2 shape 0/1 ood_174_color_object_5_2_shape
+175 1 3 color 1/1 ood_175_color_object_1_3_color
+176 1 3 shape 0/1 ood_176_color_object_1_3_shape
+177 5 2 color 0/1 ood_177_color_object_5_2_color
+178 5 2 shape 0/1 ood_178_color_object_5_2_shape
+179 0 4 color 0/1 ood_179_color_object_0_4_color
+180 0 4 shape 0/1 ood_180_color_object_0_4_shape
+181 2 3 color 0/1 ood_181_color_object_2_3_color
+182 2 3 shape 0/1 ood_182_color_object_2_3_shape
+183 1 4 color 0/1 ood_183_color_object_1_4_color
+184 1 4 shape 0/1 ood_184_color_object_1_4_shape
+185 2 5 color 0/1 ood_185_color_object_2_5_color
+186 2 5 shape 0/1 ood_186_color_object_2_5_shape
+187 4 0 color 0/1 ood_187_color_object_4_0_color
+188 4 0 shape 0/1 ood_188_color_object_4_0_shape
+189 5 1 color 0/1 ood_189_color_object_5_1_color
+190 5 1 shape 0/1 ood_190_color_object_5_1_shape
+191 2 1 color 0/1 ood_191_color_object_2_1_color
+192 2 1 shape 0/1 ood_192_color_object_2_1_shape
+193 1 0 color 0/1 ood_193_color_object_1_0_color
+194 1 0 shape 0/1 ood_194_color_object_1_0_shape
+195 2 1 color 0/1 ood_195_color_object_2_1_color
+196 2 1 shape 0/1 ood_196_color_object_2_1_shape
+197 2 1 color 0/1 ood_197_color_object_2_1_color
+198 2 1 shape 0/1 ood_198_color_object_2_1_shape
+199 1 2 color 0/1 ood_199_color_object_1_2_color
+200 1 2 shape 1/1 ood_200_color_object_1_2_shape
+201 4 1 color 1/1 ood_201_color_object_4_1_color
+202 4 1 shape 1/1 ood_202_color_object_4_1_shape
+203 1 4 color 1/1 ood_203_color_object_1_4_color
+204 1 4 shape 1/1 ood_204_color_object_1_4_shape
+205 4 2 color 0/1 ood_205_color_object_4_2_color
+206 4 2 shape 1/1 ood_206_color_object_4_2_shape
+207 5 4 color 0/1 ood_207_color_object_5_4_color
+208 5 4 shape 0/1 ood_208_color_object_5_4_shape
+209 4 1 color 0/1 ood_209_color_object_4_1_color
+210 4 1 shape 1/1 ood_210_color_object_4_1_shape
+211 4 0 color 1/1 ood_211_color_object_4_0_color
+212 4 0 shape 0/1 ood_212_color_object_4_0_shape
+213 0 3 color 0/1 ood_213_color_object_0_3_color
+214 0 3 shape 0/1 ood_214_color_object_0_3_shape
+215 3 5 color 0/1 ood_215_color_object_3_5_color
+216 3 5 shape 0/1 ood_216_color_object_3_5_shape
+217 4 0 color 0/1 ood_217_color_object_4_0_color
+218 4 0 shape 0/1 ood_218_color_object_4_0_shape
+219 1 0 color 1/1 ood_219_color_object_1_0_color
+220 1 0 shape 0/1 ood_220_color_object_1_0_shape
+221 3 2 color 0/1 ood_221_color_object_3_2_color
+222 3 2 shape 0/1 ood_222_color_object_3_2_shape
+223 4 3 color 0/1 ood_223_color_object_4_3_color
+224 4 3 shape 1/1 ood_224_color_object_4_3_shape
+225 1 3 color 0/1 ood_225_color_object_1_3_color
+226 1 3 shape 0/1 ood_226_color_object_1_3_shape
+227 1 0 color 0/1 ood_227_color_object_1_0_color
+228 1 0 shape 1/1 ood_228_color_object_1_0_shape
+229 2 5 color 0/1 ood_229_color_object_2_5_color
+230 2 5 shape 0/1 ood_230_color_object_2_5_shape
+231 2 3 color 0/1 ood_231_color_object_2_3_color
+232 2 3 shape 0/1 ood_232_color_object_2_3_shape
+233 1 4 color 0/1 ood_233_color_object_1_4_color
+234 1 4 shape 0/1 ood_234_color_object_1_4_shape
+235 5 4 color 0/1 ood_235_color_object_5_4_color
+236 5 4 shape 0/1 ood_236_color_object_5_4_shape
+237 4 0 color 0/1 ood_237_color_object_4_0_color
+238 4 0 shape 1/1 ood_238_color_object_4_0_shape
+239 4 2 color 0/1 ood_239_color_object_4_2_color
+240 4 2 shape 0/1 ood_240_color_object_4_2_shape
+241 3 2 color 0/1 ood_241_color_object_3_2_color
+242 3 2 shape 0/1 ood_242_color_object_3_2_shape
+243 1 3 color 0/1 ood_243_color_object_1_3_color
+244 1 3 shape 0/1 ood_244_color_object_1_3_shape
+245 4 1 color 1/1 ood_245_color_object_4_1_color
+246 4 1 shape 0/1 ood_246_color_object_4_1_shape
+247 2 0 color 0/1 ood_247_color_object_2_0_color
+248 2 0 shape 0/1 ood_248_color_object_2_0_shape
+249 5 1 color 0/1 ood_249_color_object_5_1_color
+250 5 1 shape 0/1 ood_250_color_object_5_1_shape
+251 4 5 color 0/1 ood_251_color_object_4_5_color
+252 4 5 shape 0/1 ood_252_color_object_4_5_shape
+253 4 5 color 0/1 ood_253_color_object_4_5_color
+254 4 5 shape 0/1 ood_254_color_object_4_5_shape
+255 0 2 color 0/1 ood_255_color_object_0_2_color
+256 0 2 shape 0/1 ood_256_color_object_0_2_shape
+257 1 3 color 0/1 ood_257_color_object_1_3_color
+258 1 3 shape 0/1 ood_258_color_object_1_3_shape
+259 5 1 color 0/1 ood_259_color_object_5_1_color
+260 5 1 shape 0/1 ood_260_color_object_5_1_shape
+261 0 2 color 1/1 ood_261_color_object_0_2_color
+262 0 2 shape 0/1 ood_262_color_object_0_2_shape
+263 5 0 color 0/1 ood_263_color_object_5_0_color
+264 5 0 shape 0/1 ood_264_color_object_5_0_shape
+265 2 0 color 0/1 ood_265_color_object_2_0_color
+266 2 0 shape 1/1 ood_266_color_object_2_0_shape
+267 2 3 color 0/1 ood_267_color_object_2_3_color
+268 2 3 shape 0/1 ood_268_color_object_2_3_shape
+269 1 4 color 1/1 ood_269_color_object_1_4_color
+270 1 4 shape 0/1 ood_270_color_object_1_4_shape
+271 0 3 color 0/1 ood_271_color_object_0_3_color
+272 0 3 shape 0/1 ood_272_color_object_0_3_shape
+273 1 2 color 0/1 ood_273_color_object_1_2_color
+274 1 2 shape 1/1 ood_274_color_object_1_2_shape
+275 5 4 color 0/1 ood_275_color_object_5_4_color
+276 5 4 shape 0/1 ood_276_color_object_5_4_shape
+277 3 4 color 0/1 ood_277_color_object_3_4_color
+278 3 4 shape 0/1 ood_278_color_object_3_4_shape
+279 5 3 color 0/1 ood_279_color_object_5_3_color
+281 4 2 color 1/1 ood_281_color_object_4_2_color
+282 4 2 shape 0/1 ood_282_color_object_4_2_shape
+283 2 0 color 0/1 ood_283_color_object_2_0_color
+284 2 0 shape 0/1 ood_284_color_object_2_0_shape
+285 1 2 color 1/1 ood_285_color_object_1_2_color
+286 1 2 shape 0/1 ood_286_color_object_1_2_shape
+287 4 0 color 0/1 ood_287_color_object_4_0_color
+288 4 0 shape 0/1 ood_288_color_object_4_0_shape
+289 3 0 color 0/1 ood_289_color_object_3_0_color
+290 3 0 shape 0/1 ood_290_color_object_3_0_shape
+291 2 3 color 0/1 ood_291_color_object_2_3_color
+292 2 3 shape 1/1 ood_292_color_object_2_3_shape
+293 5 3 color 0/1 ood_293_color_object_5_3_color
+294 5 3 shape 0/1 ood_294_color_object_5_3_shape
+295 5 4 color 1/1 ood_295_color_object_5_4_color
+296 5 4 shape 0/1 ood_296_color_object_5_4_shape
+297 4 0 color 0/1 ood_297_color_object_4_0_color
+298 4 0 shape 0/1 ood_298_color_object_4_0_shape
+299 2 5 color 0/1 ood_299_color_object_2_5_color
+300 2 5 shape 0/1 ood_300_color_object_2_5_shape
+301 0 5 color 0/1 ood_301_color_object_0_5_color
+302 0 5 shape 0/1 ood_302_color_object_0_5_shape
+303 1 4 color 0/1 ood_303_color_object_1_4_color
+304 1 4 shape 0/1 ood_304_color_object_1_4_shape
+305 0 5 color 0/1 ood_305_color_object_0_5_color
+306 0 5 shape 0/1 ood_306_color_object_0_5_shape
+307 1 3 color 0/1 ood_307_color_object_1_3_color
+308 1 3 shape 0/1 ood_308_color_object_1_3_shape
+309 4 3 color 0/1 ood_309_color_object_4_3_color
+310 4 3 shape 0/1 ood_310_color_object_4_3_shape
+311 3 2 color 0/1 ood_311_color_object_3_2_color
+312 3 2 shape 0/1 ood_312_color_object_3_2_shape
+313 3 2 color 0/1 ood_313_color_object_3_2_color
+314 3 2 shape 0/1 ood_314_color_object_3_2_shape
+315 1 4 color 0/1 ood_315_color_object_1_4_color
+316 1 4 shape 0/1 ood_316_color_object_1_4_shape
+317 4 3 color 0/1 ood_317_color_object_4_3_color
+318 4 3 shape 0/1 ood_318_color_object_4_3_shape
+319 3 4 color 0/1 ood_319_color_object_3_4_color
+320 3 4 shape 0/1 ood_320_color_object_3_4_shape
+321 2 4 color 0/1 ood_321_color_object_2_4_color
+322 2 4 shape 0/1 ood_322_color_object_2_4_shape
+323 5 3 color 0/1 ood_323_color_object_5_3_color
+324 5 3 shape 0/1 ood_324_color_object_5_3_shape
+325 3 4 color 0/1 ood_325_color_object_3_4_color
+326 3 4 shape 0/1 ood_326_color_object_3_4_shape
+327 2 3 color 0/1 ood_327_color_object_2_3_color
+328 2 3 shape 0/1 ood_328_color_object_2_3_shape
+329 2 1 color 0/1 ood_329_color_object_2_1_color
+330 2 1 shape 0/1 ood_330_color_object_2_1_shape
+331 1 3 color 0/1 ood_331_color_object_1_3_color
+332 1 3 shape 0/1 ood_332_color_object_1_3_shape
+333 0 5 color 0/1 ood_333_color_object_0_5_color
+334 0 5 shape 0/1 ood_334_color_object_0_5_shape
+335 3 1 color 0/1 ood_335_color_object_3_1_color
+336 3 1 shape 0/1 ood_336_color_object_3_1_shape
+337 3 0 color 0/1 ood_337_color_object_3_0_color
+338 3 0 shape 0/1 ood_338_color_object_3_0_shape
+339 0 3 color 0/1 ood_339_color_object_0_3_color
+340 0 3 shape 0/1 ood_340_color_object_0_3_shape
+341 4 5 color 0/1 ood_341_color_object_4_5_color
+342 4 5 shape 0/1 ood_342_color_object_4_5_shape
+343 0 2 color 0/1 ood_343_color_object_0_2_color
+344 0 2 shape 0/1 ood_344_color_object_0_2_shape
+345 5 2 color 0/1 ood_345_color_object_5_2_color
+346 5 2 shape 0/1 ood_346_color_object_5_2_shape
+347 0 4 color 1/1 ood_347_color_object_0_4_color
+348 0 4 shape 0/1 ood_348_color_object_0_4_shape
+349 0 5 color 0/1 ood_349_color_object_0_5_color
+350 0 5 shape 0/1 ood_350_color_object_0_5_shape
+351 4 0 color 0/1 ood_351_color_object_4_0_color
+352 4 0 shape 0/1 ood_352_color_object_4_0_shape
+353 1 0 color 0/1 ood_353_color_object_1_0_color
+354 1 0 shape 1/1 ood_354_color_object_1_0_shape
+355 5 0 color 0/1 ood_355_color_object_5_0_color
+356 5 0 shape 0/1 ood_356_color_object_5_0_shape
+357 4 1 color 0/1 ood_357_color_object_4_1_color
+358 4 1 shape 0/1 ood_358_color_object_4_1_shape
+359 2 4 color 0/1 ood_359_color_object_2_4_color
+360 2 4 shape 0/1 ood_360_color_object_2_4_shape
+361 3 5 color 0/1 ood_361_color_object_3_5_color
+362 3 5 shape 0/1 ood_362_color_object_3_5_shape
+363 0 3 color 0/1 ood_363_color_object_0_3_color
+364 0 3 shape 0/1 ood_364_color_object_0_3_shape
+365 2 3 color 1/1 ood_365_color_object_2_3_color
+366 2 3 shape 0/1 ood_366_color_object_2_3_shape
+367 2 3 color 0/1 ood_367_color_object_2_3_color
+368 2 3 shape 0/1 ood_368_color_object_2_3_shape
+369 3 5 color 1/1 ood_369_color_object_3_5_color
+370 3 5 shape 0/1 ood_370_color_object_3_5_shape
+371 2 5 color 0/1 ood_371_color_object_2_5_color
+372 2 5 shape 0/1 ood_372_color_object_2_5_shape
+373 3 1 color 0/1 ood_373_color_object_3_1_color
+374 3 1 shape 0/1 ood_374_color_object_3_1_shape
+375 1 4 color 0/1 ood_375_color_object_1_4_color
+376 1 4 shape 0/1 ood_376_color_object_1_4_shape
+377 3 2 color 0/1 ood_377_color_object_3_2_color
+378 3 2 shape 0/1 ood_378_color_object_3_2_shape
+379 5 2 color 0/1 ood_379_color_object_5_2_color
+380 5 2 shape 0/1 ood_380_color_object_5_2_shape
+381 0 1 color 0/1 ood_381_color_object_0_1_color
+382 0 1 shape 1/1 ood_382_color_object_0_1_shape
+383 4 1 color 0/1 ood_383_color_object_4_1_color
+384 4 1 shape 0/1 ood_384_color_object_4_1_shape
+385 4 3 color 0/1 ood_385_color_object_4_3_color
+386 4 3 shape 0/1 ood_386_color_object_4_3_shape
+387 0 4 color 0/1 ood_387_color_object_0_4_color
+388 0 4 shape 0/1 ood_388_color_object_0_4_shape
+389 4 1 color 0/1 ood_389_color_object_4_1_color
+390 4 1 shape 0/1 ood_390_color_object_4_1_shape
+391 5 3 color 0/1 ood_391_color_object_5_3_color
+392 5 3 shape 0/1 ood_392_color_object_5_3_shape
+393 3 2 color 1/1 ood_393_color_object_3_2_color
+394 3 2 shape 0/1 ood_394_color_object_3_2_shape
+395 4 5 color 0/1 ood_395_color_object_4_5_color
+396 4 5 shape 0/1 ood_396_color_object_4_5_shape
+397 1 4 color 0/1 ood_397_color_object_1_4_color
+398 1 4 shape 0/1 ood_398_color_object_1_4_shape
+399 4 5 color 0/1 ood_399_color_object_4_5_color
+400 4 5 shape 1/1 ood_400_color_object_4_5_shape
+
+overall_color_success=7/60 (11.7%)
+overall_shape_success=4/60 (6.7%)

results/genie/conflict_env/genie_color_size_ood_seed42.txt

@@ -0,0 +1,421 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=color_size
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/color_size
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 5 4 color 0/1 ood_001_color_size_5_4_color
+2 5 4 size 0/1 ood_002_color_size_5_4_size
+3 0 1 color 0/1 ood_003_color_size_0_1_color
+4 0 1 size 0/1 ood_004_color_size_0_1_size
+5 0 1 color 0/1 ood_005_color_size_0_1_color
+6 0 1 size 0/1 ood_006_color_size_0_1_size
+7 5 4 color 0/1 ood_007_color_size_5_4_color
+8 5 4 size 0/1 ood_008_color_size_5_4_size
+9 2 3 color 0/1 ood_009_color_size_2_3_color
+10 2 3 size 0/1 ood_010_color_size_2_3_size
+11 1 0 color 0/1 ood_011_color_size_1_0_color
+12 1 0 size 0/1 ood_012_color_size_1_0_size
+13 1 0 color 0/1 ood_013_color_size_1_0_color
+14 1 0 size 0/1 ood_014_color_size_1_0_size
+15 1 0 color 0/1 ood_015_color_size_1_0_color
+16 1 0 size 0/1 ood_016_color_size_1_0_size
+17 5 4 color 0/1 ood_017_color_size_5_4_color
+18 5 4 size 0/1 ood_018_color_size_5_4_size
+19 0 1 color 0/1 ood_019_color_size_0_1_color
+20 0 1 size 0/1 ood_020_color_size_0_1_size
+21 5 4 color 0/1 ood_021_color_size_5_4_color
+22 5 4 size 0/1 ood_022_color_size_5_4_size
+23 5 4 color 0/1 ood_023_color_size_5_4_color
+24 5 4 size 0/1 ood_024_color_size_5_4_size
+25 4 5 color 0/1 ood_025_color_size_4_5_color
+26 4 5 size 0/1 ood_026_color_size_4_5_size
+27 0 1 color 0/1 ood_027_color_size_0_1_color
+28 0 1 size 0/1 ood_028_color_size_0_1_size
+29 4 5 color 0/1 ood_029_color_size_4_5_color
+30 4 5 size 0/1 ood_030_color_size_4_5_size
+31 3 2 color 0/1 ood_031_color_size_3_2_color
+32 3 2 size 0/1 ood_032_color_size_3_2_size
+33 0 1 color 0/1 ood_033_color_size_0_1_color
+34 0 1 size 0/1 ood_034_color_size_0_1_size
+35 0 1 color 0/1 ood_035_color_size_0_1_color
+36 0 1 size 0/1 ood_036_color_size_0_1_size
+37 0 1 color 0/1 ood_037_color_size_0_1_color
+38 0 1 size 0/1 ood_038_color_size_0_1_size
+39 1 0 color 0/1 ood_039_color_size_1_0_color
+40 1 0 size 0/1 ood_040_color_size_1_0_size
+41 1 0 color 0/1 ood_041_color_size_1_0_color
+42 1 0 size 0/1 ood_042_color_size_1_0_size
+43 4 5 color 0/1 ood_043_color_size_4_5_color
+44 4 5 size 0/1 ood_044_color_size_4_5_size
+45 4 5 color 0/1 ood_045_color_size_4_5_color
+46 4 5 size 0/1 ood_046_color_size_4_5_size
+47 0 1 color 0/1 ood_047_color_size_0_1_color
+48 0 1 size 0/1 ood_048_color_size_0_1_size
+49 4 5 color 0/1 ood_049_color_size_4_5_color
+50 4 5 size 0/1 ood_050_color_size_4_5_size
+51 1 0 color 0/1 ood_051_color_size_1_0_color
+52 1 0 size 0/1 ood_052_color_size_1_0_size
+53 5 4 color 0/1 ood_053_color_size_5_4_color
+54 5 4 size 0/1 ood_054_color_size_5_4_size
+55 5 4 color 0/1 ood_055_color_size_5_4_color
+56 5 4 size 0/1 ood_056_color_size_5_4_size
+57 5 4 color 0/1 ood_057_color_size_5_4_color
+58 5 4 size 0/1 ood_058_color_size_5_4_size
+59 4 5 color 0/1 ood_059_color_size_4_5_color
+60 4 5 size 0/1 ood_060_color_size_4_5_size
+61 3 2 color 0/1 ood_061_color_size_3_2_color
+62 3 2 size 0/1 ood_062_color_size_3_2_size
+63 1 0 color 0/1 ood_063_color_size_1_0_color
+64 1 0 size 0/1 ood_064_color_size_1_0_size
+65 3 2 color 0/1 ood_065_color_size_3_2_color
+66 3 2 size 0/1 ood_066_color_size_3_2_size
+67 4 5 color 0/1 ood_067_color_size_4_5_color
+68 4 5 size 0/1 ood_068_color_size_4_5_size
+69 2 3 color 0/1 ood_069_color_size_2_3_color
+70 2 3 size 0/1 ood_070_color_size_2_3_size
+71 0 1 color 0/1 ood_071_color_size_0_1_color
+72 0 1 size 0/1 ood_072_color_size_0_1_size
+73 1 0 color 0/1 ood_073_color_size_1_0_color
+74 1 0 size 0/1 ood_074_color_size_1_0_size
+75 5 4 color 0/1 ood_075_color_size_5_4_color
+76 5 4 size 0/1 ood_076_color_size_5_4_size
+77 3 2 color 0/1 ood_077_color_size_3_2_color
+78 3 2 size 0/1 ood_078_color_size_3_2_size
+79 2 3 color 0/1 ood_079_color_size_2_3_color
+80 2 3 size 0/1 ood_080_color_size_2_3_size
+81 2 3 color 0/1 ood_081_color_size_2_3_color
+82 2 3 size 0/1 ood_082_color_size_2_3_size
+83 1 0 color 0/1 ood_083_color_size_1_0_color
+84 1 0 size 0/1 ood_084_color_size_1_0_size
+85 1 0 color 0/1 ood_085_color_size_1_0_color
+86 1 0 size 0/1 ood_086_color_size_1_0_size
+87 2 3 color 1/1 ood_087_color_size_2_3_color
+88 2 3 size 0/1 ood_088_color_size_2_3_size
+89 0 1 color 0/1 ood_089_color_size_0_1_color
+90 0 1 size 0/1 ood_090_color_size_0_1_size
+91 0 1 color 0/1 ood_091_color_size_0_1_color
+92 0 1 size 0/1 ood_092_color_size_0_1_size
+93 3 2 color 0/1 ood_093_color_size_3_2_color
+94 3 2 size 0/1 ood_094_color_size_3_2_size
+95 0 1 color 0/1 ood_095_color_size_0_1_color
+96 0 1 size 0/1 ood_096_color_size_0_1_size
+97 2 3 color 0/1 ood_097_color_size_2_3_color
+98 2 3 size 0/1 ood_098_color_size_2_3_size
+99 2 3 color 1/1 ood_099_color_size_2_3_color
+100 2 3 size 0/1 ood_100_color_size_2_3_size
+101 4 5 color 0/1 ood_101_color_size_4_5_color
+102 4 5 size 0/1 ood_102_color_size_4_5_size
+103 2 3 color 0/1 ood_103_color_size_2_3_color
+104 2 3 size 0/1 ood_104_color_size_2_3_size
+105 0 1 color 0/1 ood_105_color_size_0_1_color
+106 0 1 size 0/1 ood_106_color_size_0_1_size
+107 5 4 color 0/1 ood_107_color_size_5_4_color
+108 5 4 size 0/1 ood_108_color_size_5_4_size
+109 3 2 color 0/1 ood_109_color_size_3_2_color
+110 3 2 size 0/1 ood_110_color_size_3_2_size
+111 4 5 color 0/1 ood_111_color_size_4_5_color
+112 4 5 size 0/1 ood_112_color_size_4_5_size
+113 0 1 color 0/1 ood_113_color_size_0_1_color
+114 0 1 size 0/1 ood_114_color_size_0_1_size
+115 3 2 color 0/1 ood_115_color_size_3_2_color
+116 3 2 size 0/1 ood_116_color_size_3_2_size
+117 0 1 color 0/1 ood_117_color_size_0_1_color
+118 0 1 size 0/1 ood_118_color_size_0_1_size
+119 4 5 color 0/1 ood_119_color_size_4_5_color
+120 4 5 size 0/1 ood_120_color_size_4_5_size
+121 2 3 color 0/1 ood_121_color_size_2_3_color
+122 2 3 size 0/1 ood_122_color_size_2_3_size
+123 5 4 color 0/1 ood_123_color_size_5_4_color
+124 5 4 size 0/1 ood_124_color_size_5_4_size
+125 4 5 color 0/1 ood_125_color_size_4_5_color
+126 4 5 size 0/1 ood_126_color_size_4_5_size
+127 2 3 color 0/1 ood_127_color_size_2_3_color
+128 2 3 size 0/1 ood_128_color_size_2_3_size
+129 4 5 color 0/1 ood_129_color_size_4_5_color
+130 4 5 size 0/1 ood_130_color_size_4_5_size
+131 1 0 color 0/1 ood_131_color_size_1_0_color
+132 1 0 size 0/1 ood_132_color_size_1_0_size
+133 5 4 color 0/1 ood_133_color_size_5_4_color
+134 5 4 size 0/1 ood_134_color_size_5_4_size
+135 0 1 color 0/1 ood_135_color_size_0_1_color
+136 0 1 size 0/1 ood_136_color_size_0_1_size
+137 0 1 color 0/1 ood_137_color_size_0_1_color
+138 0 1 size 0/1 ood_138_color_size_0_1_size
+139 5 4 color 0/1 ood_139_color_size_5_4_color
+140 5 4 size 0/1 ood_140_color_size_5_4_size
+141 1 0 color 0/1 ood_141_color_size_1_0_color
+142 1 0 size 0/1 ood_142_color_size_1_0_size
+143 2 3 color 0/1 ood_143_color_size_2_3_color
+144 2 3 size 0/1 ood_144_color_size_2_3_size
+145 0 1 color 0/1 ood_145_color_size_0_1_color
+146 0 1 size 0/1 ood_146_color_size_0_1_size
+147 1 0 color 0/1 ood_147_color_size_1_0_color
+148 1 0 size 0/1 ood_148_color_size_1_0_size
+149 0 1 color 0/1 ood_149_color_size_0_1_color
+150 0 1 size 0/1 ood_150_color_size_0_1_size
+151 3 2 color 0/1 ood_151_color_size_3_2_color
+152 3 2 size 0/1 ood_152_color_size_3_2_size
+153 2 3 color 0/1 ood_153_color_size_2_3_color
+154 2 3 size 0/1 ood_154_color_size_2_3_size
+155 3 2 color 0/1 ood_155_color_size_3_2_color
+156 3 2 size 0/1 ood_156_color_size_3_2_size
+157 5 4 color 0/1 ood_157_color_size_5_4_color
+158 5 4 size 0/1 ood_158_color_size_5_4_size
+159 2 3 color 0/1 ood_159_color_size_2_3_color
+160 2 3 size 0/1 ood_160_color_size_2_3_size
+161 1 0 color 0/1 ood_161_color_size_1_0_color
+162 1 0 size 0/1 ood_162_color_size_1_0_size
+163 2 3 color 0/1 ood_163_color_size_2_3_color
+164 2 3 size 0/1 ood_164_color_size_2_3_size
+165 2 3 color 0/1 ood_165_color_size_2_3_color
+166 2 3 size 0/1 ood_166_color_size_2_3_size
+167 1 0 color 0/1 ood_167_color_size_1_0_color
+168 1 0 size 0/1 ood_168_color_size_1_0_size
+169 5 4 color 0/1 ood_169_color_size_5_4_color
+170 5 4 size 0/1 ood_170_color_size_5_4_size
+171 2 3 color 0/1 ood_171_color_size_2_3_color
+172 2 3 size 0/1 ood_172_color_size_2_3_size
+173 5 4 color 0/1 ood_173_color_size_5_4_color
+174 5 4 size 0/1 ood_174_color_size_5_4_size
+175 5 4 color 0/1 ood_175_color_size_5_4_color
+176 5 4 size 0/1 ood_176_color_size_5_4_size
+177 5 4 color 0/1 ood_177_color_size_5_4_color
+178 5 4 size 0/1 ood_178_color_size_5_4_size
+179 0 1 color 0/1 ood_179_color_size_0_1_color
+180 0 1 size 0/1 ood_180_color_size_0_1_size
+181 4 5 color 0/1 ood_181_color_size_4_5_color
+182 4 5 size 0/1 ood_182_color_size_4_5_size
+183 5 4 color 0/1 ood_183_color_size_5_4_color
+184 5 4 size 0/1 ood_184_color_size_5_4_size
+185 1 0 color 0/1 ood_185_color_size_1_0_color
+186 1 0 size 0/1 ood_186_color_size_1_0_size
+187 4 5 color 0/1 ood_187_color_size_4_5_color
+188 4 5 size 0/1 ood_188_color_size_4_5_size
+189 5 4 color 0/1 ood_189_color_size_5_4_color
+190 5 4 size 0/1 ood_190_color_size_5_4_size
+191 1 0 color 0/1 ood_191_color_size_1_0_color
+192 1 0 size 0/1 ood_192_color_size_1_0_size
+193 1 0 color 0/1 ood_193_color_size_1_0_color
+194 1 0 size 0/1 ood_194_color_size_1_0_size
+195 3 2 color 0/1 ood_195_color_size_3_2_color
+196 3 2 size 0/1 ood_196_color_size_3_2_size
+197 3 2 color 0/1 ood_197_color_size_3_2_color
+198 3 2 size 0/1 ood_198_color_size_3_2_size
+199 2 3 color 0/1 ood_199_color_size_2_3_color
+200 2 3 size 0/1 ood_200_color_size_2_3_size
+201 5 4 color 0/1 ood_201_color_size_5_4_color
+202 5 4 size 0/1 ood_202_color_size_5_4_size
+203 5 4 color 0/1 ood_203_color_size_5_4_color
+204 5 4 size 0/1 ood_204_color_size_5_4_size
+205 4 5 color 0/1 ood_205_color_size_4_5_color
+206 4 5 size 0/1 ood_206_color_size_4_5_size
+207 1 0 color 0/1 ood_207_color_size_1_0_color
+208 1 0 size 0/1 ood_208_color_size_1_0_size
+209 5 4 color 0/1 ood_209_color_size_5_4_color
+210 5 4 size 0/1 ood_210_color_size_5_4_size
+211 2 3 color 0/1 ood_211_color_size_2_3_color
+212 2 3 size 0/1 ood_212_color_size_2_3_size
+213 0 1 color 0/1 ood_213_color_size_0_1_color
+214 0 1 size 0/1 ood_214_color_size_0_1_size
+215 1 0 color 0/1 ood_215_color_size_1_0_color
+216 1 0 size 0/1 ood_216_color_size_1_0_size
+217 0 1 color 0/1 ood_217_color_size_0_1_color
+218 0 1 size 0/1 ood_218_color_size_0_1_size
+219 2 3 color 0/1 ood_219_color_size_2_3_color
+220 2 3 size 0/1 ood_220_color_size_2_3_size
+221 3 2 color 0/1 ood_221_color_size_3_2_color
+222 3 2 size 0/1 ood_222_color_size_3_2_size
+223 2 3 color 0/1 ood_223_color_size_2_3_color
+224 2 3 size 0/1 ood_224_color_size_2_3_size
+225 0 1 color 0/1 ood_225_color_size_0_1_color
+226 0 1 size 0/1 ood_226_color_size_0_1_size
+227 1 0 color 0/1 ood_227_color_size_1_0_color
+228 1 0 size 0/1 ood_228_color_size_1_0_size
+229 4 5 color 0/1 ood_229_color_size_4_5_color
+230 4 5 size 0/1 ood_230_color_size_4_5_size
+231 5 4 color 0/1 ood_231_color_size_5_4_color
+232 5 4 size 0/1 ood_232_color_size_5_4_size
+233 2 3 color 0/1 ood_233_color_size_2_3_color
+234 2 3 size 0/1 ood_234_color_size_2_3_size
+235 1 0 color 0/1 ood_235_color_size_1_0_color
+236 1 0 size 0/1 ood_236_color_size_1_0_size
+237 5 4 color 0/1 ood_237_color_size_5_4_color
+238 5 4 size 0/1 ood_238_color_size_5_4_size
+239 3 2 color 0/1 ood_239_color_size_3_2_color
+240 3 2 size 0/1 ood_240_color_size_3_2_size
+241 3 2 color 0/1 ood_241_color_size_3_2_color
+242 3 2 size 0/1 ood_242_color_size_3_2_size
+243 5 4 color 0/1 ood_243_color_size_5_4_color
+244 5 4 size 0/1 ood_244_color_size_5_4_size
+245 3 2 color 0/1 ood_245_color_size_3_2_color
+246 3 2 size 0/1 ood_246_color_size_3_2_size
+247 1 0 color 0/1 ood_247_color_size_1_0_color
+248 1 0 size 0/1 ood_248_color_size_1_0_size
+249 2 3 color 0/1 ood_249_color_size_2_3_color
+250 2 3 size 0/1 ood_250_color_size_2_3_size
+251 1 0 color 0/1 ood_251_color_size_1_0_color
+252 1 0 size 0/1 ood_252_color_size_1_0_size
+253 1 0 color 0/1 ood_253_color_size_1_0_color
+254 1 0 size 0/1 ood_254_color_size_1_0_size
+255 5 4 color 0/1 ood_255_color_size_5_4_color
+256 5 4 size 0/1 ood_256_color_size_5_4_size
+257 4 5 color 0/1 ood_257_color_size_4_5_color
+258 4 5 size 0/1 ood_258_color_size_4_5_size
+259 4 5 color 0/1 ood_259_color_size_4_5_color
+260 4 5 size 0/1 ood_260_color_size_4_5_size
+261 2 3 color 0/1 ood_261_color_size_2_3_color
+262 2 3 size 0/1 ood_262_color_size_2_3_size
+263 5 4 color 0/1 ood_263_color_size_5_4_color
+264 5 4 size 0/1 ood_264_color_size_5_4_size
+265 4 5 color 0/1 ood_265_color_size_4_5_color
+266 4 5 size 0/1 ood_266_color_size_4_5_size
+267 3 2 color 0/1 ood_267_color_size_3_2_color
+268 3 2 size 0/1 ood_268_color_size_3_2_size
+269 4 5 color 0/1 ood_269_color_size_4_5_color
+270 4 5 size 0/1 ood_270_color_size_4_5_size
+271 3 2 color 0/1 ood_271_color_size_3_2_color
+272 3 2 size 0/1 ood_272_color_size_3_2_size
+273 2 3 color 0/1 ood_273_color_size_2_3_color
+274 2 3 size 0/1 ood_274_color_size_2_3_size
+275 1 0 color 0/1 ood_275_color_size_1_0_color
+276 1 0 size 0/1 ood_276_color_size_1_0_size
+277 1 0 color 0/1 ood_277_color_size_1_0_color
+278 1 0 size 0/1 ood_278_color_size_1_0_size
+279 4 5 color 0/1 ood_279_color_size_4_5_color
+280 4 5 size 0/1 ood_280_color_size_4_5_size
+281 3 2 color 0/1 ood_281_color_size_3_2_color
+282 3 2 size 0/1 ood_282_color_size_3_2_size
+283 0 1 color 0/1 ood_283_color_size_0_1_color
+284 0 1 size 0/1 ood_284_color_size_0_1_size
+285 0 1 color 0/1 ood_285_color_size_0_1_color
+286 0 1 size 0/1 ood_286_color_size_0_1_size
+287 0 1 color 0/1 ood_287_color_size_0_1_color
+288 0 1 size 0/1 ood_288_color_size_0_1_size
+289 1 0 color 0/1 ood_289_color_size_1_0_color
+290 1 0 size 0/1 ood_290_color_size_1_0_size
+291 5 4 color 0/1 ood_291_color_size_5_4_color
+292 5 4 size 0/1 ood_292_color_size_5_4_size
+293 1 0 color 0/1 ood_293_color_size_1_0_color
+294 1 0 size 0/1 ood_294_color_size_1_0_size
+295 5 4 color 0/1 ood_295_color_size_5_4_color
+296 5 4 size 0/1 ood_296_color_size_5_4_size
+297 3 2 color 0/1 ood_297_color_size_3_2_color
+298 3 2 size 0/1 ood_298_color_size_3_2_size
+299 4 5 color 0/1 ood_299_color_size_4_5_color
+300 4 5 size 0/1 ood_300_color_size_4_5_size
+301 0 1 color 0/1 ood_301_color_size_0_1_color
+302 0 1 size 0/1 ood_302_color_size_0_1_size
+303 3 2 color 0/1 ood_303_color_size_3_2_color
+304 3 2 size 0/1 ood_304_color_size_3_2_size
+305 3 2 color 0/1 ood_305_color_size_3_2_color
+306 3 2 size 0/1 ood_306_color_size_3_2_size
+307 4 5 color 0/1 ood_307_color_size_4_5_color
+308 4 5 size 0/1 ood_308_color_size_4_5_size
+309 3 2 color 0/1 ood_309_color_size_3_2_color
+310 3 2 size 0/1 ood_310_color_size_3_2_size
+311 4 5 color 0/1 ood_311_color_size_4_5_color
+312 4 5 size 0/1 ood_312_color_size_4_5_size
+313 2 3 color 0/1 ood_313_color_size_2_3_color
+314 2 3 size 0/1 ood_314_color_size_2_3_size
+315 4 5 color 0/1 ood_315_color_size_4_5_color
+316 4 5 size 0/1 ood_316_color_size_4_5_size
+317 0 1 color 0/1 ood_317_color_size_0_1_color
+318 0 1 size 0/1 ood_318_color_size_0_1_size
+319 5 4 color 0/1 ood_319_color_size_5_4_color
+320 5 4 size 0/1 ood_320_color_size_5_4_size
+321 5 4 color 0/1 ood_321_color_size_5_4_color
+322 5 4 size 0/1 ood_322_color_size_5_4_size
+323 0 1 color 0/1 ood_323_color_size_0_1_color
+324 0 1 size 0/1 ood_324_color_size_0_1_size
+325 5 4 color 0/1 ood_325_color_size_5_4_color
+326 5 4 size 0/1 ood_326_color_size_5_4_size
+327 4 5 color 0/1 ood_327_color_size_4_5_color
+328 4 5 size 0/1 ood_328_color_size_4_5_size
+329 2 3 color 0/1 ood_329_color_size_2_3_color
+330 2 3 size 0/1 ood_330_color_size_2_3_size
+331 5 4 color 0/1 ood_331_color_size_5_4_color
+332 5 4 size 0/1 ood_332_color_size_5_4_size
+333 2 3 color 0/1 ood_333_color_size_2_3_color
+334 2 3 size 0/1 ood_334_color_size_2_3_size
+335 0 1 color 0/1 ood_335_color_size_0_1_color
+336 0 1 size 0/1 ood_336_color_size_0_1_size
+337 2 3 color 0/1 ood_337_color_size_2_3_color
+338 2 3 size 0/1 ood_338_color_size_2_3_size
+339 3 2 color 0/1 ood_339_color_size_3_2_color
+340 3 2 size 0/1 ood_340_color_size_3_2_size
+341 1 0 color 0/1 ood_341_color_size_1_0_color
+342 1 0 size 0/1 ood_342_color_size_1_0_size
+343 3 2 color 0/1 ood_343_color_size_3_2_color
+344 3 2 size 0/1 ood_344_color_size_3_2_size
+345 0 1 color 0/1 ood_345_color_size_0_1_color
+346 0 1 size 0/1 ood_346_color_size_0_1_size
+347 5 4 color 1/1 ood_347_color_size_5_4_color
+348 5 4 size 0/1 ood_348_color_size_5_4_size
+349 5 4 color 0/1 ood_349_color_size_5_4_color
+350 5 4 size 0/1 ood_350_color_size_5_4_size
+351 2 3 color 0/1 ood_351_color_size_2_3_color
+352 2 3 size 0/1 ood_352_color_size_2_3_size
+353 4 5 color 0/1 ood_353_color_size_4_5_color
+354 4 5 size 0/1 ood_354_color_size_4_5_size
+355 1 0 color 0/1 ood_355_color_size_1_0_color
+356 1 0 size 0/1 ood_356_color_size_1_0_size
+357 4 5 color 0/1 ood_357_color_size_4_5_color
+358 4 5 size 0/1 ood_358_color_size_4_5_size
+359 0 1 color 0/1 ood_359_color_size_0_1_color
+360 0 1 size 0/1 ood_360_color_size_0_1_size
+361 5 4 color 0/1 ood_361_color_size_5_4_color
+362 5 4 size 0/1 ood_362_color_size_5_4_size
+363 2 3 color 0/1 ood_363_color_size_2_3_color
+364 2 3 size 0/1 ood_364_color_size_2_3_size
+365 5 4 color 0/1 ood_365_color_size_5_4_color
+366 5 4 size 0/1 ood_366_color_size_5_4_size
+367 4 5 color 0/1 ood_367_color_size_4_5_color
+368 4 5 size 0/1 ood_368_color_size_4_5_size
+369 4 5 color 0/1 ood_369_color_size_4_5_color
+370 4 5 size 0/1 ood_370_color_size_4_5_size
+371 1 0 color 0/1 ood_371_color_size_1_0_color
+372 1 0 size 0/1 ood_372_color_size_1_0_size
+373 1 0 color 0/1 ood_373_color_size_1_0_color
+374 1 0 size 0/1 ood_374_color_size_1_0_size
+375 2 3 color 0/1 ood_375_color_size_2_3_color
+376 2 3 size 0/1 ood_376_color_size_2_3_size
+377 1 0 color 0/1 ood_377_color_size_1_0_color
+378 1 0 size 0/1 ood_378_color_size_1_0_size
+379 4 5 color 0/1 ood_379_color_size_4_5_color
+380 4 5 size 0/1 ood_380_color_size_4_5_size
+381 4 5 color 0/1 ood_381_color_size_4_5_color
+382 4 5 size 0/1 ood_382_color_size_4_5_size
+383 0 1 color 0/1 ood_383_color_size_0_1_color
+384 0 1 size 0/1 ood_384_color_size_0_1_size
+385 4 5 color 0/1 ood_385_color_size_4_5_color
+386 4 5 size 0/1 ood_386_color_size_4_5_size
+387 2 3 color 0/1 ood_387_color_size_2_3_color
+388 2 3 size 0/1 ood_388_color_size_2_3_size
+389 3 2 color 0/1 ood_389_color_size_3_2_color
+390 3 2 size 0/1 ood_390_color_size_3_2_size
+391 0 1 color 0/1 ood_391_color_size_0_1_color
+392 0 1 size 0/1 ood_392_color_size_0_1_size
+393 0 1 color 0/1 ood_393_color_size_0_1_color
+394 0 1 size 0/1 ood_394_color_size_0_1_size
+395 2 3 color 0/1 ood_395_color_size_2_3_color
+396 2 3 size 0/1 ood_396_color_size_2_3_size
+397 2 3 color 0/1 ood_397_color_size_2_3_color
+398 2 3 size 0/1 ood_398_color_size_2_3_size
+399 1 0 color 0/1 ood_399_color_size_1_0_color
+400 1 0 size 0/1 ood_400_color_size_1_0_size
+
+overall_color_success=3/200 (1.5%)
+overall_size_success=0/200 (0.0%)

results/genie/conflict_env/genie_color_spatial_ood_seed42.txt

@@ -0,0 +1,421 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=color_spatial
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/color_spatial
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 0 4 color 0/1 ood_001_color_spatial_0_4_color
+2 0 4 spatial 0/1 ood_002_color_spatial_0_4_spatial
+3 0 1 color 0/1 ood_003_color_spatial_0_1_color
+4 0 1 spatial 1/1 ood_004_color_spatial_0_1_spatial
+5 2 0 color 0/1 ood_005_color_spatial_2_0_color
+6 2 0 spatial 0/1 ood_006_color_spatial_2_0_spatial
+7 1 4 color 0/1 ood_007_color_spatial_1_4_color
+8 1 4 spatial 0/1 ood_008_color_spatial_1_4_spatial
+9 1 4 color 1/1 ood_009_color_spatial_1_4_color
+10 1 4 spatial 0/1 ood_010_color_spatial_1_4_spatial
+11 1 0 color 0/1 ood_011_color_spatial_1_0_color
+12 1 0 spatial 0/1 ood_012_color_spatial_1_0_spatial
+13 0 4 color 0/1 ood_013_color_spatial_0_4_color
+14 0 4 spatial 0/1 ood_014_color_spatial_0_4_spatial
+15 4 1 color 0/1 ood_015_color_spatial_4_1_color
+16 4 1 spatial 1/1 ood_016_color_spatial_4_1_spatial
+17 0 3 color 0/1 ood_017_color_spatial_0_3_color
+18 0 3 spatial 1/1 ood_018_color_spatial_0_3_spatial
+19 4 2 color 0/1 ood_019_color_spatial_4_2_color
+20 4 2 spatial 0/1 ood_020_color_spatial_4_2_spatial
+21 3 1 color 0/1 ood_021_color_spatial_3_1_color
+22 3 1 spatial 0/1 ood_022_color_spatial_3_1_spatial
+23 0 2 color 0/1 ood_023_color_spatial_0_2_color
+24 0 2 spatial 0/1 ood_024_color_spatial_0_2_spatial
+25 0 1 color 0/1 ood_025_color_spatial_0_1_color
+26 0 1 spatial 0/1 ood_026_color_spatial_0_1_spatial
+27 0 3 color 0/1 ood_027_color_spatial_0_3_color
+28 0 3 spatial 1/1 ood_028_color_spatial_0_3_spatial
+29 1 3 color 0/1 ood_029_color_spatial_1_3_color
+30 1 3 spatial 1/1 ood_030_color_spatial_1_3_spatial
+31 1 4 color 0/1 ood_031_color_spatial_1_4_color
+32 1 4 spatial 0/1 ood_032_color_spatial_1_4_spatial
+33 4 0 color 0/1 ood_033_color_spatial_4_0_color
+34 4 0 spatial 0/1 ood_034_color_spatial_4_0_spatial
+35 4 3 color 0/1 ood_035_color_spatial_4_3_color
+36 4 3 spatial 0/1 ood_036_color_spatial_4_3_spatial
+37 0 1 color 0/1 ood_037_color_spatial_0_1_color
+38 0 1 spatial 1/1 ood_038_color_spatial_0_1_spatial
+39 4 1 color 0/1 ood_039_color_spatial_4_1_color
+40 4 1 spatial 1/1 ood_040_color_spatial_4_1_spatial
+41 1 3 color 0/1 ood_041_color_spatial_1_3_color
+42 1 3 spatial 1/1 ood_042_color_spatial_1_3_spatial
+43 4 1 color 0/1 ood_043_color_spatial_4_1_color
+44 4 1 spatial 1/1 ood_044_color_spatial_4_1_spatial
+45 3 1 color 0/1 ood_045_color_spatial_3_1_color
+46 3 1 spatial 0/1 ood_046_color_spatial_3_1_spatial
+47 1 4 color 1/1 ood_047_color_spatial_1_4_color
+48 1 4 spatial 0/1 ood_048_color_spatial_1_4_spatial
+49 3 2 color 0/1 ood_049_color_spatial_3_2_color
+50 3 2 spatial 0/1 ood_050_color_spatial_3_2_spatial
+51 4 2 color 0/1 ood_051_color_spatial_4_2_color
+52 4 2 spatial 0/1 ood_052_color_spatial_4_2_spatial
+53 2 0 color 0/1 ood_053_color_spatial_2_0_color
+54 2 0 spatial 0/1 ood_054_color_spatial_2_0_spatial
+55 0 1 color 0/1 ood_055_color_spatial_0_1_color
+56 0 1 spatial 1/1 ood_056_color_spatial_0_1_spatial
+57 1 2 color 0/1 ood_057_color_spatial_1_2_color
+58 1 2 spatial 1/1 ood_058_color_spatial_1_2_spatial
+59 3 1 color 0/1 ood_059_color_spatial_3_1_color
+60 3 1 spatial 0/1 ood_060_color_spatial_3_1_spatial
+61 2 3 color 0/1 ood_061_color_spatial_2_3_color
+62 2 3 spatial 0/1 ood_062_color_spatial_2_3_spatial
+63 2 0 color 0/1 ood_063_color_spatial_2_0_color
+64 2 0 spatial 0/1 ood_064_color_spatial_2_0_spatial
+65 1 0 color 0/1 ood_065_color_spatial_1_0_color
+66 1 0 spatial 1/1 ood_066_color_spatial_1_0_spatial
+67 1 3 color 0/1 ood_067_color_spatial_1_3_color
+68 1 3 spatial 0/1 ood_068_color_spatial_1_3_spatial
+69 2 3 color 0/1 ood_069_color_spatial_2_3_color
+70 2 3 spatial 0/1 ood_070_color_spatial_2_3_spatial
+71 0 4 color 0/1 ood_071_color_spatial_0_4_color
+72 0 4 spatial 0/1 ood_072_color_spatial_0_4_spatial
+73 0 3 color 0/1 ood_073_color_spatial_0_3_color
+74 0 3 spatial 0/1 ood_074_color_spatial_0_3_spatial
+75 3 0 color 0/1 ood_075_color_spatial_3_0_color
+76 3 0 spatial 0/1 ood_076_color_spatial_3_0_spatial
+77 0 4 color 0/1 ood_077_color_spatial_0_4_color
+78 0 4 spatial 0/1 ood_078_color_spatial_0_4_spatial
+79 2 4 color 0/1 ood_079_color_spatial_2_4_color
+80 2 4 spatial 1/1 ood_080_color_spatial_2_4_spatial
+81 2 4 color 0/1 ood_081_color_spatial_2_4_color
+82 2 4 spatial 0/1 ood_082_color_spatial_2_4_spatial
+83 4 3 color 0/1 ood_083_color_spatial_4_3_color
+84 4 3 spatial 1/1 ood_084_color_spatial_4_3_spatial
+85 2 0 color 0/1 ood_085_color_spatial_2_0_color
+86 2 0 spatial 1/1 ood_086_color_spatial_2_0_spatial
+87 0 2 color 0/1 ood_087_color_spatial_0_2_color
+88 0 2 spatial 0/1 ood_088_color_spatial_0_2_spatial
+89 3 2 color 0/1 ood_089_color_spatial_3_2_color
+90 3 2 spatial 0/1 ood_090_color_spatial_3_2_spatial
+91 4 1 color 0/1 ood_091_color_spatial_4_1_color
+92 4 1 spatial 1/1 ood_092_color_spatial_4_1_spatial
+93 0 4 color 0/1 ood_093_color_spatial_0_4_color
+94 0 4 spatial 0/1 ood_094_color_spatial_0_4_spatial
+95 3 0 color 0/1 ood_095_color_spatial_3_0_color
+96 3 0 spatial 1/1 ood_096_color_spatial_3_0_spatial
+97 0 3 color 0/1 ood_097_color_spatial_0_3_color
+98 0 3 spatial 1/1 ood_098_color_spatial_0_3_spatial
+99 4 1 color 0/1 ood_099_color_spatial_4_1_color
+100 4 1 spatial 1/1 ood_100_color_spatial_4_1_spatial
+101 2 1 color 0/1 ood_101_color_spatial_2_1_color
+102 2 1 spatial 1/1 ood_102_color_spatial_2_1_spatial
+103 4 3 color 0/1 ood_103_color_spatial_4_3_color
+104 4 3 spatial 0/1 ood_104_color_spatial_4_3_spatial
+105 2 4 color 0/1 ood_105_color_spatial_2_4_color
+106 2 4 spatial 1/1 ood_106_color_spatial_2_4_spatial
+107 4 2 color 0/1 ood_107_color_spatial_4_2_color
+108 4 2 spatial 0/1 ood_108_color_spatial_4_2_spatial
+109 1 3 color 0/1 ood_109_color_spatial_1_3_color
+110 1 3 spatial 0/1 ood_110_color_spatial_1_3_spatial
+111 0 3 color 0/1 ood_111_color_spatial_0_3_color
+112 0 3 spatial 0/1 ood_112_color_spatial_0_3_spatial
+113 0 2 color 0/1 ood_113_color_spatial_0_2_color
+114 0 2 spatial 1/1 ood_114_color_spatial_0_2_spatial
+115 1 4 color 1/1 ood_115_color_spatial_1_4_color
+116 1 4 spatial 1/1 ood_116_color_spatial_1_4_spatial
+117 2 1 color 0/1 ood_117_color_spatial_2_1_color
+118 2 1 spatial 1/1 ood_118_color_spatial_2_1_spatial
+119 0 3 color 0/1 ood_119_color_spatial_0_3_color
+120 0 3 spatial 0/1 ood_120_color_spatial_0_3_spatial
+121 1 4 color 1/1 ood_121_color_spatial_1_4_color
+122 1 4 spatial 0/1 ood_122_color_spatial_1_4_spatial
+123 0 4 color 0/1 ood_123_color_spatial_0_4_color
+124 0 4 spatial 0/1 ood_124_color_spatial_0_4_spatial
+125 3 0 color 0/1 ood_125_color_spatial_3_0_color
+126 3 0 spatial 0/1 ood_126_color_spatial_3_0_spatial
+127 2 0 color 0/1 ood_127_color_spatial_2_0_color
+128 2 0 spatial 1/1 ood_128_color_spatial_2_0_spatial
+129 3 2 color 0/1 ood_129_color_spatial_3_2_color
+130 3 2 spatial 0/1 ood_130_color_spatial_3_2_spatial
+131 2 4 color 0/1 ood_131_color_spatial_2_4_color
+132 2 4 spatial 1/1 ood_132_color_spatial_2_4_spatial
+133 1 2 color 1/1 ood_133_color_spatial_1_2_color
+134 1 2 spatial 0/1 ood_134_color_spatial_1_2_spatial
+135 2 4 color 0/1 ood_135_color_spatial_2_4_color
+136 2 4 spatial 1/1 ood_136_color_spatial_2_4_spatial
+137 2 4 color 0/1 ood_137_color_spatial_2_4_color
+138 2 4 spatial 1/1 ood_138_color_spatial_2_4_spatial
+139 1 3 color 0/1 ood_139_color_spatial_1_3_color
+140 1 3 spatial 0/1 ood_140_color_spatial_1_3_spatial
+141 2 0 color 0/1 ood_141_color_spatial_2_0_color
+142 2 0 spatial 0/1 ood_142_color_spatial_2_0_spatial
+143 0 3 color 0/1 ood_143_color_spatial_0_3_color
+144 0 3 spatial 0/1 ood_144_color_spatial_0_3_spatial
+145 4 3 color 0/1 ood_145_color_spatial_4_3_color
+146 4 3 spatial 0/1 ood_146_color_spatial_4_3_spatial
+147 1 2 color 0/1 ood_147_color_spatial_1_2_color
+148 1 2 spatial 0/1 ood_148_color_spatial_1_2_spatial
+149 4 1 color 0/1 ood_149_color_spatial_4_1_color
+150 4 1 spatial 1/1 ood_150_color_spatial_4_1_spatial
+151 1 4 color 0/1 ood_151_color_spatial_1_4_color
+152 1 4 spatial 0/1 ood_152_color_spatial_1_4_spatial
+153 1 2 color 0/1 ood_153_color_spatial_1_2_color
+154 1 2 spatial 0/1 ood_154_color_spatial_1_2_spatial
+155 3 2 color 1/1 ood_155_color_spatial_3_2_color
+156 3 2 spatial 0/1 ood_156_color_spatial_3_2_spatial
+157 3 0 color 0/1 ood_157_color_spatial_3_0_color
+158 3 0 spatial 0/1 ood_158_color_spatial_3_0_spatial
+159 2 0 color 0/1 ood_159_color_spatial_2_0_color
+160 2 0 spatial 1/1 ood_160_color_spatial_2_0_spatial
+161 4 1 color 0/1 ood_161_color_spatial_4_1_color
+162 4 1 spatial 1/1 ood_162_color_spatial_4_1_spatial
+163 1 4 color 1/1 ood_163_color_spatial_1_4_color
+164 1 4 spatial 0/1 ood_164_color_spatial_1_4_spatial
+165 2 3 color 0/1 ood_165_color_spatial_2_3_color
+166 2 3 spatial 0/1 ood_166_color_spatial_2_3_spatial
+167 0 2 color 1/1 ood_167_color_spatial_0_2_color
+168 0 2 spatial 0/1 ood_168_color_spatial_0_2_spatial
+169 1 4 color 0/1 ood_169_color_spatial_1_4_color
+170 1 4 spatial 0/1 ood_170_color_spatial_1_4_spatial
+171 0 2 color 0/1 ood_171_color_spatial_0_2_color
+172 0 2 spatial 0/1 ood_172_color_spatial_0_2_spatial
+173 2 3 color 0/1 ood_173_color_spatial_2_3_color
+174 2 3 spatial 0/1 ood_174_color_spatial_2_3_spatial
+175 3 0 color 0/1 ood_175_color_spatial_3_0_color
+176 3 0 spatial 0/1 ood_176_color_spatial_3_0_spatial
+177 2 0 color 0/1 ood_177_color_spatial_2_0_color
+178 2 0 spatial 1/1 ood_178_color_spatial_2_0_spatial
+179 0 3 color 0/1 ood_179_color_spatial_0_3_color
+180 0 3 spatial 0/1 ood_180_color_spatial_0_3_spatial
+181 1 3 color 0/1 ood_181_color_spatial_1_3_color
+182 1 3 spatial 0/1 ood_182_color_spatial_1_3_spatial
+183 4 2 color 1/1 ood_183_color_spatial_4_2_color
+184 4 2 spatial 0/1 ood_184_color_spatial_4_2_spatial
+185 2 3 color 0/1 ood_185_color_spatial_2_3_color
+186 2 3 spatial 1/1 ood_186_color_spatial_2_3_spatial
+187 1 3 color 0/1 ood_187_color_spatial_1_3_color
+188 1 3 spatial 0/1 ood_188_color_spatial_1_3_spatial
+189 3 4 color 0/1 ood_189_color_spatial_3_4_color
+190 3 4 spatial 1/1 ood_190_color_spatial_3_4_spatial
+191 3 0 color 0/1 ood_191_color_spatial_3_0_color
+192 3 0 spatial 0/1 ood_192_color_spatial_3_0_spatial
+193 3 2 color 1/1 ood_193_color_spatial_3_2_color
+194 3 2 spatial 0/1 ood_194_color_spatial_3_2_spatial
+195 1 0 color 0/1 ood_195_color_spatial_1_0_color
+196 1 0 spatial 1/1 ood_196_color_spatial_1_0_spatial
+197 2 0 color 0/1 ood_197_color_spatial_2_0_color
+198 2 0 spatial 0/1 ood_198_color_spatial_2_0_spatial
+199 1 0 color 0/1 ood_199_color_spatial_1_0_color
+200 1 0 spatial 1/1 ood_200_color_spatial_1_0_spatial
+201 1 4 color 1/1 ood_201_color_spatial_1_4_color
+202 1 4 spatial 0/1 ood_202_color_spatial_1_4_spatial
+203 4 1 color 0/1 ood_203_color_spatial_4_1_color
+204 4 1 spatial 1/1 ood_204_color_spatial_4_1_spatial
+205 4 1 color 0/1 ood_205_color_spatial_4_1_color
+206 4 1 spatial 1/1 ood_206_color_spatial_4_1_spatial
+207 2 0 color 0/1 ood_207_color_spatial_2_0_color
+208 2 0 spatial 1/1 ood_208_color_spatial_2_0_spatial
+209 4 2 color 0/1 ood_209_color_spatial_4_2_color
+210 4 2 spatial 0/1 ood_210_color_spatial_4_2_spatial
+211 3 1 color 0/1 ood_211_color_spatial_3_1_color
+212 3 1 spatial 0/1 ood_212_color_spatial_3_1_spatial
+213 4 2 color 0/1 ood_213_color_spatial_4_2_color
+214 4 2 spatial 0/1 ood_214_color_spatial_4_2_spatial
+215 3 0 color 0/1 ood_215_color_spatial_3_0_color
+216 3 0 spatial 0/1 ood_216_color_spatial_3_0_spatial
+217 2 4 color 0/1 ood_217_color_spatial_2_4_color
+218 2 4 spatial 0/1 ood_218_color_spatial_2_4_spatial
+219 1 4 color 1/1 ood_219_color_spatial_1_4_color
+220 1 4 spatial 0/1 ood_220_color_spatial_1_4_spatial
+221 1 0 color 0/1 ood_221_color_spatial_1_0_color
+222 1 0 spatial 0/1 ood_222_color_spatial_1_0_spatial
+223 4 0 color 0/1 ood_223_color_spatial_4_0_color
+224 4 0 spatial 1/1 ood_224_color_spatial_4_0_spatial
+225 3 4 color 0/1 ood_225_color_spatial_3_4_color
+226 3 4 spatial 0/1 ood_226_color_spatial_3_4_spatial
+227 0 3 color 0/1 ood_227_color_spatial_0_3_color
+228 0 3 spatial 1/1 ood_228_color_spatial_0_3_spatial
+229 0 2 color 1/1 ood_229_color_spatial_0_2_color
+230 0 2 spatial 0/1 ood_230_color_spatial_0_2_spatial
+231 0 4 color 0/1 ood_231_color_spatial_0_4_color
+232 0 4 spatial 0/1 ood_232_color_spatial_0_4_spatial
+233 1 0 color 0/1 ood_233_color_spatial_1_0_color
+234 1 0 spatial 1/1 ood_234_color_spatial_1_0_spatial
+235 1 2 color 0/1 ood_235_color_spatial_1_2_color
+236 1 2 spatial 0/1 ood_236_color_spatial_1_2_spatial
+237 3 1 color 0/1 ood_237_color_spatial_3_1_color
+238 3 1 spatial 0/1 ood_238_color_spatial_3_1_spatial
+239 4 3 color 0/1 ood_239_color_spatial_4_3_color
+240 4 3 spatial 1/1 ood_240_color_spatial_4_3_spatial
+241 0 3 color 0/1 ood_241_color_spatial_0_3_color
+242 0 3 spatial 1/1 ood_242_color_spatial_0_3_spatial
+243 3 0 color 1/1 ood_243_color_spatial_3_0_color
+244 3 0 spatial 1/1 ood_244_color_spatial_3_0_spatial
+245 3 0 color 0/1 ood_245_color_spatial_3_0_color
+246 3 0 spatial 1/1 ood_246_color_spatial_3_0_spatial
+247 4 3 color 0/1 ood_247_color_spatial_4_3_color
+248 4 3 spatial 1/1 ood_248_color_spatial_4_3_spatial
+249 3 2 color 1/1 ood_249_color_spatial_3_2_color
+250 3 2 spatial 0/1 ood_250_color_spatial_3_2_spatial
+251 4 0 color 0/1 ood_251_color_spatial_4_0_color
+252 4 0 spatial 0/1 ood_252_color_spatial_4_0_spatial
+253 2 0 color 0/1 ood_253_color_spatial_2_0_color
+254 2 0 spatial 1/1 ood_254_color_spatial_2_0_spatial
+255 4 1 color 0/1 ood_255_color_spatial_4_1_color
+256 4 1 spatial 1/1 ood_256_color_spatial_4_1_spatial
+257 0 1 color 0/1 ood_257_color_spatial_0_1_color
+258 0 1 spatial 0/1 ood_258_color_spatial_0_1_spatial
+259 0 4 color 0/1 ood_259_color_spatial_0_4_color
+260 0 4 spatial 0/1 ood_260_color_spatial_0_4_spatial
+261 4 1 color 0/1 ood_261_color_spatial_4_1_color
+262 4 1 spatial 1/1 ood_262_color_spatial_4_1_spatial
+263 2 0 color 0/1 ood_263_color_spatial_2_0_color
+264 2 0 spatial 1/1 ood_264_color_spatial_2_0_spatial
+265 2 3 color 0/1 ood_265_color_spatial_2_3_color
+266 2 3 spatial 1/1 ood_266_color_spatial_2_3_spatial
+267 0 4 color 0/1 ood_267_color_spatial_0_4_color
+268 0 4 spatial 0/1 ood_268_color_spatial_0_4_spatial
+269 2 1 color 0/1 ood_269_color_spatial_2_1_color
+270 2 1 spatial 1/1 ood_270_color_spatial_2_1_spatial
+271 3 1 color 0/1 ood_271_color_spatial_3_1_color
+272 3 1 spatial 0/1 ood_272_color_spatial_3_1_spatial
+273 1 2 color 0/1 ood_273_color_spatial_1_2_color
+274 1 2 spatial 0/1 ood_274_color_spatial_1_2_spatial
+275 3 2 color 1/1 ood_275_color_spatial_3_2_color
+276 3 2 spatial 0/1 ood_276_color_spatial_3_2_spatial
+277 0 1 color 0/1 ood_277_color_spatial_0_1_color
+278 0 1 spatial 0/1 ood_278_color_spatial_0_1_spatial
+279 2 0 color 0/1 ood_279_color_spatial_2_0_color
+280 2 0 spatial 0/1 ood_280_color_spatial_2_0_spatial
+281 4 0 color 0/1 ood_281_color_spatial_4_0_color
+282 4 0 spatial 1/1 ood_282_color_spatial_4_0_spatial
+283 1 2 color 0/1 ood_283_color_spatial_1_2_color
+284 1 2 spatial 0/1 ood_284_color_spatial_1_2_spatial
+285 4 0 color 0/1 ood_285_color_spatial_4_0_color
+286 4 0 spatial 0/1 ood_286_color_spatial_4_0_spatial
+287 0 4 color 0/1 ood_287_color_spatial_0_4_color
+288 0 4 spatial 0/1 ood_288_color_spatial_0_4_spatial
+289 2 1 color 0/1 ood_289_color_spatial_2_1_color
+290 2 1 spatial 1/1 ood_290_color_spatial_2_1_spatial
+291 4 0 color 0/1 ood_291_color_spatial_4_0_color
+292 4 0 spatial 0/1 ood_292_color_spatial_4_0_spatial
+293 4 3 color 0/1 ood_293_color_spatial_4_3_color
+294 4 3 spatial 0/1 ood_294_color_spatial_4_3_spatial
+295 1 3 color 0/1 ood_295_color_spatial_1_3_color
+296 1 3 spatial 0/1 ood_296_color_spatial_1_3_spatial
+297 1 0 color 0/1 ood_297_color_spatial_1_0_color
+298 1 0 spatial 1/1 ood_298_color_spatial_1_0_spatial
+299 2 4 color 0/1 ood_299_color_spatial_2_4_color
+300 2 4 spatial 1/1 ood_300_color_spatial_2_4_spatial
+301 1 2 color 0/1 ood_301_color_spatial_1_2_color
+302 1 2 spatial 0/1 ood_302_color_spatial_1_2_spatial
+303 4 1 color 0/1 ood_303_color_spatial_4_1_color
+304 4 1 spatial 1/1 ood_304_color_spatial_4_1_spatial
+305 4 0 color 0/1 ood_305_color_spatial_4_0_color
+306 4 0 spatial 0/1 ood_306_color_spatial_4_0_spatial
+307 0 1 color 0/1 ood_307_color_spatial_0_1_color
+308 0 1 spatial 1/1 ood_308_color_spatial_0_1_spatial
+309 4 3 color 0/1 ood_309_color_spatial_4_3_color
+310 4 3 spatial 0/1 ood_310_color_spatial_4_3_spatial
+311 2 3 color 0/1 ood_311_color_spatial_2_3_color
+312 2 3 spatial 1/1 ood_312_color_spatial_2_3_spatial
+313 3 4 color 0/1 ood_313_color_spatial_3_4_color
+314 3 4 spatial 0/1 ood_314_color_spatial_3_4_spatial
+315 0 1 color 0/1 ood_315_color_spatial_0_1_color
+316 0 1 spatial 0/1 ood_316_color_spatial_0_1_spatial
+317 0 4 color 0/1 ood_317_color_spatial_0_4_color
+318 0 4 spatial 0/1 ood_318_color_spatial_0_4_spatial
+319 2 4 color 0/1 ood_319_color_spatial_2_4_color
+320 2 4 spatial 0/1 ood_320_color_spatial_2_4_spatial
+321 2 1 color 0/1 ood_321_color_spatial_2_1_color
+322 2 1 spatial 1/1 ood_322_color_spatial_2_1_spatial
+323 1 4 color 0/1 ood_323_color_spatial_1_4_color
+324 1 4 spatial 0/1 ood_324_color_spatial_1_4_spatial
+325 0 2 color 0/1 ood_325_color_spatial_0_2_color
+326 0 2 spatial 1/1 ood_326_color_spatial_0_2_spatial
+327 1 4 color 0/1 ood_327_color_spatial_1_4_color
+328 1 4 spatial 0/1 ood_328_color_spatial_1_4_spatial
+329 4 2 color 0/1 ood_329_color_spatial_4_2_color
+330 4 2 spatial 0/1 ood_330_color_spatial_4_2_spatial
+331 0 3 color 0/1 ood_331_color_spatial_0_3_color
+332 0 3 spatial 0/1 ood_332_color_spatial_0_3_spatial
+333 0 3 color 0/1 ood_333_color_spatial_0_3_color
+334 0 3 spatial 0/1 ood_334_color_spatial_0_3_spatial
+335 3 4 color 0/1 ood_335_color_spatial_3_4_color
+336 3 4 spatial 1/1 ood_336_color_spatial_3_4_spatial
+337 0 3 color 1/1 ood_337_color_spatial_0_3_color
+338 0 3 spatial 1/1 ood_338_color_spatial_0_3_spatial
+339 4 1 color 0/1 ood_339_color_spatial_4_1_color
+340 4 1 spatial 1/1 ood_340_color_spatial_4_1_spatial
+341 1 0 color 1/1 ood_341_color_spatial_1_0_color
+342 1 0 spatial 0/1 ood_342_color_spatial_1_0_spatial
+343 1 0 color 0/1 ood_343_color_spatial_1_0_color
+344 1 0 spatial 1/1 ood_344_color_spatial_1_0_spatial
+345 3 4 color 0/1 ood_345_color_spatial_3_4_color
+346 3 4 spatial 1/1 ood_346_color_spatial_3_4_spatial
+347 4 1 color 0/1 ood_347_color_spatial_4_1_color
+348 4 1 spatial 1/1 ood_348_color_spatial_4_1_spatial
+349 1 2 color 0/1 ood_349_color_spatial_1_2_color
+350 1 2 spatial 0/1 ood_350_color_spatial_1_2_spatial
+351 2 0 color 0/1 ood_351_color_spatial_2_0_color
+352 2 0 spatial 1/1 ood_352_color_spatial_2_0_spatial
+353 4 0 color 0/1 ood_353_color_spatial_4_0_color
+354 4 0 spatial 0/1 ood_354_color_spatial_4_0_spatial
+355 4 3 color 0/1 ood_355_color_spatial_4_3_color
+356 4 3 spatial 0/1 ood_356_color_spatial_4_3_spatial
+357 3 1 color 0/1 ood_357_color_spatial_3_1_color
+358 3 1 spatial 0/1 ood_358_color_spatial_3_1_spatial
+359 1 3 color 0/1 ood_359_color_spatial_1_3_color
+360 1 3 spatial 0/1 ood_360_color_spatial_1_3_spatial
+361 4 1 color 0/1 ood_361_color_spatial_4_1_color
+362 4 1 spatial 1/1 ood_362_color_spatial_4_1_spatial
+363 1 3 color 0/1 ood_363_color_spatial_1_3_color
+364 1 3 spatial 0/1 ood_364_color_spatial_1_3_spatial
+365 2 1 color 0/1 ood_365_color_spatial_2_1_color
+366 2 1 spatial 1/1 ood_366_color_spatial_2_1_spatial
+367 3 0 color 0/1 ood_367_color_spatial_3_0_color
+368 3 0 spatial 1/1 ood_368_color_spatial_3_0_spatial
+369 2 4 color 0/1 ood_369_color_spatial_2_4_color
+370 2 4 spatial 1/1 ood_370_color_spatial_2_4_spatial
+371 3 2 color 0/1 ood_371_color_spatial_3_2_color
+372 3 2 spatial 0/1 ood_372_color_spatial_3_2_spatial
+373 4 0 color 0/1 ood_373_color_spatial_4_0_color
+374 4 0 spatial 1/1 ood_374_color_spatial_4_0_spatial
+375 3 2 color 1/1 ood_375_color_spatial_3_2_color
+376 3 2 spatial 0/1 ood_376_color_spatial_3_2_spatial
+377 0 4 color 0/1 ood_377_color_spatial_0_4_color
+378 0 4 spatial 0/1 ood_378_color_spatial_0_4_spatial
+379 1 4 color 0/1 ood_379_color_spatial_1_4_color
+380 1 4 spatial 0/1 ood_380_color_spatial_1_4_spatial
+381 1 4 color 1/1 ood_381_color_spatial_1_4_color
+382 1 4 spatial 0/1 ood_382_color_spatial_1_4_spatial
+383 0 3 color 0/1 ood_383_color_spatial_0_3_color
+384 0 3 spatial 0/1 ood_384_color_spatial_0_3_spatial
+385 2 3 color 0/1 ood_385_color_spatial_2_3_color
+386 2 3 spatial 0/1 ood_386_color_spatial_2_3_spatial
+387 0 1 color 0/1 ood_387_color_spatial_0_1_color
+388 0 1 spatial 1/1 ood_388_color_spatial_0_1_spatial
+389 4 2 color 0/1 ood_389_color_spatial_4_2_color
+390 4 2 spatial 0/1 ood_390_color_spatial_4_2_spatial
+391 4 1 color 0/1 ood_391_color_spatial_4_1_color
+392 4 1 spatial 1/1 ood_392_color_spatial_4_1_spatial
+393 1 4 color 1/1 ood_393_color_spatial_1_4_color
+394 1 4 spatial 0/1 ood_394_color_spatial_1_4_spatial
+395 4 2 color 0/1 ood_395_color_spatial_4_2_color
+396 4 2 spatial 0/1 ood_396_color_spatial_4_2_spatial
+397 1 4 color 0/1 ood_397_color_spatial_1_4_color
+398 1 4 spatial 0/1 ood_398_color_spatial_1_4_spatial
+399 0 1 color 0/1 ood_399_color_spatial_0_1_color
+400 0 1 spatial 1/1 ood_400_color_spatial_0_1_spatial
+
+overall_color_success=21/200 (10.5%)
+overall_spatial_success=77/200 (38.5%)

results/genie/conflict_env/genie_size_object_ood_seed42.txt

@@ -0,0 +1,420 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=size_object
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/size_object
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 5 4 size 0/1 ood_001_size_object_5_4_size
+2 5 4 shape 0/1 ood_002_size_object_5_4_shape
+3 0 1 size 0/1 ood_003_size_object_0_1_size
+4 0 1 shape 0/1 ood_004_size_object_0_1_shape
+5 0 1 size 0/1 ood_005_size_object_0_1_size
+6 0 1 shape 0/1 ood_006_size_object_0_1_shape
+7 5 4 size 0/1 ood_007_size_object_5_4_size
+8 5 4 shape 0/1 ood_008_size_object_5_4_shape
+9 2 3 size 0/1 ood_009_size_object_2_3_size
+10 2 3 shape 0/1 ood_010_size_object_2_3_shape
+11 1 0 size 0/1 ood_011_size_object_1_0_size
+12 1 0 shape 0/1 ood_012_size_object_1_0_shape
+13 1 0 size 0/1 ood_013_size_object_1_0_size
+14 1 0 shape 0/1 ood_014_size_object_1_0_shape
+15 1 0 size 0/1 ood_015_size_object_1_0_size
+16 1 0 shape 0/1 ood_016_size_object_1_0_shape
+17 5 4 size 0/1 ood_017_size_object_5_4_size
+18 5 4 shape 0/1 ood_018_size_object_5_4_shape
+19 0 1 size 0/1 ood_019_size_object_0_1_size
+20 0 1 shape 0/1 ood_020_size_object_0_1_shape
+21 5 4 size 0/1 ood_021_size_object_5_4_size
+22 5 4 shape 0/1 ood_022_size_object_5_4_shape
+23 5 4 size 0/1 ood_023_size_object_5_4_size
+24 5 4 shape 0/1 ood_024_size_object_5_4_shape
+25 4 5 size 0/1 ood_025_size_object_4_5_size
+26 4 5 shape 0/1 ood_026_size_object_4_5_shape
+27 0 1 size 0/1 ood_027_size_object_0_1_size
+28 0 1 shape 0/1 ood_028_size_object_0_1_shape
+29 4 5 size 0/1 ood_029_size_object_4_5_size
+30 4 5 shape 0/1 ood_030_size_object_4_5_shape
+31 3 2 size 0/1 ood_031_size_object_3_2_size
+32 3 2 shape 0/1 ood_032_size_object_3_2_shape
+33 0 1 size 0/1 ood_033_size_object_0_1_size
+34 0 1 shape 0/1 ood_034_size_object_0_1_shape
+35 0 1 size 0/1 ood_035_size_object_0_1_size
+36 0 1 shape 0/1 ood_036_size_object_0_1_shape
+37 0 1 size 0/1 ood_037_size_object_0_1_size
+38 0 1 shape 0/1 ood_038_size_object_0_1_shape
+39 1 0 size 0/1 ood_039_size_object_1_0_size
+40 1 0 shape 0/1 ood_040_size_object_1_0_shape
+41 1 0 size 0/1 ood_041_size_object_1_0_size
+42 1 0 shape 0/1 ood_042_size_object_1_0_shape
+43 4 5 size 0/1 ood_043_size_object_4_5_size
+44 4 5 shape 0/1 ood_044_size_object_4_5_shape
+45 4 5 size 0/1 ood_045_size_object_4_5_size
+46 4 5 shape 0/1 ood_046_size_object_4_5_shape
+47 0 1 size 0/1 ood_047_size_object_0_1_size
+48 0 1 shape 0/1 ood_048_size_object_0_1_shape
+49 4 5 size 0/1 ood_049_size_object_4_5_size
+50 4 5 shape 0/1 ood_050_size_object_4_5_shape
+51 1 0 size 0/1 ood_051_size_object_1_0_size
+52 1 0 shape 0/1 ood_052_size_object_1_0_shape
+53 5 4 size 0/1 ood_053_size_object_5_4_size
+54 5 4 shape 0/1 ood_054_size_object_5_4_shape
+55 5 4 size 0/1 ood_055_size_object_5_4_size
+56 5 4 shape 0/1 ood_056_size_object_5_4_shape
+57 5 4 size 0/1 ood_057_size_object_5_4_size
+58 5 4 shape 0/1 ood_058_size_object_5_4_shape
+59 4 5 size 0/1 ood_059_size_object_4_5_size
+60 4 5 shape 0/1 ood_060_size_object_4_5_shape
+61 3 2 size 1/1 ood_061_size_object_3_2_size
+62 3 2 shape 0/1 ood_062_size_object_3_2_shape
+63 1 0 size 0/1 ood_063_size_object_1_0_size
+64 1 0 shape 0/1 ood_064_size_object_1_0_shape
+65 3 2 size 0/1 ood_065_size_object_3_2_size
+66 3 2 shape 0/1 ood_066_size_object_3_2_shape
+67 4 5 size 0/1 ood_067_size_object_4_5_size
+68 4 5 shape 0/1 ood_068_size_object_4_5_shape
+69 2 3 size 0/1 ood_069_size_object_2_3_size
+70 2 3 shape 0/1 ood_070_size_object_2_3_shape
+71 0 1 size 0/1 ood_071_size_object_0_1_size
+72 0 1 shape 0/1 ood_072_size_object_0_1_shape
+73 1 0 size 0/1 ood_073_size_object_1_0_size
+74 1 0 shape 0/1 ood_074_size_object_1_0_shape
+75 5 4 size 0/1 ood_075_size_object_5_4_size
+76 5 4 shape 0/1 ood_076_size_object_5_4_shape
+77 3 2 size 0/1 ood_077_size_object_3_2_size
+78 3 2 shape 0/1 ood_078_size_object_3_2_shape
+79 2 3 size 0/1 ood_079_size_object_2_3_size
+80 2 3 shape 0/1 ood_080_size_object_2_3_shape
+81 2 3 size 0/1 ood_081_size_object_2_3_size
+82 2 3 shape 0/1 ood_082_size_object_2_3_shape
+83 1 0 size 0/1 ood_083_size_object_1_0_size
+84 1 0 shape 0/1 ood_084_size_object_1_0_shape
+85 1 0 size 0/1 ood_085_size_object_1_0_size
+86 1 0 shape 0/1 ood_086_size_object_1_0_shape
+87 2 3 size 0/1 ood_087_size_object_2_3_size
+88 2 3 shape 0/1 ood_088_size_object_2_3_shape
+89 0 1 size 0/1 ood_089_size_object_0_1_size
+90 0 1 shape 0/1 ood_090_size_object_0_1_shape
+91 0 1 size 0/1 ood_091_size_object_0_1_size
+92 0 1 shape 0/1 ood_092_size_object_0_1_shape
+93 3 2 size 0/1 ood_093_size_object_3_2_size
+94 3 2 shape 0/1 ood_094_size_object_3_2_shape
+95 0 1 size 0/1 ood_095_size_object_0_1_size
+96 0 1 shape 0/1 ood_096_size_object_0_1_shape
+97 2 3 size 0/1 ood_097_size_object_2_3_size
+98 2 3 shape 0/1 ood_098_size_object_2_3_shape
+99 2 3 size 0/1 ood_099_size_object_2_3_size
+100 2 3 shape 1/1 ood_100_size_object_2_3_shape
+101 4 5 size 0/1 ood_101_size_object_4_5_size
+102 4 5 shape 0/1 ood_102_size_object_4_5_shape
+103 2 3 size 0/1 ood_103_size_object_2_3_size
+104 2 3 shape 0/1 ood_104_size_object_2_3_shape
+105 0 1 size 0/1 ood_105_size_object_0_1_size
+106 0 1 shape 0/1 ood_106_size_object_0_1_shape
+107 5 4 size 0/1 ood_107_size_object_5_4_size
+108 5 4 shape 0/1 ood_108_size_object_5_4_shape
+109 3 2 size 0/1 ood_109_size_object_3_2_size
+110 3 2 shape 0/1 ood_110_size_object_3_2_shape
+111 4 5 size 0/1 ood_111_size_object_4_5_size
+112 4 5 shape 0/1 ood_112_size_object_4_5_shape
+113 0 1 size 0/1 ood_113_size_object_0_1_size
+114 0 1 shape 0/1 ood_114_size_object_0_1_shape
+115 3 2 size 0/1 ood_115_size_object_3_2_size
+116 3 2 shape 0/1 ood_116_size_object_3_2_shape
+117 0 1 size 0/1 ood_117_size_object_0_1_size
+118 0 1 shape 0/1 ood_118_size_object_0_1_shape
+119 4 5 size 0/1 ood_119_size_object_4_5_size
+120 4 5 shape 0/1 ood_120_size_object_4_5_shape
+121 2 3 size 0/1 ood_121_size_object_2_3_size
+122 2 3 shape 0/1 ood_122_size_object_2_3_shape
+123 5 4 size 0/1 ood_123_size_object_5_4_size
+124 5 4 shape 0/1 ood_124_size_object_5_4_shape
+125 4 5 size 0/1 ood_125_size_object_4_5_size
+126 4 5 shape 0/1 ood_126_size_object_4_5_shape
+127 2 3 size 0/1 ood_127_size_object_2_3_size
+128 2 3 shape 0/1 ood_128_size_object_2_3_shape
+129 4 5 size 0/1 ood_129_size_object_4_5_size
+130 4 5 shape 0/1 ood_130_size_object_4_5_shape
+131 1 0 size 0/1 ood_131_size_object_1_0_size
+132 1 0 shape 0/1 ood_132_size_object_1_0_shape
+133 5 4 size 0/1 ood_133_size_object_5_4_size
+134 5 4 shape 0/1 ood_134_size_object_5_4_shape
+135 0 1 size 0/1 ood_135_size_object_0_1_size
+136 0 1 shape 0/1 ood_136_size_object_0_1_shape
+137 0 1 size 0/1 ood_137_size_object_0_1_size
+138 0 1 shape 0/1 ood_138_size_object_0_1_shape
+139 5 4 size 0/1 ood_139_size_object_5_4_size
+140 5 4 shape 0/1 ood_140_size_object_5_4_shape
+141 1 0 size 0/1 ood_141_size_object_1_0_size
+142 1 0 shape 0/1 ood_142_size_object_1_0_shape
+143 2 3 size 0/1 ood_143_size_object_2_3_size
+144 2 3 shape 0/1 ood_144_size_object_2_3_shape
+145 0 1 size 0/1 ood_145_size_object_0_1_size
+146 0 1 shape 0/1 ood_146_size_object_0_1_shape
+147 1 0 size 0/1 ood_147_size_object_1_0_size
+148 1 0 shape 0/1 ood_148_size_object_1_0_shape
+149 0 1 size 0/1 ood_149_size_object_0_1_size
+150 0 1 shape 0/1 ood_150_size_object_0_1_shape
+151 3 2 size 0/1 ood_151_size_object_3_2_size
+152 3 2 shape 0/1 ood_152_size_object_3_2_shape
+153 2 3 size 0/1 ood_153_size_object_2_3_size
+154 2 3 shape 0/1 ood_154_size_object_2_3_shape
+155 3 2 size 0/1 ood_155_size_object_3_2_size
+156 3 2 shape 0/1 ood_156_size_object_3_2_shape
+157 5 4 size 0/1 ood_157_size_object_5_4_size
+158 5 4 shape 0/1 ood_158_size_object_5_4_shape
+159 2 3 size 0/1 ood_159_size_object_2_3_size
+160 2 3 shape 0/1 ood_160_size_object_2_3_shape
+161 1 0 size 0/1 ood_161_size_object_1_0_size
+162 1 0 shape 0/1 ood_162_size_object_1_0_shape
+163 2 3 size 0/1 ood_163_size_object_2_3_size
+164 2 3 shape 0/1 ood_164_size_object_2_3_shape
+165 2 3 size 0/1 ood_165_size_object_2_3_size
+166 2 3 shape 0/1 ood_166_size_object_2_3_shape
+167 1 0 size 0/1 ood_167_size_object_1_0_size
+168 1 0 shape 0/1 ood_168_size_object_1_0_shape
+169 5 4 size 0/1 ood_169_size_object_5_4_size
+170 5 4 shape 0/1 ood_170_size_object_5_4_shape
+171 2 3 size 0/1 ood_171_size_object_2_3_size
+172 2 3 shape 0/1 ood_172_size_object_2_3_shape
+173 5 4 size 0/1 ood_173_size_object_5_4_size
+174 5 4 shape 0/1 ood_174_size_object_5_4_shape
+175 5 4 size 0/1 ood_175_size_object_5_4_size
+176 5 4 shape 0/1 ood_176_size_object_5_4_shape
+177 5 4 size 0/1 ood_177_size_object_5_4_size
+178 5 4 shape 0/1 ood_178_size_object_5_4_shape
+179 0 1 size 0/1 ood_179_size_object_0_1_size
+180 0 1 shape 0/1 ood_180_size_object_0_1_shape
+181 4 5 size 0/1 ood_181_size_object_4_5_size
+182 4 5 shape 0/1 ood_182_size_object_4_5_shape
+183 5 4 size 0/1 ood_183_size_object_5_4_size
+184 5 4 shape 0/1 ood_184_size_object_5_4_shape
+185 1 0 size 0/1 ood_185_size_object_1_0_size
+186 1 0 shape 0/1 ood_186_size_object_1_0_shape
+187 4 5 size 0/1 ood_187_size_object_4_5_size
+188 4 5 shape 0/1 ood_188_size_object_4_5_shape
+189 5 4 size 0/1 ood_189_size_object_5_4_size
+190 5 4 shape 0/1 ood_190_size_object_5_4_shape
+191 1 0 size 0/1 ood_191_size_object_1_0_size
+192 1 0 shape 0/1 ood_192_size_object_1_0_shape
+193 1 0 size 0/1 ood_193_size_object_1_0_size
+194 1 0 shape 0/1 ood_194_size_object_1_0_shape
+195 3 2 size 1/1 ood_195_size_object_3_2_size
+196 3 2 shape 0/1 ood_196_size_object_3_2_shape
+197 3 2 size 0/1 ood_197_size_object_3_2_size
+198 3 2 shape 0/1 ood_198_size_object_3_2_shape
+199 2 3 size 0/1 ood_199_size_object_2_3_size
+200 2 3 shape 0/1 ood_200_size_object_2_3_shape
+201 5 4 size 0/1 ood_201_size_object_5_4_size
+202 5 4 shape 0/1 ood_202_size_object_5_4_shape
+203 5 4 size 0/1 ood_203_size_object_5_4_size
+204 5 4 shape 0/1 ood_204_size_object_5_4_shape
+205 4 5 size 0/1 ood_205_size_object_4_5_size
+206 4 5 shape 0/1 ood_206_size_object_4_5_shape
+207 1 0 size 0/1 ood_207_size_object_1_0_size
+208 1 0 shape 0/1 ood_208_size_object_1_0_shape
+209 5 4 size 0/1 ood_209_size_object_5_4_size
+210 5 4 shape 0/1 ood_210_size_object_5_4_shape
+211 2 3 size 0/1 ood_211_size_object_2_3_size
+212 2 3 shape 0/1 ood_212_size_object_2_3_shape
+213 0 1 size 0/1 ood_213_size_object_0_1_size
+214 0 1 shape 0/1 ood_214_size_object_0_1_shape
+215 1 0 size 0/1 ood_215_size_object_1_0_size
+216 1 0 shape 0/1 ood_216_size_object_1_0_shape
+217 0 1 size 0/1 ood_217_size_object_0_1_size
+218 0 1 shape 0/1 ood_218_size_object_0_1_shape
+219 2 3 size 0/1 ood_219_size_object_2_3_size
+220 2 3 shape 0/1 ood_220_size_object_2_3_shape
+221 3 2 size 0/1 ood_221_size_object_3_2_size
+222 3 2 shape 0/1 ood_222_size_object_3_2_shape
+223 2 3 size 0/1 ood_223_size_object_2_3_size
+224 2 3 shape 0/1 ood_224_size_object_2_3_shape
+225 0 1 size 0/1 ood_225_size_object_0_1_size
+226 0 1 shape 0/1 ood_226_size_object_0_1_shape
+227 1 0 size 0/1 ood_227_size_object_1_0_size
+228 1 0 shape 0/1 ood_228_size_object_1_0_shape
+229 4 5 size 0/1 ood_229_size_object_4_5_size
+230 4 5 shape 0/1 ood_230_size_object_4_5_shape
+231 5 4 size 0/1 ood_231_size_object_5_4_size
+232 5 4 shape 0/1 ood_232_size_object_5_4_shape
+233 2 3 size 0/1 ood_233_size_object_2_3_size
+234 2 3 shape 0/1 ood_234_size_object_2_3_shape
+235 1 0 size 0/1 ood_235_size_object_1_0_size
+236 1 0 shape 0/1 ood_236_size_object_1_0_shape
+237 5 4 size 0/1 ood_237_size_object_5_4_size
+238 5 4 shape 0/1 ood_238_size_object_5_4_shape
+239 3 2 size 0/1 ood_239_size_object_3_2_size
+240 3 2 shape 0/1 ood_240_size_object_3_2_shape
+241 3 2 size 0/1 ood_241_size_object_3_2_size
+242 3 2 shape 0/1 ood_242_size_object_3_2_shape
+243 5 4 size 0/1 ood_243_size_object_5_4_size
+244 5 4 shape 0/1 ood_244_size_object_5_4_shape
+245 3 2 size 1/1 ood_245_size_object_3_2_size
+246 3 2 shape 0/1 ood_246_size_object_3_2_shape
+247 1 0 size 0/1 ood_247_size_object_1_0_size
+248 1 0 shape 0/1 ood_248_size_object_1_0_shape
+249 2 3 size 0/1 ood_249_size_object_2_3_size
+250 2 3 shape 0/1 ood_250_size_object_2_3_shape
+251 1 0 size 0/1 ood_251_size_object_1_0_size
+252 1 0 shape 0/1 ood_252_size_object_1_0_shape
+253 1 0 size 0/1 ood_253_size_object_1_0_size
+254 1 0 shape 0/1 ood_254_size_object_1_0_shape
+255 5 4 size 0/1 ood_255_size_object_5_4_size
+256 5 4 shape 0/1 ood_256_size_object_5_4_shape
+257 4 5 size 0/1 ood_257_size_object_4_5_size
+258 4 5 shape 0/1 ood_258_size_object_4_5_shape
+259 4 5 size 0/1 ood_259_size_object_4_5_size
+260 4 5 shape 0/1 ood_260_size_object_4_5_shape
+261 2 3 size 0/1 ood_261_size_object_2_3_size
+262 2 3 shape 0/1 ood_262_size_object_2_3_shape
+263 5 4 size 0/1 ood_263_size_object_5_4_size
+264 5 4 shape 0/1 ood_264_size_object_5_4_shape
+265 4 5 size 0/1 ood_265_size_object_4_5_size
+266 4 5 shape 0/1 ood_266_size_object_4_5_shape
+267 3 2 size 0/1 ood_267_size_object_3_2_size
+268 3 2 shape 0/1 ood_268_size_object_3_2_shape
+269 4 5 size 0/1 ood_269_size_object_4_5_size
+270 4 5 shape 0/1 ood_270_size_object_4_5_shape
+271 3 2 size 0/1 ood_271_size_object_3_2_size
+272 3 2 shape 0/1 ood_272_size_object_3_2_shape
+273 2 3 size 0/1 ood_273_size_object_2_3_size
+274 2 3 shape 0/1 ood_274_size_object_2_3_shape
+275 1 0 size 0/1 ood_275_size_object_1_0_size
+276 1 0 shape 0/1 ood_276_size_object_1_0_shape
+277 1 0 size 0/1 ood_277_size_object_1_0_size
+278 1 0 shape 0/1 ood_278_size_object_1_0_shape
+279 4 5 size 0/1 ood_279_size_object_4_5_size
+280 4 5 shape 0/1 ood_280_size_object_4_5_shape
+281 3 2 size 0/1 ood_281_size_object_3_2_size
+282 3 2 shape 0/1 ood_282_size_object_3_2_shape
+283 0 1 size 0/1 ood_283_size_object_0_1_size
+284 0 1 shape 0/1 ood_284_size_object_0_1_shape
+285 0 1 size 0/1 ood_285_size_object_0_1_size
+286 0 1 shape 0/1 ood_286_size_object_0_1_shape
+287 0 1 size 0/1 ood_287_size_object_0_1_size
+288 0 1 shape 0/1 ood_288_size_object_0_1_shape
+289 1 0 size 0/1 ood_289_size_object_1_0_size
+290 1 0 shape 0/1 ood_290_size_object_1_0_shape
+291 5 4 size 0/1 ood_291_size_object_5_4_size
+292 5 4 shape 0/1 ood_292_size_object_5_4_shape
+293 1 0 size 0/1 ood_293_size_object_1_0_size
+294 1 0 shape 0/1 ood_294_size_object_1_0_shape
+295 5 4 size 0/1 ood_295_size_object_5_4_size
+296 5 4 shape 0/1 ood_296_size_object_5_4_shape
+297 3 2 size 0/1 ood_297_size_object_3_2_size
+298 3 2 shape 0/1 ood_298_size_object_3_2_shape
+299 4 5 size 0/1 ood_299_size_object_4_5_size
+300 4 5 shape 0/1 ood_300_size_object_4_5_shape
+301 0 1 size 0/1 ood_301_size_object_0_1_size
+302 0 1 shape 0/1 ood_302_size_object_0_1_shape
+303 3 2 size 0/1 ood_303_size_object_3_2_size
+304 3 2 shape 0/1 ood_304_size_object_3_2_shape
+305 3 2 size 0/1 ood_305_size_object_3_2_size
+306 3 2 shape 0/1 ood_306_size_object_3_2_shape
+307 4 5 size 0/1 ood_307_size_object_4_5_size
+308 4 5 shape 0/1 ood_308_size_object_4_5_shape
+309 3 2 size 1/1 ood_309_size_object_3_2_size
+310 3 2 shape 0/1 ood_310_size_object_3_2_shape
+311 4 5 size 0/1 ood_311_size_object_4_5_size
+312 4 5 shape 0/1 ood_312_size_object_4_5_shape
+313 2 3 size 0/1 ood_313_size_object_2_3_size
+314 2 3 shape 0/1 ood_314_size_object_2_3_shape
+315 4 5 size 0/1 ood_315_size_object_4_5_size
+316 4 5 shape 0/1 ood_316_size_object_4_5_shape
+317 0 1 size 0/1 ood_317_size_object_0_1_size
+318 0 1 shape 0/1 ood_318_size_object_0_1_shape
+319 5 4 size 0/1 ood_319_size_object_5_4_size
+320 5 4 shape 0/1 ood_320_size_object_5_4_shape
+321 5 4 size 0/1 ood_321_size_object_5_4_size
+322 5 4 shape 0/1 ood_322_size_object_5_4_shape
+323 0 1 size 0/1 ood_323_size_object_0_1_size
+324 0 1 shape 0/1 ood_324_size_object_0_1_shape
+325 5 4 size 0/1 ood_325_size_object_5_4_size
+326 5 4 shape 0/1 ood_326_size_object_5_4_shape
+327 4 5 size 0/1 ood_327_size_object_4_5_size
+328 4 5 shape 0/1 ood_328_size_object_4_5_shape
+330 2 3 shape 0/1 ood_330_size_object_2_3_shape
+331 5 4 size 0/1 ood_331_size_object_5_4_size
+332 5 4 shape 0/1 ood_332_size_object_5_4_shape
+333 2 3 size 0/1 ood_333_size_object_2_3_size
+334 2 3 shape 0/1 ood_334_size_object_2_3_shape
+335 0 1 size 0/1 ood_335_size_object_0_1_size
+336 0 1 shape 0/1 ood_336_size_object_0_1_shape
+337 2 3 size 0/1 ood_337_size_object_2_3_size
+338 2 3 shape 0/1 ood_338_size_object_2_3_shape
+339 3 2 size 0/1 ood_339_size_object_3_2_size
+340 3 2 shape 0/1 ood_340_size_object_3_2_shape
+341 1 0 size 0/1 ood_341_size_object_1_0_size
+342 1 0 shape 0/1 ood_342_size_object_1_0_shape
+343 3 2 size 0/1 ood_343_size_object_3_2_size
+344 3 2 shape 0/1 ood_344_size_object_3_2_shape
+345 0 1 size 0/1 ood_345_size_object_0_1_size
+346 0 1 shape 0/1 ood_346_size_object_0_1_shape
+347 5 4 size 0/1 ood_347_size_object_5_4_size
+348 5 4 shape 0/1 ood_348_size_object_5_4_shape
+349 5 4 size 0/1 ood_349_size_object_5_4_size
+350 5 4 shape 0/1 ood_350_size_object_5_4_shape
+351 2 3 size 0/1 ood_351_size_object_2_3_size
+352 2 3 shape 0/1 ood_352_size_object_2_3_shape
+353 4 5 size 0/1 ood_353_size_object_4_5_size
+354 4 5 shape 0/1 ood_354_size_object_4_5_shape
+355 1 0 size 0/1 ood_355_size_object_1_0_size
+356 1 0 shape 0/1 ood_356_size_object_1_0_shape
+357 4 5 size 0/1 ood_357_size_object_4_5_size
+358 4 5 shape 0/1 ood_358_size_object_4_5_shape
+359 0 1 size 0/1 ood_359_size_object_0_1_size
+360 0 1 shape 0/1 ood_360_size_object_0_1_shape
+361 5 4 size 0/1 ood_361_size_object_5_4_size
+362 5 4 shape 0/1 ood_362_size_object_5_4_shape
+363 2 3 size 0/1 ood_363_size_object_2_3_size
+364 2 3 shape 0/1 ood_364_size_object_2_3_shape
+365 5 4 size 0/1 ood_365_size_object_5_4_size
+366 5 4 shape 0/1 ood_366_size_object_5_4_shape
+367 4 5 size 0/1 ood_367_size_object_4_5_size
+368 4 5 shape 0/1 ood_368_size_object_4_5_shape
+369 4 5 size 0/1 ood_369_size_object_4_5_size
+370 4 5 shape 0/1 ood_370_size_object_4_5_shape
+371 1 0 size 0/1 ood_371_size_object_1_0_size
+372 1 0 shape 0/1 ood_372_size_object_1_0_shape
+373 1 0 size 0/1 ood_373_size_object_1_0_size
+374 1 0 shape 0/1 ood_374_size_object_1_0_shape
+375 2 3 size 0/1 ood_375_size_object_2_3_size
+376 2 3 shape 0/1 ood_376_size_object_2_3_shape
+377 1 0 size 0/1 ood_377_size_object_1_0_size
+378 1 0 shape 0/1 ood_378_size_object_1_0_shape
+379 4 5 size 0/1 ood_379_size_object_4_5_size
+380 4 5 shape 0/1 ood_380_size_object_4_5_shape
+381 4 5 size 0/1 ood_381_size_object_4_5_size
+382 4 5 shape 0/1 ood_382_size_object_4_5_shape
+383 0 1 size 0/1 ood_383_size_object_0_1_size
+384 0 1 shape 0/1 ood_384_size_object_0_1_shape
+385 4 5 size 0/1 ood_385_size_object_4_5_size
+386 4 5 shape 0/1 ood_386_size_object_4_5_shape
+387 2 3 size 0/1 ood_387_size_object_2_3_size
+388 2 3 shape 0/1 ood_388_size_object_2_3_shape
+389 3 2 size 0/1 ood_389_size_object_3_2_size
+390 3 2 shape 0/1 ood_390_size_object_3_2_shape
+391 0 1 size 0/1 ood_391_size_object_0_1_size
+392 0 1 shape 0/1 ood_392_size_object_0_1_shape
+393 0 1 size 0/1 ood_393_size_object_0_1_size
+394 0 1 shape 0/1 ood_394_size_object_0_1_shape
+395 2 3 size 0/1 ood_395_size_object_2_3_size
+396 2 3 shape 0/1 ood_396_size_object_2_3_shape
+397 2 3 size 0/1 ood_397_size_object_2_3_size
+398 2 3 shape 0/1 ood_398_size_object_2_3_shape
+399 1 0 size 0/1 ood_399_size_object_1_0_size
+400 1 0 shape 0/1 ood_400_size_object_1_0_shape
+
+overall_size_success=0/35 (0.0%)
+overall_shape_success=0/36 (0.0%)

results/genie/conflict_env/genie_spatial_object_ood_seed42.txt

@@ -0,0 +1,420 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=spatial_object
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/spatial_object
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 0 4 spatial 0/1 ood_001_spatial_object_0_4_spatial
+2 0 4 shape 0/1 ood_002_spatial_object_0_4_shape
+3 0 1 spatial 0/1 ood_003_spatial_object_0_1_spatial
+4 0 1 shape 0/1 ood_004_spatial_object_0_1_shape
+5 2 0 spatial 0/1 ood_005_spatial_object_2_0_spatial
+6 2 0 shape 0/1 ood_006_spatial_object_2_0_shape
+7 1 4 spatial 0/1 ood_007_spatial_object_1_4_spatial
+8 1 4 shape 0/1 ood_008_spatial_object_1_4_shape
+9 1 4 spatial 0/1 ood_009_spatial_object_1_4_spatial
+10 1 4 shape 0/1 ood_010_spatial_object_1_4_shape
+11 1 0 spatial 0/1 ood_011_spatial_object_1_0_spatial
+12 1 0 shape 0/1 ood_012_spatial_object_1_0_shape
+13 0 4 spatial 0/1 ood_013_spatial_object_0_4_spatial
+14 0 4 shape 0/1 ood_014_spatial_object_0_4_shape
+15 4 1 spatial 0/1 ood_015_spatial_object_4_1_spatial
+16 4 1 shape 0/1 ood_016_spatial_object_4_1_shape
+17 0 3 spatial 0/1 ood_017_spatial_object_0_3_spatial
+18 0 3 shape 0/1 ood_018_spatial_object_0_3_shape
+19 4 2 spatial 0/1 ood_019_spatial_object_4_2_spatial
+20 4 2 shape 1/1 ood_020_spatial_object_4_2_shape
+21 3 1 spatial 0/1 ood_021_spatial_object_3_1_spatial
+22 3 1 shape 0/1 ood_022_spatial_object_3_1_shape
+23 0 2 spatial 0/1 ood_023_spatial_object_0_2_spatial
+24 0 2 shape 0/1 ood_024_spatial_object_0_2_shape
+25 0 1 spatial 0/1 ood_025_spatial_object_0_1_spatial
+26 0 1 shape 0/1 ood_026_spatial_object_0_1_shape
+27 0 3 spatial 0/1 ood_027_spatial_object_0_3_spatial
+28 0 3 shape 0/1 ood_028_spatial_object_0_3_shape
+29 1 3 spatial 0/1 ood_029_spatial_object_1_3_spatial
+30 1 3 shape 0/1 ood_030_spatial_object_1_3_shape
+31 1 4 spatial 0/1 ood_031_spatial_object_1_4_spatial
+32 1 4 shape 1/1 ood_032_spatial_object_1_4_shape
+33 4 0 spatial 0/1 ood_033_spatial_object_4_0_spatial
+34 4 0 shape 0/1 ood_034_spatial_object_4_0_shape
+35 4 3 spatial 0/1 ood_035_spatial_object_4_3_spatial
+36 4 3 shape 0/1 ood_036_spatial_object_4_3_shape
+37 0 1 spatial 0/1 ood_037_spatial_object_0_1_spatial
+38 0 1 shape 0/1 ood_038_spatial_object_0_1_shape
+39 4 1 spatial 0/1 ood_039_spatial_object_4_1_spatial
+40 4 1 shape 0/1 ood_040_spatial_object_4_1_shape
+41 1 3 spatial 0/1 ood_041_spatial_object_1_3_spatial
+42 1 3 shape 0/1 ood_042_spatial_object_1_3_shape
+43 4 1 spatial 0/1 ood_043_spatial_object_4_1_spatial
+44 4 1 shape 0/1 ood_044_spatial_object_4_1_shape
+45 3 1 spatial 0/1 ood_045_spatial_object_3_1_spatial
+46 3 1 shape 0/1 ood_046_spatial_object_3_1_shape
+47 1 4 spatial 0/1 ood_047_spatial_object_1_4_spatial
+48 1 4 shape 1/1 ood_048_spatial_object_1_4_shape
+49 3 2 spatial 1/1 ood_049_spatial_object_3_2_spatial
+50 3 2 shape 0/1 ood_050_spatial_object_3_2_shape
+51 4 2 spatial 0/1 ood_051_spatial_object_4_2_spatial
+52 4 2 shape 1/1 ood_052_spatial_object_4_2_shape
+53 2 0 spatial 0/1 ood_053_spatial_object_2_0_spatial
+54 2 0 shape 0/1 ood_054_spatial_object_2_0_shape
+55 0 1 spatial 0/1 ood_055_spatial_object_0_1_spatial
+56 0 1 shape 0/1 ood_056_spatial_object_0_1_shape
+57 1 2 spatial 0/1 ood_057_spatial_object_1_2_spatial
+58 1 2 shape 0/1 ood_058_spatial_object_1_2_shape
+59 3 1 spatial 0/1 ood_059_spatial_object_3_1_spatial
+60 3 1 shape 0/1 ood_060_spatial_object_3_1_shape
+61 2 3 spatial 0/1 ood_061_spatial_object_2_3_spatial
+62 2 3 shape 0/1 ood_062_spatial_object_2_3_shape
+63 2 0 spatial 0/1 ood_063_spatial_object_2_0_spatial
+64 2 0 shape 0/1 ood_064_spatial_object_2_0_shape
+65 1 0 spatial 0/1 ood_065_spatial_object_1_0_spatial
+66 1 0 shape 0/1 ood_066_spatial_object_1_0_shape
+67 1 3 spatial 0/1 ood_067_spatial_object_1_3_spatial
+68 1 3 shape 0/1 ood_068_spatial_object_1_3_shape
+69 2 3 spatial 0/1 ood_069_spatial_object_2_3_spatial
+70 2 3 shape 0/1 ood_070_spatial_object_2_3_shape
+71 0 4 spatial 0/1 ood_071_spatial_object_0_4_spatial
+72 0 4 shape 0/1 ood_072_spatial_object_0_4_shape
+73 0 3 spatial 0/1 ood_073_spatial_object_0_3_spatial
+74 0 3 shape 0/1 ood_074_spatial_object_0_3_shape
+75 3 0 spatial 0/1 ood_075_spatial_object_3_0_spatial
+76 3 0 shape 0/1 ood_076_spatial_object_3_0_shape
+77 0 4 spatial 0/1 ood_077_spatial_object_0_4_spatial
+78 0 4 shape 0/1 ood_078_spatial_object_0_4_shape
+79 2 4 spatial 0/1 ood_079_spatial_object_2_4_spatial
+80 2 4 shape 0/1 ood_080_spatial_object_2_4_shape
+81 2 4 spatial 0/1 ood_081_spatial_object_2_4_spatial
+82 2 4 shape 0/1 ood_082_spatial_object_2_4_shape
+83 4 3 spatial 0/1 ood_083_spatial_object_4_3_spatial
+84 4 3 shape 0/1 ood_084_spatial_object_4_3_shape
+85 2 0 spatial 0/1 ood_085_spatial_object_2_0_spatial
+86 2 0 shape 0/1 ood_086_spatial_object_2_0_shape
+87 0 2 spatial 0/1 ood_087_spatial_object_0_2_spatial
+88 0 2 shape 0/1 ood_088_spatial_object_0_2_shape
+89 3 2 spatial 1/1 ood_089_spatial_object_3_2_spatial
+90 3 2 shape 0/1 ood_090_spatial_object_3_2_shape
+91 4 1 spatial 0/1 ood_091_spatial_object_4_1_spatial
+92 4 1 shape 0/1 ood_092_spatial_object_4_1_shape
+93 0 4 spatial 0/1 ood_093_spatial_object_0_4_spatial
+94 0 4 shape 0/1 ood_094_spatial_object_0_4_shape
+95 3 0 spatial 0/1 ood_095_spatial_object_3_0_spatial
+96 3 0 shape 0/1 ood_096_spatial_object_3_0_shape
+97 0 3 spatial 0/1 ood_097_spatial_object_0_3_spatial
+98 0 3 shape 0/1 ood_098_spatial_object_0_3_shape
+99 4 1 spatial 0/1 ood_099_spatial_object_4_1_spatial
+100 4 1 shape 0/1 ood_100_spatial_object_4_1_shape
+101 2 1 spatial 0/1 ood_101_spatial_object_2_1_spatial
+102 2 1 shape 0/1 ood_102_spatial_object_2_1_shape
+103 4 3 spatial 0/1 ood_103_spatial_object_4_3_spatial
+104 4 3 shape 0/1 ood_104_spatial_object_4_3_shape
+105 2 4 spatial 0/1 ood_105_spatial_object_2_4_spatial
+106 2 4 shape 0/1 ood_106_spatial_object_2_4_shape
+107 4 2 spatial 0/1 ood_107_spatial_object_4_2_spatial
+108 4 2 shape 1/1 ood_108_spatial_object_4_2_shape
+109 1 3 spatial 0/1 ood_109_spatial_object_1_3_spatial
+110 1 3 shape 1/1 ood_110_spatial_object_1_3_shape
+111 0 3 spatial 0/1 ood_111_spatial_object_0_3_spatial
+112 0 3 shape 0/1 ood_112_spatial_object_0_3_shape
+113 0 2 spatial 0/1 ood_113_spatial_object_0_2_spatial
+114 0 2 shape 0/1 ood_114_spatial_object_0_2_shape
+115 1 4 spatial 0/1 ood_115_spatial_object_1_4_spatial
+116 1 4 shape 1/1 ood_116_spatial_object_1_4_shape
+117 2 1 spatial 0/1 ood_117_spatial_object_2_1_spatial
+118 2 1 shape 1/1 ood_118_spatial_object_2_1_shape
+119 0 3 spatial 0/1 ood_119_spatial_object_0_3_spatial
+120 0 3 shape 0/1 ood_120_spatial_object_0_3_shape
+121 1 4 spatial 0/1 ood_121_spatial_object_1_4_spatial
+122 1 4 shape 1/1 ood_122_spatial_object_1_4_shape
+123 0 4 spatial 0/1 ood_123_spatial_object_0_4_spatial
+124 0 4 shape 0/1 ood_124_spatial_object_0_4_shape
+125 3 0 spatial 0/1 ood_125_spatial_object_3_0_spatial
+126 3 0 shape 1/1 ood_126_spatial_object_3_0_shape
+127 2 0 spatial 0/1 ood_127_spatial_object_2_0_spatial
+128 2 0 shape 0/1 ood_128_spatial_object_2_0_shape
+129 3 2 spatial 0/1 ood_129_spatial_object_3_2_spatial
+130 3 2 shape 0/1 ood_130_spatial_object_3_2_shape
+131 2 4 spatial 0/1 ood_131_spatial_object_2_4_spatial
+132 2 4 shape 0/1 ood_132_spatial_object_2_4_shape
+133 1 2 spatial 0/1 ood_133_spatial_object_1_2_spatial
+134 1 2 shape 0/1 ood_134_spatial_object_1_2_shape
+135 2 4 spatial 1/1 ood_135_spatial_object_2_4_spatial
+136 2 4 shape 0/1 ood_136_spatial_object_2_4_shape
+137 2 4 spatial 0/1 ood_137_spatial_object_2_4_spatial
+138 2 4 shape 0/1 ood_138_spatial_object_2_4_shape
+139 1 3 spatial 0/1 ood_139_spatial_object_1_3_spatial
+140 1 3 shape 0/1 ood_140_spatial_object_1_3_shape
+141 2 0 spatial 0/1 ood_141_spatial_object_2_0_spatial
+142 2 0 shape 0/1 ood_142_spatial_object_2_0_shape
+143 0 3 spatial 0/1 ood_143_spatial_object_0_3_spatial
+144 0 3 shape 0/1 ood_144_spatial_object_0_3_shape
+145 4 3 spatial 0/1 ood_145_spatial_object_4_3_spatial
+146 4 3 shape 0/1 ood_146_spatial_object_4_3_shape
+147 1 2 spatial 0/1 ood_147_spatial_object_1_2_spatial
+148 1 2 shape 0/1 ood_148_spatial_object_1_2_shape
+149 4 1 spatial 0/1 ood_149_spatial_object_4_1_spatial
+150 4 1 shape 0/1 ood_150_spatial_object_4_1_shape
+151 1 4 spatial 0/1 ood_151_spatial_object_1_4_spatial
+152 1 4 shape 0/1 ood_152_spatial_object_1_4_shape
+153 1 2 spatial 0/1 ood_153_spatial_object_1_2_spatial
+154 1 2 shape 0/1 ood_154_spatial_object_1_2_shape
+155 3 2 spatial 0/1 ood_155_spatial_object_3_2_spatial
+156 3 2 shape 0/1 ood_156_spatial_object_3_2_shape
+157 3 0 spatial 0/1 ood_157_spatial_object_3_0_spatial
+158 3 0 shape 0/1 ood_158_spatial_object_3_0_shape
+159 2 0 spatial 0/1 ood_159_spatial_object_2_0_spatial
+160 2 0 shape 0/1 ood_160_spatial_object_2_0_shape
+161 4 1 spatial 0/1 ood_161_spatial_object_4_1_spatial
+162 4 1 shape 0/1 ood_162_spatial_object_4_1_shape
+163 1 4 spatial 0/1 ood_163_spatial_object_1_4_spatial
+164 1 4 shape 1/1 ood_164_spatial_object_1_4_shape
+165 2 3 spatial 0/1 ood_165_spatial_object_2_3_spatial
+166 2 3 shape 0/1 ood_166_spatial_object_2_3_shape
+167 0 2 spatial 0/1 ood_167_spatial_object_0_2_spatial
+168 0 2 shape 0/1 ood_168_spatial_object_0_2_shape
+169 1 4 spatial 0/1 ood_169_spatial_object_1_4_spatial
+170 1 4 shape 1/1 ood_170_spatial_object_1_4_shape
+171 0 2 spatial 0/1 ood_171_spatial_object_0_2_spatial
+172 0 2 shape 0/1 ood_172_spatial_object_0_2_shape
+173 2 3 spatial 0/1 ood_173_spatial_object_2_3_spatial
+174 2 3 shape 0/1 ood_174_spatial_object_2_3_shape
+175 3 0 spatial 0/1 ood_175_spatial_object_3_0_spatial
+176 3 0 shape 0/1 ood_176_spatial_object_3_0_shape
+177 2 0 spatial 0/1 ood_177_spatial_object_2_0_spatial
+178 2 0 shape 1/1 ood_178_spatial_object_2_0_shape
+179 0 3 spatial 0/1 ood_179_spatial_object_0_3_spatial
+180 0 3 shape 0/1 ood_180_spatial_object_0_3_shape
+181 1 3 spatial 0/1 ood_181_spatial_object_1_3_spatial
+182 1 3 shape 1/1 ood_182_spatial_object_1_3_shape
+183 4 2 spatial 0/1 ood_183_spatial_object_4_2_spatial
+184 4 2 shape 0/1 ood_184_spatial_object_4_2_shape
+185 2 3 spatial 0/1 ood_185_spatial_object_2_3_spatial
+186 2 3 shape 0/1 ood_186_spatial_object_2_3_shape
+187 1 3 spatial 0/1 ood_187_spatial_object_1_3_spatial
+188 1 3 shape 0/1 ood_188_spatial_object_1_3_shape
+189 3 4 spatial 0/1 ood_189_spatial_object_3_4_spatial
+190 3 4 shape 0/1 ood_190_spatial_object_3_4_shape
+191 3 0 spatial 0/1 ood_191_spatial_object_3_0_spatial
+192 3 0 shape 0/1 ood_192_spatial_object_3_0_shape
+193 3 2 spatial 0/1 ood_193_spatial_object_3_2_spatial
+194 3 2 shape 0/1 ood_194_spatial_object_3_2_shape
+195 1 0 spatial 0/1 ood_195_spatial_object_1_0_spatial
+196 1 0 shape 0/1 ood_196_spatial_object_1_0_shape
+197 2 0 spatial 0/1 ood_197_spatial_object_2_0_spatial
+198 2 0 shape 0/1 ood_198_spatial_object_2_0_shape
+199 1 0 spatial 0/1 ood_199_spatial_object_1_0_spatial
+200 1 0 shape 0/1 ood_200_spatial_object_1_0_shape
+201 1 4 spatial 0/1 ood_201_spatial_object_1_4_spatial
+202 1 4 shape 0/1 ood_202_spatial_object_1_4_shape
+203 4 1 spatial 0/1 ood_203_spatial_object_4_1_spatial
+204 4 1 shape 0/1 ood_204_spatial_object_4_1_shape
+205 4 1 spatial 0/1 ood_205_spatial_object_4_1_spatial
+206 4 1 shape 0/1 ood_206_spatial_object_4_1_shape
+207 2 0 spatial 0/1 ood_207_spatial_object_2_0_spatial
+208 2 0 shape 0/1 ood_208_spatial_object_2_0_shape
+209 4 2 spatial 0/1 ood_209_spatial_object_4_2_spatial
+210 4 2 shape 0/1 ood_210_spatial_object_4_2_shape
+211 3 1 spatial 0/1 ood_211_spatial_object_3_1_spatial
+212 3 1 shape 0/1 ood_212_spatial_object_3_1_shape
+213 4 2 spatial 0/1 ood_213_spatial_object_4_2_spatial
+214 4 2 shape 1/1 ood_214_spatial_object_4_2_shape
+215 3 0 spatial 0/1 ood_215_spatial_object_3_0_spatial
+216 3 0 shape 0/1 ood_216_spatial_object_3_0_shape
+217 2 4 spatial 1/1 ood_217_spatial_object_2_4_spatial
+218 2 4 shape 0/1 ood_218_spatial_object_2_4_shape
+219 1 4 spatial 0/1 ood_219_spatial_object_1_4_spatial
+220 1 4 shape 0/1 ood_220_spatial_object_1_4_shape
+221 1 0 spatial 0/1 ood_221_spatial_object_1_0_spatial
+222 1 0 shape 0/1 ood_222_spatial_object_1_0_shape
+223 4 0 spatial 0/1 ood_223_spatial_object_4_0_spatial
+224 4 0 shape 0/1 ood_224_spatial_object_4_0_shape
+225 3 4 spatial 0/1 ood_225_spatial_object_3_4_spatial
+226 3 4 shape 0/1 ood_226_spatial_object_3_4_shape
+227 0 3 spatial 0/1 ood_227_spatial_object_0_3_spatial
+228 0 3 shape 0/1 ood_228_spatial_object_0_3_shape
+229 0 2 spatial 0/1 ood_229_spatial_object_0_2_spatial
+230 0 2 shape 0/1 ood_230_spatial_object_0_2_shape
+231 0 4 spatial 0/1 ood_231_spatial_object_0_4_spatial
+232 0 4 shape 0/1 ood_232_spatial_object_0_4_shape
+233 1 0 spatial 0/1 ood_233_spatial_object_1_0_spatial
+234 1 0 shape 0/1 ood_234_spatial_object_1_0_shape
+235 1 2 spatial 0/1 ood_235_spatial_object_1_2_spatial
+236 1 2 shape 0/1 ood_236_spatial_object_1_2_shape
+237 3 1 spatial 0/1 ood_237_spatial_object_3_1_spatial
+238 3 1 shape 0/1 ood_238_spatial_object_3_1_shape
+239 4 3 spatial 0/1 ood_239_spatial_object_4_3_spatial
+240 4 3 shape 1/1 ood_240_spatial_object_4_3_shape
+241 0 3 spatial 0/1 ood_241_spatial_object_0_3_spatial
+242 0 3 shape 1/1 ood_242_spatial_object_0_3_shape
+243 3 0 spatial 0/1 ood_243_spatial_object_3_0_spatial
+244 3 0 shape 0/1 ood_244_spatial_object_3_0_shape
+245 3 0 spatial 0/1 ood_245_spatial_object_3_0_spatial
+246 3 0 shape 0/1 ood_246_spatial_object_3_0_shape
+247 4 3 spatial 0/1 ood_247_spatial_object_4_3_spatial
+248 4 3 shape 0/1 ood_248_spatial_object_4_3_shape
+249 3 2 spatial 0/1 ood_249_spatial_object_3_2_spatial
+250 3 2 shape 0/1 ood_250_spatial_object_3_2_shape
+251 4 0 spatial 0/1 ood_251_spatial_object_4_0_spatial
+252 4 0 shape 1/1 ood_252_spatial_object_4_0_shape
+253 2 0 spatial 0/1 ood_253_spatial_object_2_0_spatial
+254 2 0 shape 1/1 ood_254_spatial_object_2_0_shape
+255 4 1 spatial 0/1 ood_255_spatial_object_4_1_spatial
+256 4 1 shape 0/1 ood_256_spatial_object_4_1_shape
+257 0 1 spatial 0/1 ood_257_spatial_object_0_1_spatial
+258 0 1 shape 0/1 ood_258_spatial_object_0_1_shape
+259 0 4 spatial 0/1 ood_259_spatial_object_0_4_spatial
+260 0 4 shape 0/1 ood_260_spatial_object_0_4_shape
+261 4 1 spatial 0/1 ood_261_spatial_object_4_1_spatial
+262 4 1 shape 0/1 ood_262_spatial_object_4_1_shape
+263 2 0 spatial 0/1 ood_263_spatial_object_2_0_spatial
+264 2 0 shape 0/1 ood_264_spatial_object_2_0_shape
+266 2 3 shape 0/1 ood_266_spatial_object_2_3_shape
+267 0 4 spatial 0/1 ood_267_spatial_object_0_4_spatial
+268 0 4 shape 0/1 ood_268_spatial_object_0_4_shape
+269 2 1 spatial 0/1 ood_269_spatial_object_2_1_spatial
+270 2 1 shape 0/1 ood_270_spatial_object_2_1_shape
+271 3 1 spatial 0/1 ood_271_spatial_object_3_1_spatial
+272 3 1 shape 0/1 ood_272_spatial_object_3_1_shape
+273 1 2 spatial 0/1 ood_273_spatial_object_1_2_spatial
+274 1 2 shape 0/1 ood_274_spatial_object_1_2_shape
+275 3 2 spatial 0/1 ood_275_spatial_object_3_2_spatial
+276 3 2 shape 0/1 ood_276_spatial_object_3_2_shape
+277 0 1 spatial 0/1 ood_277_spatial_object_0_1_spatial
+278 0 1 shape 0/1 ood_278_spatial_object_0_1_shape
+279 2 0 spatial 0/1 ood_279_spatial_object_2_0_spatial
+280 2 0 shape 0/1 ood_280_spatial_object_2_0_shape
+281 4 0 spatial 0/1 ood_281_spatial_object_4_0_spatial
+282 4 0 shape 0/1 ood_282_spatial_object_4_0_shape
+283 1 2 spatial 0/1 ood_283_spatial_object_1_2_spatial
+284 1 2 shape 0/1 ood_284_spatial_object_1_2_shape
+285 4 0 spatial 0/1 ood_285_spatial_object_4_0_spatial
+286 4 0 shape 0/1 ood_286_spatial_object_4_0_shape
+287 0 4 spatial 0/1 ood_287_spatial_object_0_4_spatial
+288 0 4 shape 0/1 ood_288_spatial_object_0_4_shape
+289 2 1 spatial 0/1 ood_289_spatial_object_2_1_spatial
+290 2 1 shape 0/1 ood_290_spatial_object_2_1_shape
+291 4 0 spatial 0/1 ood_291_spatial_object_4_0_spatial
+292 4 0 shape 0/1 ood_292_spatial_object_4_0_shape
+293 4 3 spatial 0/1 ood_293_spatial_object_4_3_spatial
+294 4 3 shape 0/1 ood_294_spatial_object_4_3_shape
+295 1 3 spatial 0/1 ood_295_spatial_object_1_3_spatial
+296 1 3 shape 0/1 ood_296_spatial_object_1_3_shape
+297 1 0 spatial 0/1 ood_297_spatial_object_1_0_spatial
+298 1 0 shape 1/1 ood_298_spatial_object_1_0_shape
+299 2 4 spatial 0/1 ood_299_spatial_object_2_4_spatial
+300 2 4 shape 0/1 ood_300_spatial_object_2_4_shape
+301 1 2 spatial 0/1 ood_301_spatial_object_1_2_spatial
+302 1 2 shape 1/1 ood_302_spatial_object_1_2_shape
+303 4 1 spatial 0/1 ood_303_spatial_object_4_1_spatial
+304 4 1 shape 0/1 ood_304_spatial_object_4_1_shape
+305 4 0 spatial 0/1 ood_305_spatial_object_4_0_spatial
+306 4 0 shape 0/1 ood_306_spatial_object_4_0_shape
+307 0 1 spatial 0/1 ood_307_spatial_object_0_1_spatial
+308 0 1 shape 0/1 ood_308_spatial_object_0_1_shape
+309 4 3 spatial 0/1 ood_309_spatial_object_4_3_spatial
+310 4 3 shape 1/1 ood_310_spatial_object_4_3_shape
+311 2 3 spatial 0/1 ood_311_spatial_object_2_3_spatial
+312 2 3 shape 1/1 ood_312_spatial_object_2_3_shape
+313 3 4 spatial 0/1 ood_313_spatial_object_3_4_spatial
+314 3 4 shape 0/1 ood_314_spatial_object_3_4_shape
+315 0 1 spatial 0/1 ood_315_spatial_object_0_1_spatial
+316 0 1 shape 0/1 ood_316_spatial_object_0_1_shape
+317 0 4 spatial 0/1 ood_317_spatial_object_0_4_spatial
+318 0 4 shape 0/1 ood_318_spatial_object_0_4_shape
+319 2 4 spatial 1/1 ood_319_spatial_object_2_4_spatial
+320 2 4 shape 0/1 ood_320_spatial_object_2_4_shape
+321 2 1 spatial 0/1 ood_321_spatial_object_2_1_spatial
+322 2 1 shape 0/1 ood_322_spatial_object_2_1_shape
+323 1 4 spatial 0/1 ood_323_spatial_object_1_4_spatial
+324 1 4 shape 0/1 ood_324_spatial_object_1_4_shape
+325 0 2 spatial 0/1 ood_325_spatial_object_0_2_spatial
+326 0 2 shape 0/1 ood_326_spatial_object_0_2_shape
+327 1 4 spatial 0/1 ood_327_spatial_object_1_4_spatial
+328 1 4 shape 1/1 ood_328_spatial_object_1_4_shape
+329 4 2 spatial 0/1 ood_329_spatial_object_4_2_spatial
+330 4 2 shape 0/1 ood_330_spatial_object_4_2_shape
+331 0 3 spatial 0/1 ood_331_spatial_object_0_3_spatial
+332 0 3 shape 0/1 ood_332_spatial_object_0_3_shape
+333 0 3 spatial 0/1 ood_333_spatial_object_0_3_spatial
+334 0 3 shape 0/1 ood_334_spatial_object_0_3_shape
+335 3 4 spatial 0/1 ood_335_spatial_object_3_4_spatial
+336 3 4 shape 0/1 ood_336_spatial_object_3_4_shape
+337 0 3 spatial 0/1 ood_337_spatial_object_0_3_spatial
+338 0 3 shape 0/1 ood_338_spatial_object_0_3_shape
+339 4 1 spatial 0/1 ood_339_spatial_object_4_1_spatial
+340 4 1 shape 0/1 ood_340_spatial_object_4_1_shape
+341 1 0 spatial 0/1 ood_341_spatial_object_1_0_spatial
+342 1 0 shape 1/1 ood_342_spatial_object_1_0_shape
+343 1 0 spatial 0/1 ood_343_spatial_object_1_0_spatial
+344 1 0 shape 0/1 ood_344_spatial_object_1_0_shape
+345 3 4 spatial 0/1 ood_345_spatial_object_3_4_spatial
+346 3 4 shape 0/1 ood_346_spatial_object_3_4_shape
+347 4 1 spatial 0/1 ood_347_spatial_object_4_1_spatial
+348 4 1 shape 0/1 ood_348_spatial_object_4_1_shape
+349 1 2 spatial 0/1 ood_349_spatial_object_1_2_spatial
+350 1 2 shape 0/1 ood_350_spatial_object_1_2_shape
+351 2 0 spatial 0/1 ood_351_spatial_object_2_0_spatial
+352 2 0 shape 0/1 ood_352_spatial_object_2_0_shape
+353 4 0 spatial 0/1 ood_353_spatial_object_4_0_spatial
+354 4 0 shape 0/1 ood_354_spatial_object_4_0_shape
+355 4 3 spatial 0/1 ood_355_spatial_object_4_3_spatial
+356 4 3 shape 0/1 ood_356_spatial_object_4_3_shape
+357 3 1 spatial 0/1 ood_357_spatial_object_3_1_spatial
+358 3 1 shape 0/1 ood_358_spatial_object_3_1_shape
+359 1 3 spatial 0/1 ood_359_spatial_object_1_3_spatial
+360 1 3 shape 1/1 ood_360_spatial_object_1_3_shape
+361 4 1 spatial 0/1 ood_361_spatial_object_4_1_spatial
+362 4 1 shape 0/1 ood_362_spatial_object_4_1_shape
+363 1 3 spatial 0/1 ood_363_spatial_object_1_3_spatial
+364 1 3 shape 0/1 ood_364_spatial_object_1_3_shape
+365 2 1 spatial 0/1 ood_365_spatial_object_2_1_spatial
+366 2 1 shape 0/1 ood_366_spatial_object_2_1_shape
+367 3 0 spatial 0/1 ood_367_spatial_object_3_0_spatial
+368 3 0 shape 0/1 ood_368_spatial_object_3_0_shape
+369 2 4 spatial 0/1 ood_369_spatial_object_2_4_spatial
+370 2 4 shape 0/1 ood_370_spatial_object_2_4_shape
+371 3 2 spatial 0/1 ood_371_spatial_object_3_2_spatial
+372 3 2 shape 0/1 ood_372_spatial_object_3_2_shape
+373 4 0 spatial 0/1 ood_373_spatial_object_4_0_spatial
+374 4 0 shape 0/1 ood_374_spatial_object_4_0_shape
+375 3 2 spatial 0/1 ood_375_spatial_object_3_2_spatial
+376 3 2 shape 0/1 ood_376_spatial_object_3_2_shape
+377 0 4 spatial 0/1 ood_377_spatial_object_0_4_spatial
+378 0 4 shape 0/1 ood_378_spatial_object_0_4_shape
+379 1 4 spatial 0/1 ood_379_spatial_object_1_4_spatial
+380 1 4 shape 0/1 ood_380_spatial_object_1_4_shape
+381 1 4 spatial 0/1 ood_381_spatial_object_1_4_spatial
+382 1 4 shape 1/1 ood_382_spatial_object_1_4_shape
+383 0 3 spatial 0/1 ood_383_spatial_object_0_3_spatial
+384 0 3 shape 0/1 ood_384_spatial_object_0_3_shape
+385 2 3 spatial 0/1 ood_385_spatial_object_2_3_spatial
+386 2 3 shape 0/1 ood_386_spatial_object_2_3_shape
+387 0 1 spatial 0/1 ood_387_spatial_object_0_1_spatial
+388 0 1 shape 0/1 ood_388_spatial_object_0_1_shape
+389 4 2 spatial 0/1 ood_389_spatial_object_4_2_spatial
+390 4 2 shape 1/1 ood_390_spatial_object_4_2_shape
+391 4 1 spatial 0/1 ood_391_spatial_object_4_1_spatial
+392 4 1 shape 0/1 ood_392_spatial_object_4_1_shape
+393 1 4 spatial 0/1 ood_393_spatial_object_1_4_spatial
+394 1 4 shape 1/1 ood_394_spatial_object_1_4_shape
+395 4 2 spatial 0/1 ood_395_spatial_object_4_2_spatial
+396 4 2 shape 1/1 ood_396_spatial_object_4_2_shape
+397 1 4 spatial 0/1 ood_397_spatial_object_1_4_spatial
+398 1 4 shape 0/1 ood_398_spatial_object_1_4_shape
+399 0 1 spatial 0/1 ood_399_spatial_object_0_1_spatial
+400 0 1 shape 0/1 ood_400_spatial_object_0_1_shape
+
+overall_spatial_success=1/67 (1.5%)
+overall_shape_success=11/68 (16.2%)

results/genie/conflict_env/genie_spatial_size_ood_seed42.txt

@@ -0,0 +1,421 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=spatial_size
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/spatial_size
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 0 1 spatial 1/1 ood_001_spatial_size_0_1_spatial
+2 0 1 size 0/1 ood_002_spatial_size_0_1_size
+3 0 1 spatial 0/1 ood_003_spatial_size_0_1_spatial
+4 0 1 size 0/1 ood_004_spatial_size_0_1_size
+5 2 3 spatial 1/1 ood_005_spatial_size_2_3_spatial
+6 2 3 size 1/1 ood_006_spatial_size_2_3_size
+7 1 0 spatial 0/1 ood_007_spatial_size_1_0_spatial
+8 1 0 size 0/1 ood_008_spatial_size_1_0_size
+9 1 0 spatial 0/1 ood_009_spatial_size_1_0_spatial
+10 1 0 size 0/1 ood_010_spatial_size_1_0_size
+11 1 0 spatial 1/1 ood_011_spatial_size_1_0_spatial
+12 1 0 size 0/1 ood_012_spatial_size_1_0_size
+13 0 1 spatial 1/1 ood_013_spatial_size_0_1_spatial
+14 0 1 size 0/1 ood_014_spatial_size_0_1_size
+15 0 1 spatial 1/1 ood_015_spatial_size_0_1_spatial
+16 0 1 size 0/1 ood_016_spatial_size_0_1_size
+17 3 2 spatial 1/1 ood_017_spatial_size_3_2_spatial
+18 3 2 size 0/1 ood_018_spatial_size_3_2_size
+19 0 1 spatial 1/1 ood_019_spatial_size_0_1_spatial
+20 0 1 size 0/1 ood_020_spatial_size_0_1_size
+21 0 1 spatial 0/1 ood_021_spatial_size_0_1_spatial
+22 0 1 size 0/1 ood_022_spatial_size_0_1_size
+23 0 1 spatial 1/1 ood_023_spatial_size_0_1_spatial
+24 0 1 size 0/1 ood_024_spatial_size_0_1_size
+25 1 0 spatial 0/1 ood_025_spatial_size_1_0_spatial
+26 1 0 size 0/1 ood_026_spatial_size_1_0_size
+27 1 0 spatial 0/1 ood_027_spatial_size_1_0_spatial
+28 1 0 size 0/1 ood_028_spatial_size_1_0_size
+29 0 1 spatial 1/1 ood_029_spatial_size_0_1_spatial
+30 0 1 size 0/1 ood_030_spatial_size_0_1_size
+31 1 0 spatial 0/1 ood_031_spatial_size_1_0_spatial
+32 1 0 size 0/1 ood_032_spatial_size_1_0_size
+33 3 2 spatial 0/1 ood_033_spatial_size_3_2_spatial
+34 3 2 size 0/1 ood_034_spatial_size_3_2_size
+35 1 0 spatial 0/1 ood_035_spatial_size_1_0_spatial
+36 1 0 size 0/1 ood_036_spatial_size_1_0_size
+37 3 2 spatial 0/1 ood_037_spatial_size_3_2_spatial
+38 3 2 size 0/1 ood_038_spatial_size_3_2_size
+39 2 3 spatial 0/1 ood_039_spatial_size_2_3_spatial
+40 2 3 size 0/1 ood_040_spatial_size_2_3_size
+41 0 1 spatial 0/1 ood_041_spatial_size_0_1_spatial
+42 0 1 size 0/1 ood_042_spatial_size_0_1_size
+43 1 0 spatial 0/1 ood_043_spatial_size_1_0_spatial
+44 1 0 size 0/1 ood_044_spatial_size_1_0_size
+45 3 2 spatial 0/1 ood_045_spatial_size_3_2_spatial
+46 3 2 size 0/1 ood_046_spatial_size_3_2_size
+47 2 3 spatial 0/1 ood_047_spatial_size_2_3_spatial
+48 2 3 size 1/1 ood_048_spatial_size_2_3_size
+49 2 3 spatial 0/1 ood_049_spatial_size_2_3_spatial
+50 2 3 size 1/1 ood_050_spatial_size_2_3_size
+51 1 0 spatial 0/1 ood_051_spatial_size_1_0_spatial
+52 1 0 size 0/1 ood_052_spatial_size_1_0_size
+53 1 0 spatial 0/1 ood_053_spatial_size_1_0_spatial
+54 1 0 size 0/1 ood_054_spatial_size_1_0_size
+55 2 3 spatial 0/1 ood_055_spatial_size_2_3_spatial
+56 2 3 size 1/1 ood_056_spatial_size_2_3_size
+57 0 1 spatial 0/1 ood_057_spatial_size_0_1_spatial
+58 0 1 size 0/1 ood_058_spatial_size_0_1_size
+59 0 1 spatial 0/1 ood_059_spatial_size_0_1_spatial
+60 0 1 size 0/1 ood_060_spatial_size_0_1_size
+61 3 2 spatial 1/1 ood_061_spatial_size_3_2_spatial
+62 3 2 size 0/1 ood_062_spatial_size_3_2_size
+63 0 1 spatial 0/1 ood_063_spatial_size_0_1_spatial
+64 0 1 size 0/1 ood_064_spatial_size_0_1_size
+65 2 3 spatial 0/1 ood_065_spatial_size_2_3_spatial
+66 2 3 size 1/1 ood_066_spatial_size_2_3_size
+67 2 3 spatial 0/1 ood_067_spatial_size_2_3_spatial
+68 2 3 size 1/1 ood_068_spatial_size_2_3_size
+69 2 3 spatial 0/1 ood_069_spatial_size_2_3_spatial
+70 2 3 size 1/1 ood_070_spatial_size_2_3_size
+71 0 1 spatial 1/1 ood_071_spatial_size_0_1_spatial
+72 0 1 size 0/1 ood_072_spatial_size_0_1_size
+73 3 2 spatial 1/1 ood_073_spatial_size_3_2_spatial
+74 3 2 size 0/1 ood_074_spatial_size_3_2_size
+75 0 1 spatial 1/1 ood_075_spatial_size_0_1_spatial
+76 0 1 size 0/1 ood_076_spatial_size_0_1_size
+77 3 2 spatial 0/1 ood_077_spatial_size_3_2_spatial
+78 3 2 size 0/1 ood_078_spatial_size_3_2_size
+79 0 1 spatial 0/1 ood_079_spatial_size_0_1_spatial
+80 0 1 size 0/1 ood_080_spatial_size_0_1_size
+81 2 3 spatial 0/1 ood_081_spatial_size_2_3_spatial
+82 2 3 size 0/1 ood_082_spatial_size_2_3_size
+83 2 3 spatial 0/1 ood_083_spatial_size_2_3_spatial
+84 2 3 size 1/1 ood_084_spatial_size_2_3_size
+85 1 0 spatial 0/1 ood_085_spatial_size_1_0_spatial
+86 1 0 size 0/1 ood_086_spatial_size_1_0_size
+87 0 1 spatial 0/1 ood_087_spatial_size_0_1_spatial
+88 0 1 size 0/1 ood_088_spatial_size_0_1_size
+89 0 1 spatial 0/1 ood_089_spatial_size_0_1_spatial
+90 0 1 size 0/1 ood_090_spatial_size_0_1_size
+91 1 0 spatial 0/1 ood_091_spatial_size_1_0_spatial
+92 1 0 size 0/1 ood_092_spatial_size_1_0_size
+93 2 3 spatial 1/1 ood_093_spatial_size_2_3_spatial
+94 2 3 size 1/1 ood_094_spatial_size_2_3_size
+95 0 1 spatial 0/1 ood_095_spatial_size_0_1_spatial
+96 0 1 size 0/1 ood_096_spatial_size_0_1_size
+97 1 0 spatial 0/1 ood_097_spatial_size_1_0_spatial
+98 1 0 size 0/1 ood_098_spatial_size_1_0_size
+99 0 1 spatial 0/1 ood_099_spatial_size_0_1_spatial
+100 0 1 size 0/1 ood_100_spatial_size_0_1_size
+101 3 2 spatial 1/1 ood_101_spatial_size_3_2_spatial
+102 3 2 size 0/1 ood_102_spatial_size_3_2_size
+103 2 3 spatial 0/1 ood_103_spatial_size_2_3_spatial
+104 2 3 size 1/1 ood_104_spatial_size_2_3_size
+105 3 2 spatial 1/1 ood_105_spatial_size_3_2_spatial
+106 3 2 size 0/1 ood_106_spatial_size_3_2_size
+107 2 3 spatial 0/1 ood_107_spatial_size_2_3_spatial
+108 2 3 size 1/1 ood_108_spatial_size_2_3_size
+109 1 0 spatial 1/1 ood_109_spatial_size_1_0_spatial
+110 1 0 size 0/1 ood_110_spatial_size_1_0_size
+111 2 3 spatial 0/1 ood_111_spatial_size_2_3_spatial
+112 2 3 size 1/1 ood_112_spatial_size_2_3_size
+113 2 3 spatial 0/1 ood_113_spatial_size_2_3_spatial
+114 2 3 size 1/1 ood_114_spatial_size_2_3_size
+115 1 0 spatial 0/1 ood_115_spatial_size_1_0_spatial
+116 1 0 size 0/1 ood_116_spatial_size_1_0_size
+117 2 3 spatial 0/1 ood_117_spatial_size_2_3_spatial
+118 2 3 size 1/1 ood_118_spatial_size_2_3_size
+119 0 1 spatial 1/1 ood_119_spatial_size_0_1_spatial
+120 0 1 size 0/1 ood_120_spatial_size_0_1_size
+121 1 0 spatial 0/1 ood_121_spatial_size_1_0_spatial
+122 1 0 size 0/1 ood_122_spatial_size_1_0_size
+123 1 0 spatial 0/1 ood_123_spatial_size_1_0_spatial
+124 1 0 size 0/1 ood_124_spatial_size_1_0_size
+125 1 0 spatial 0/1 ood_125_spatial_size_1_0_spatial
+126 1 0 size 0/1 ood_126_spatial_size_1_0_size
+127 3 2 spatial 1/1 ood_127_spatial_size_3_2_spatial
+128 3 2 size 0/1 ood_128_spatial_size_3_2_size
+129 3 2 spatial 0/1 ood_129_spatial_size_3_2_spatial
+130 3 2 size 0/1 ood_130_spatial_size_3_2_size
+131 2 3 spatial 0/1 ood_131_spatial_size_2_3_spatial
+132 2 3 size 1/1 ood_132_spatial_size_2_3_size
+133 1 0 spatial 0/1 ood_133_spatial_size_1_0_spatial
+134 1 0 size 0/1 ood_134_spatial_size_1_0_size
+135 2 3 spatial 0/1 ood_135_spatial_size_2_3_spatial
+136 2 3 size 1/1 ood_136_spatial_size_2_3_size
+137 0 1 spatial 0/1 ood_137_spatial_size_0_1_spatial
+138 0 1 size 0/1 ood_138_spatial_size_0_1_size
+139 1 0 spatial 0/1 ood_139_spatial_size_1_0_spatial
+140 1 0 size 0/1 ood_140_spatial_size_1_0_size
+141 0 1 spatial 0/1 ood_141_spatial_size_0_1_spatial
+142 0 1 size 0/1 ood_142_spatial_size_0_1_size
+143 2 3 spatial 0/1 ood_143_spatial_size_2_3_spatial
+144 2 3 size 1/1 ood_144_spatial_size_2_3_size
+145 3 2 spatial 0/1 ood_145_spatial_size_3_2_spatial
+146 3 2 size 0/1 ood_146_spatial_size_3_2_size
+147 2 3 spatial 0/1 ood_147_spatial_size_2_3_spatial
+148 2 3 size 1/1 ood_148_spatial_size_2_3_size
+149 0 1 spatial 0/1 ood_149_spatial_size_0_1_spatial
+150 0 1 size 0/1 ood_150_spatial_size_0_1_size
+151 1 0 spatial 0/1 ood_151_spatial_size_1_0_spatial
+152 1 0 size 0/1 ood_152_spatial_size_1_0_size
+153 2 3 spatial 0/1 ood_153_spatial_size_2_3_spatial
+154 2 3 size 0/1 ood_154_spatial_size_2_3_size
+155 1 0 spatial 1/1 ood_155_spatial_size_1_0_spatial
+156 1 0 size 0/1 ood_156_spatial_size_1_0_size
+157 3 2 spatial 0/1 ood_157_spatial_size_3_2_spatial
+158 3 2 size 0/1 ood_158_spatial_size_3_2_size
+159 3 2 spatial 1/1 ood_159_spatial_size_3_2_spatial
+160 3 2 size 0/1 ood_160_spatial_size_3_2_size
+161 3 2 spatial 0/1 ood_161_spatial_size_3_2_spatial
+162 3 2 size 0/1 ood_162_spatial_size_3_2_size
+163 1 0 spatial 1/1 ood_163_spatial_size_1_0_spatial
+164 1 0 size 0/1 ood_164_spatial_size_1_0_size
+165 2 3 spatial 0/1 ood_165_spatial_size_2_3_spatial
+166 2 3 size 1/1 ood_166_spatial_size_2_3_size
+167 1 0 spatial 0/1 ood_167_spatial_size_1_0_spatial
+168 1 0 size 0/1 ood_168_spatial_size_1_0_size
+169 1 0 spatial 0/1 ood_169_spatial_size_1_0_spatial
+170 1 0 size 0/1 ood_170_spatial_size_1_0_size
+171 2 3 spatial 0/1 ood_171_spatial_size_2_3_spatial
+172 2 3 size 1/1 ood_172_spatial_size_2_3_size
+173 3 2 spatial 1/1 ood_173_spatial_size_3_2_spatial
+174 3 2 size 0/1 ood_174_spatial_size_3_2_size
+175 3 2 spatial 0/1 ood_175_spatial_size_3_2_spatial
+176 3 2 size 0/1 ood_176_spatial_size_3_2_size
+177 2 3 spatial 0/1 ood_177_spatial_size_2_3_spatial
+178 2 3 size 1/1 ood_178_spatial_size_2_3_size
+179 1 0 spatial 0/1 ood_179_spatial_size_1_0_spatial
+180 1 0 size 0/1 ood_180_spatial_size_1_0_size
+181 1 0 spatial 0/1 ood_181_spatial_size_1_0_spatial
+182 1 0 size 0/1 ood_182_spatial_size_1_0_size
+183 3 2 spatial 1/1 ood_183_spatial_size_3_2_spatial
+184 3 2 size 0/1 ood_184_spatial_size_3_2_size
+185 0 1 spatial 1/1 ood_185_spatial_size_0_1_spatial
+186 0 1 size 0/1 ood_186_spatial_size_0_1_size
+187 0 1 spatial 0/1 ood_187_spatial_size_0_1_spatial
+188 0 1 size 0/1 ood_188_spatial_size_0_1_size
+189 0 1 spatial 1/1 ood_189_spatial_size_0_1_spatial
+190 0 1 size 0/1 ood_190_spatial_size_0_1_size
+191 1 0 spatial 0/1 ood_191_spatial_size_1_0_spatial
+192 1 0 size 0/1 ood_192_spatial_size_1_0_size
+193 1 0 spatial 0/1 ood_193_spatial_size_1_0_spatial
+194 1 0 size 0/1 ood_194_spatial_size_1_0_size
+195 3 2 spatial 1/1 ood_195_spatial_size_3_2_spatial
+196 3 2 size 0/1 ood_196_spatial_size_3_2_size
+197 0 1 spatial 0/1 ood_197_spatial_size_0_1_spatial
+198 0 1 size 0/1 ood_198_spatial_size_0_1_size
+199 3 2 spatial 1/1 ood_199_spatial_size_3_2_spatial
+200 3 2 size 0/1 ood_200_spatial_size_3_2_size
+201 3 2 spatial 0/1 ood_201_spatial_size_3_2_spatial
+202 3 2 size 0/1 ood_202_spatial_size_3_2_size
+203 3 2 spatial 1/1 ood_203_spatial_size_3_2_spatial
+204 3 2 size 0/1 ood_204_spatial_size_3_2_size
+205 2 3 spatial 0/1 ood_205_spatial_size_2_3_spatial
+206 2 3 size 1/1 ood_206_spatial_size_2_3_size
+207 0 1 spatial 0/1 ood_207_spatial_size_0_1_spatial
+208 0 1 size 0/1 ood_208_spatial_size_0_1_size
+209 0 1 spatial 1/1 ood_209_spatial_size_0_1_spatial
+210 0 1 size 0/1 ood_210_spatial_size_0_1_size
+211 2 3 spatial 0/1 ood_211_spatial_size_2_3_spatial
+212 2 3 size 1/1 ood_212_spatial_size_2_3_size
+213 2 3 spatial 0/1 ood_213_spatial_size_2_3_spatial
+214 2 3 size 1/1 ood_214_spatial_size_2_3_size
+215 0 1 spatial 1/1 ood_215_spatial_size_0_1_spatial
+216 0 1 size 0/1 ood_216_spatial_size_0_1_size
+217 2 3 spatial 0/1 ood_217_spatial_size_2_3_spatial
+218 2 3 size 1/1 ood_218_spatial_size_2_3_size
+219 3 2 spatial 1/1 ood_219_spatial_size_3_2_spatial
+220 3 2 size 0/1 ood_220_spatial_size_3_2_size
+221 1 0 spatial 0/1 ood_221_spatial_size_1_0_spatial
+222 1 0 size 0/1 ood_222_spatial_size_1_0_size
+223 3 2 spatial 0/1 ood_223_spatial_size_3_2_spatial
+224 3 2 size 0/1 ood_224_spatial_size_3_2_size
+225 0 1 spatial 0/1 ood_225_spatial_size_0_1_spatial
+226 0 1 size 0/1 ood_226_spatial_size_0_1_size
+227 2 3 spatial 0/1 ood_227_spatial_size_2_3_spatial
+228 2 3 size 1/1 ood_228_spatial_size_2_3_size
+229 1 0 spatial 0/1 ood_229_spatial_size_1_0_spatial
+230 1 0 size 0/1 ood_230_spatial_size_1_0_size
+231 0 1 spatial 0/1 ood_231_spatial_size_0_1_spatial
+232 0 1 size 0/1 ood_232_spatial_size_0_1_size
+233 2 3 spatial 0/1 ood_233_spatial_size_2_3_spatial
+234 2 3 size 0/1 ood_234_spatial_size_2_3_size
+235 1 0 spatial 0/1 ood_235_spatial_size_1_0_spatial
+236 1 0 size 0/1 ood_236_spatial_size_1_0_size
+237 1 0 spatial 0/1 ood_237_spatial_size_1_0_spatial
+238 1 0 size 0/1 ood_238_spatial_size_1_0_size
+239 2 3 spatial 0/1 ood_239_spatial_size_2_3_spatial
+240 2 3 size 1/1 ood_240_spatial_size_2_3_size
+241 1 0 spatial 0/1 ood_241_spatial_size_1_0_spatial
+242 1 0 size 0/1 ood_242_spatial_size_1_0_size
+243 0 1 spatial 1/1 ood_243_spatial_size_0_1_spatial
+244 0 1 size 0/1 ood_244_spatial_size_0_1_size
+245 2 3 spatial 0/1 ood_245_spatial_size_2_3_spatial
+246 2 3 size 1/1 ood_246_spatial_size_2_3_size
+247 3 2 spatial 0/1 ood_247_spatial_size_3_2_spatial
+248 3 2 size 0/1 ood_248_spatial_size_3_2_size
+249 0 1 spatial 0/1 ood_249_spatial_size_0_1_spatial
+250 0 1 size 0/1 ood_250_spatial_size_0_1_size
+251 0 1 spatial 1/1 ood_251_spatial_size_0_1_spatial
+252 0 1 size 0/1 ood_252_spatial_size_0_1_size
+253 2 3 spatial 0/1 ood_253_spatial_size_2_3_spatial
+254 2 3 size 1/1 ood_254_spatial_size_2_3_size
+255 2 3 spatial 0/1 ood_255_spatial_size_2_3_spatial
+256 2 3 size 1/1 ood_256_spatial_size_2_3_size
+257 1 0 spatial 0/1 ood_257_spatial_size_1_0_spatial
+258 1 0 size 0/1 ood_258_spatial_size_1_0_size
+259 0 1 spatial 0/1 ood_259_spatial_size_0_1_spatial
+260 0 1 size 0/1 ood_260_spatial_size_0_1_size
+261 1 0 spatial 0/1 ood_261_spatial_size_1_0_spatial
+262 1 0 size 0/1 ood_262_spatial_size_1_0_size
+263 0 1 spatial 0/1 ood_263_spatial_size_0_1_spatial
+264 0 1 size 0/1 ood_264_spatial_size_0_1_size
+265 0 1 spatial 0/1 ood_265_spatial_size_0_1_spatial
+266 0 1 size 0/1 ood_266_spatial_size_0_1_size
+267 3 2 spatial 1/1 ood_267_spatial_size_3_2_spatial
+268 3 2 size 0/1 ood_268_spatial_size_3_2_size
+269 0 1 spatial 1/1 ood_269_spatial_size_0_1_spatial
+270 0 1 size 0/1 ood_270_spatial_size_0_1_size
+271 1 0 spatial 0/1 ood_271_spatial_size_1_0_spatial
+272 1 0 size 0/1 ood_272_spatial_size_1_0_size
+273 1 0 spatial 0/1 ood_273_spatial_size_1_0_spatial
+274 1 0 size 0/1 ood_274_spatial_size_1_0_size
+275 3 2 spatial 0/1 ood_275_spatial_size_3_2_spatial
+276 3 2 size 0/1 ood_276_spatial_size_3_2_size
+277 1 0 spatial 0/1 ood_277_spatial_size_1_0_spatial
+278 1 0 size 0/1 ood_278_spatial_size_1_0_size
+279 2 3 spatial 0/1 ood_279_spatial_size_2_3_spatial
+280 2 3 size 1/1 ood_280_spatial_size_2_3_size
+281 3 2 spatial 1/1 ood_281_spatial_size_3_2_spatial
+282 3 2 size 0/1 ood_282_spatial_size_3_2_size
+283 1 0 spatial 0/1 ood_283_spatial_size_1_0_spatial
+284 1 0 size 0/1 ood_284_spatial_size_1_0_size
+285 1 0 spatial 0/1 ood_285_spatial_size_1_0_spatial
+286 1 0 size 0/1 ood_286_spatial_size_1_0_size
+287 2 3 spatial 0/1 ood_287_spatial_size_2_3_spatial
+288 2 3 size 0/1 ood_288_spatial_size_2_3_size
+289 3 2 spatial 0/1 ood_289_spatial_size_3_2_spatial
+290 3 2 size 0/1 ood_290_spatial_size_3_2_size
+291 2 3 spatial 0/1 ood_291_spatial_size_2_3_spatial
+292 2 3 size 1/1 ood_292_spatial_size_2_3_size
+293 3 2 spatial 0/1 ood_293_spatial_size_3_2_spatial
+294 3 2 size 0/1 ood_294_spatial_size_3_2_size
+295 3 2 spatial 0/1 ood_295_spatial_size_3_2_spatial
+296 3 2 size 0/1 ood_296_spatial_size_3_2_size
+297 0 1 spatial 1/1 ood_297_spatial_size_0_1_spatial
+298 0 1 size 0/1 ood_298_spatial_size_0_1_size
+299 1 0 spatial 0/1 ood_299_spatial_size_1_0_spatial
+300 1 0 size 0/1 ood_300_spatial_size_1_0_size
+301 1 0 spatial 0/1 ood_301_spatial_size_1_0_spatial
+302 1 0 size 0/1 ood_302_spatial_size_1_0_size
+303 0 1 spatial 1/1 ood_303_spatial_size_0_1_spatial
+304 0 1 size 0/1 ood_304_spatial_size_0_1_size
+305 2 3 spatial 0/1 ood_305_spatial_size_2_3_spatial
+306 2 3 size 1/1 ood_306_spatial_size_2_3_size
+307 0 1 spatial 1/1 ood_307_spatial_size_0_1_spatial
+308 0 1 size 0/1 ood_308_spatial_size_0_1_size
+309 1 0 spatial 0/1 ood_309_spatial_size_1_0_spatial
+310 1 0 size 0/1 ood_310_spatial_size_1_0_size
+311 1 0 spatial 0/1 ood_311_spatial_size_1_0_spatial
+312 1 0 size 0/1 ood_312_spatial_size_1_0_size
+313 0 1 spatial 1/1 ood_313_spatial_size_0_1_spatial
+314 0 1 size 0/1 ood_314_spatial_size_0_1_size
+315 0 1 spatial 1/1 ood_315_spatial_size_0_1_spatial
+316 0 1 size 0/1 ood_316_spatial_size_0_1_size
+317 0 1 spatial 0/1 ood_317_spatial_size_0_1_spatial
+318 0 1 size 0/1 ood_318_spatial_size_0_1_size
+319 1 0 spatial 0/1 ood_319_spatial_size_1_0_spatial
+320 1 0 size 0/1 ood_320_spatial_size_1_0_size
+321 0 1 spatial 1/1 ood_321_spatial_size_0_1_spatial
+322 0 1 size 0/1 ood_322_spatial_size_0_1_size
+323 0 1 spatial 1/1 ood_323_spatial_size_0_1_spatial
+324 0 1 size 0/1 ood_324_spatial_size_0_1_size
+325 2 3 spatial 0/1 ood_325_spatial_size_2_3_spatial
+326 2 3 size 0/1 ood_326_spatial_size_2_3_size
+327 0 1 spatial 0/1 ood_327_spatial_size_0_1_spatial
+328 0 1 size 0/1 ood_328_spatial_size_0_1_size
+329 1 0 spatial 0/1 ood_329_spatial_size_1_0_spatial
+330 1 0 size 0/1 ood_330_spatial_size_1_0_size
+331 2 3 spatial 0/1 ood_331_spatial_size_2_3_spatial
+332 2 3 size 1/1 ood_332_spatial_size_2_3_size
+333 3 2 spatial 0/1 ood_333_spatial_size_3_2_spatial
+334 3 2 size 0/1 ood_334_spatial_size_3_2_size
+335 1 0 spatial 0/1 ood_335_spatial_size_1_0_spatial
+336 1 0 size 0/1 ood_336_spatial_size_1_0_size
+337 1 0 spatial 0/1 ood_337_spatial_size_1_0_spatial
+338 1 0 size 0/1 ood_338_spatial_size_1_0_size
+339 3 2 spatial 1/1 ood_339_spatial_size_3_2_spatial
+340 3 2 size 0/1 ood_340_spatial_size_3_2_size
+341 1 0 spatial 0/1 ood_341_spatial_size_1_0_spatial
+342 1 0 size 0/1 ood_342_spatial_size_1_0_size
+343 3 2 spatial 1/1 ood_343_spatial_size_3_2_spatial
+344 3 2 size 0/1 ood_344_spatial_size_3_2_size
+345 3 2 spatial 1/1 ood_345_spatial_size_3_2_spatial
+346 3 2 size 0/1 ood_346_spatial_size_3_2_size
+347 1 0 spatial 0/1 ood_347_spatial_size_1_0_spatial
+348 1 0 size 0/1 ood_348_spatial_size_1_0_size
+349 0 1 spatial 1/1 ood_349_spatial_size_0_1_spatial
+350 0 1 size 0/1 ood_350_spatial_size_0_1_size
+351 0 1 spatial 0/1 ood_351_spatial_size_0_1_spatial
+352 0 1 size 0/1 ood_352_spatial_size_0_1_size
+353 3 2 spatial 0/1 ood_353_spatial_size_3_2_spatial
+354 3 2 size 0/1 ood_354_spatial_size_3_2_size
+355 2 3 spatial 0/1 ood_355_spatial_size_2_3_spatial
+356 2 3 size 1/1 ood_356_spatial_size_2_3_size
+357 3 2 spatial 0/1 ood_357_spatial_size_3_2_spatial
+358 3 2 size 0/1 ood_358_spatial_size_3_2_size
+359 3 2 spatial 1/1 ood_359_spatial_size_3_2_spatial
+360 3 2 size 0/1 ood_360_spatial_size_3_2_size
+361 3 2 spatial 1/1 ood_361_spatial_size_3_2_spatial
+362 3 2 size 0/1 ood_362_spatial_size_3_2_size
+363 0 1 spatial 1/1 ood_363_spatial_size_0_1_spatial
+364 0 1 size 0/1 ood_364_spatial_size_0_1_size
+365 0 1 spatial 1/1 ood_365_spatial_size_0_1_spatial
+366 0 1 size 0/1 ood_366_spatial_size_0_1_size
+367 0 1 spatial 1/1 ood_367_spatial_size_0_1_spatial
+368 0 1 size 0/1 ood_368_spatial_size_0_1_size
+369 3 2 spatial 1/1 ood_369_spatial_size_3_2_spatial
+370 3 2 size 0/1 ood_370_spatial_size_3_2_size
+371 2 3 spatial 0/1 ood_371_spatial_size_2_3_spatial
+372 2 3 size 1/1 ood_372_spatial_size_2_3_size
+373 0 1 spatial 1/1 ood_373_spatial_size_0_1_spatial
+374 0 1 size 0/1 ood_374_spatial_size_0_1_size
+375 1 0 spatial 0/1 ood_375_spatial_size_1_0_spatial
+376 1 0 size 0/1 ood_376_spatial_size_1_0_size
+377 1 0 spatial 0/1 ood_377_spatial_size_1_0_spatial
+378 1 0 size 0/1 ood_378_spatial_size_1_0_size
+379 1 0 spatial 0/1 ood_379_spatial_size_1_0_spatial
+380 1 0 size 0/1 ood_380_spatial_size_1_0_size
+381 3 2 spatial 1/1 ood_381_spatial_size_3_2_spatial
+382 3 2 size 0/1 ood_382_spatial_size_3_2_size
+383 1 0 spatial 0/1 ood_383_spatial_size_1_0_spatial
+384 1 0 size 0/1 ood_384_spatial_size_1_0_size
+385 3 2 spatial 1/1 ood_385_spatial_size_3_2_spatial
+386 3 2 size 0/1 ood_386_spatial_size_3_2_size
+387 1 0 spatial 0/1 ood_387_spatial_size_1_0_spatial
+388 1 0 size 0/1 ood_388_spatial_size_1_0_size
+389 2 3 spatial 0/1 ood_389_spatial_size_2_3_spatial
+390 2 3 size 1/1 ood_390_spatial_size_2_3_size
+391 3 2 spatial 1/1 ood_391_spatial_size_3_2_spatial
+392 3 2 size 0/1 ood_392_spatial_size_3_2_size
+393 1 0 spatial 1/1 ood_393_spatial_size_1_0_spatial
+394 1 0 size 0/1 ood_394_spatial_size_1_0_size
+395 0 1 spatial 0/1 ood_395_spatial_size_0_1_spatial
+396 0 1 size 0/1 ood_396_spatial_size_0_1_size
+397 3 2 spatial 1/1 ood_397_spatial_size_3_2_spatial
+398 3 2 size 0/1 ood_398_spatial_size_3_2_size
+399 0 1 spatial 1/1 ood_399_spatial_size_0_1_spatial
+400 0 1 size 0/1 ood_400_spatial_size_0_1_size
+
+overall_spatial_success=61/200 (30.5%)
+overall_size_success=37/200 (18.5%)

results/genie/conflict_env/genie_verb_color_ood_seed42.txt

@@ -0,0 +1,421 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=verb_color
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/verb_color
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 4 0 verb 0/1 ood_001_verb_color_4_0_verb
+2 4 0 color 0/1 ood_002_verb_color_4_0_color
+3 0 4 verb 0/1 ood_003_verb_color_0_4_verb
+4 0 4 color 1/1 ood_004_verb_color_0_4_color
+5 0 1 verb 0/1 ood_005_verb_color_0_1_verb
+6 0 1 color 1/1 ood_006_verb_color_0_1_color
+7 4 3 verb 0/1 ood_007_verb_color_4_3_verb
+8 4 3 color 0/1 ood_008_verb_color_4_3_color
+9 1 4 verb 0/1 ood_009_verb_color_1_4_verb
+10 1 4 color 1/1 ood_010_verb_color_1_4_color
+11 1 3 verb 1/1 ood_011_verb_color_1_3_verb
+12 1 3 color 0/1 ood_012_verb_color_1_3_color
+13 1 3 verb 0/1 ood_013_verb_color_1_3_verb
+14 1 3 color 0/1 ood_014_verb_color_1_3_color
+15 0 5 verb 1/1 ood_015_verb_color_0_5_verb
+16 0 5 color 1/1 ood_016_verb_color_0_5_color
+17 4 3 verb 0/1 ood_017_verb_color_4_3_verb
+18 4 3 color 0/1 ood_018_verb_color_4_3_color
+19 0 4 verb 0/1 ood_019_verb_color_0_4_verb
+20 0 4 color 0/1 ood_020_verb_color_0_4_color
+21 4 1 verb 0/1 ood_021_verb_color_4_1_verb
+22 4 1 color 0/1 ood_022_verb_color_4_1_color
+23 4 3 verb 1/1 ood_023_verb_color_4_3_verb
+24 4 3 color 0/1 ood_024_verb_color_4_3_color
+25 5 3 verb 0/1 ood_025_verb_color_5_3_verb
+26 5 3 color 0/1 ood_026_verb_color_5_3_color
+27 3 2 verb 0/1 ood_027_verb_color_3_2_verb
+28 3 2 color 0/1 ood_028_verb_color_3_2_color
+29 0 3 verb 0/1 ood_029_verb_color_0_3_verb
+30 0 3 color 0/1 ood_030_verb_color_0_3_color
+31 3 4 verb 0/1 ood_031_verb_color_3_4_verb
+32 3 4 color 0/1 ood_032_verb_color_3_4_color
+33 2 4 verb 0/1 ood_033_verb_color_2_4_verb
+34 2 4 color 0/1 ood_034_verb_color_2_4_color
+35 0 2 verb 0/1 ood_035_verb_color_0_2_verb
+36 0 2 color 0/1 ood_036_verb_color_0_2_color
+37 0 1 verb 0/1 ood_037_verb_color_0_1_verb
+38 0 1 color 0/1 ood_038_verb_color_0_1_color
+39 0 3 verb 0/1 ood_039_verb_color_0_3_verb
+40 0 3 color 0/1 ood_040_verb_color_0_3_color
+41 1 2 verb 1/1 ood_041_verb_color_1_2_verb
+42 1 2 color 0/1 ood_042_verb_color_1_2_color
+43 1 3 verb 0/1 ood_043_verb_color_1_3_verb
+44 1 3 color 0/1 ood_044_verb_color_1_3_color
+45 3 1 verb 0/1 ood_045_verb_color_3_1_verb
+46 3 1 color 0/1 ood_046_verb_color_3_1_color
+47 3 5 verb 0/1 ood_047_verb_color_3_5_verb
+48 3 5 color 0/1 ood_048_verb_color_3_5_color
+49 0 1 verb 0/1 ood_049_verb_color_0_1_verb
+50 0 1 color 0/1 ood_050_verb_color_0_1_color
+51 3 2 verb 0/1 ood_051_verb_color_3_2_verb
+52 3 2 color 0/1 ood_052_verb_color_3_2_color
+53 1 2 verb 0/1 ood_053_verb_color_1_2_verb
+54 1 2 color 0/1 ood_054_verb_color_1_2_color
+55 4 2 verb 0/1 ood_055_verb_color_4_2_verb
+56 4 2 color 0/1 ood_056_verb_color_4_2_color
+57 4 0 verb 0/1 ood_057_verb_color_4_0_verb
+58 4 0 color 0/1 ood_058_verb_color_4_0_color
+59 4 2 verb 0/1 ood_059_verb_color_4_2_verb
+60 4 2 color 0/1 ood_060_verb_color_4_2_color
+61 3 2 verb 0/1 ood_061_verb_color_3_2_verb
+62 3 2 color 0/1 ood_062_verb_color_3_2_color
+63 2 4 verb 0/1 ood_063_verb_color_2_4_verb
+64 2 4 color 0/1 ood_064_verb_color_2_4_color
+65 1 3 verb 0/1 ood_065_verb_color_1_3_verb
+66 1 3 color 0/1 ood_066_verb_color_1_3_color
+67 2 5 verb 0/1 ood_067_verb_color_2_5_verb
+68 2 5 color 0/1 ood_068_verb_color_2_5_color
+69 3 4 verb 0/1 ood_069_verb_color_3_4_verb
+70 3 4 color 1/1 ood_070_verb_color_3_4_color
+71 1 4 verb 0/1 ood_071_verb_color_1_4_verb
+72 1 4 color 1/1 ood_072_verb_color_1_4_color
+73 5 0 verb 0/1 ood_073_verb_color_5_0_verb
+74 5 0 color 0/1 ood_074_verb_color_5_0_color
+75 5 2 verb 0/1 ood_075_verb_color_5_2_verb
+76 5 2 color 0/1 ood_076_verb_color_5_2_color
+77 0 1 verb 1/1 ood_077_verb_color_0_1_verb
+78 0 1 color 0/1 ood_078_verb_color_0_1_color
+79 4 5 verb 1/1 ood_079_verb_color_4_5_verb
+80 4 5 color 0/1 ood_080_verb_color_4_5_color
+81 5 0 verb 0/1 ood_081_verb_color_5_0_verb
+82 5 0 color 0/1 ood_082_verb_color_5_0_color
+83 1 0 verb 0/1 ood_083_verb_color_1_0_verb
+84 1 0 color 0/1 ood_084_verb_color_1_0_color
+85 4 2 verb 1/1 ood_085_verb_color_4_2_verb
+86 4 2 color 0/1 ood_086_verb_color_4_2_color
+87 2 4 verb 0/1 ood_087_verb_color_2_4_verb
+88 2 4 color 0/1 ood_088_verb_color_2_4_color
+89 2 0 verb 0/1 ood_089_verb_color_2_0_verb
+90 2 0 color 0/1 ood_090_verb_color_2_0_color
+91 1 4 verb 0/1 ood_091_verb_color_1_4_verb
+92 1 4 color 0/1 ood_092_verb_color_1_4_color
+93 0 5 verb 0/1 ood_093_verb_color_0_5_verb
+94 0 5 color 0/1 ood_094_verb_color_0_5_color
+95 1 2 verb 0/1 ood_095_verb_color_1_2_verb
+96 1 2 color 0/1 ood_096_verb_color_1_2_color
+97 4 5 verb 0/1 ood_097_verb_color_4_5_verb
+98 4 5 color 0/1 ood_098_verb_color_4_5_color
+99 2 0 verb 0/1 ood_099_verb_color_2_0_verb
+100 2 0 color 0/1 ood_100_verb_color_2_0_color
+101 0 4 verb 0/1 ood_101_verb_color_0_4_verb
+102 0 4 color 0/1 ood_102_verb_color_0_4_color
+103 0 3 verb 0/1 ood_103_verb_color_0_3_verb
+104 0 3 color 0/1 ood_104_verb_color_0_3_color
+105 2 3 verb 0/1 ood_105_verb_color_2_3_verb
+106 2 3 color 1/1 ood_106_verb_color_2_3_color
+107 0 4 verb 0/1 ood_107_verb_color_0_4_verb
+108 0 4 color 0/1 ood_108_verb_color_0_4_color
+109 2 1 verb 0/1 ood_109_verb_color_2_1_verb
+110 2 1 color 0/1 ood_110_verb_color_2_1_color
+111 5 2 verb 0/1 ood_111_verb_color_5_2_verb
+112 5 2 color 0/1 ood_112_verb_color_5_2_color
+113 2 1 verb 0/1 ood_113_verb_color_2_1_verb
+114 2 1 color 0/1 ood_114_verb_color_2_1_color
+115 3 5 verb 0/1 ood_115_verb_color_3_5_verb
+116 3 5 color 0/1 ood_116_verb_color_3_5_color
+117 1 4 verb 0/1 ood_117_verb_color_1_4_verb
+118 1 4 color 0/1 ood_118_verb_color_1_4_color
+119 5 0 verb 0/1 ood_119_verb_color_5_0_verb
+120 5 0 color 0/1 ood_120_verb_color_5_0_color
+121 0 2 verb 0/1 ood_121_verb_color_0_2_verb
+122 0 2 color 0/1 ood_122_verb_color_0_2_color
+123 4 3 verb 1/1 ood_123_verb_color_4_3_verb
+124 4 3 color 0/1 ood_124_verb_color_4_3_color
+125 2 5 verb 0/1 ood_125_verb_color_2_5_verb
+126 2 5 color 0/1 ood_126_verb_color_2_5_color
+127 3 2 verb 0/1 ood_127_verb_color_3_2_verb
+128 3 2 color 0/1 ood_128_verb_color_3_2_color
+129 0 4 verb 0/1 ood_129_verb_color_0_4_verb
+130 0 4 color 0/1 ood_130_verb_color_0_4_color
+131 5 4 verb 0/1 ood_131_verb_color_5_4_verb
+132 5 4 color 0/1 ood_132_verb_color_5_4_color
+133 2 3 verb 0/1 ood_133_verb_color_2_3_verb
+134 2 3 color 0/1 ood_134_verb_color_2_3_color
+135 0 3 verb 0/1 ood_135_verb_color_0_3_verb
+136 0 3 color 0/1 ood_136_verb_color_0_3_color
+137 3 2 verb 0/1 ood_137_verb_color_3_2_verb
+138 3 2 color 0/1 ood_138_verb_color_3_2_color
+139 1 5 verb 1/1 ood_139_verb_color_1_5_verb
+140 1 5 color 0/1 ood_140_verb_color_1_5_color
+141 5 1 verb 0/1 ood_141_verb_color_5_1_verb
+142 5 1 color 0/1 ood_142_verb_color_5_1_color
+143 4 0 verb 0/1 ood_143_verb_color_4_0_verb
+144 4 0 color 0/1 ood_144_verb_color_4_0_color
+145 3 5 verb 0/1 ood_145_verb_color_3_5_verb
+146 3 5 color 0/1 ood_146_verb_color_3_5_color
+147 5 3 verb 0/1 ood_147_verb_color_5_3_verb
+148 5 3 color 0/1 ood_148_verb_color_5_3_color
+149 5 2 verb 0/1 ood_149_verb_color_5_2_verb
+150 5 2 color 0/1 ood_150_verb_color_5_2_color
+151 2 1 verb 0/1 ood_151_verb_color_2_1_verb
+152 2 1 color 0/1 ood_152_verb_color_2_1_color
+153 3 4 verb 0/1 ood_153_verb_color_3_4_verb
+154 3 4 color 0/1 ood_154_verb_color_3_4_color
+155 1 2 verb 0/1 ood_155_verb_color_1_2_verb
+156 1 2 color 1/1 ood_156_verb_color_1_2_color
+157 4 2 verb 1/1 ood_157_verb_color_4_2_verb
+158 4 2 color 0/1 ood_158_verb_color_4_2_color
+159 0 3 verb 0/1 ood_159_verb_color_0_3_verb
+160 0 3 color 0/1 ood_160_verb_color_0_3_color
+161 0 2 verb 0/1 ood_161_verb_color_0_2_verb
+162 0 2 color 0/1 ood_162_verb_color_0_2_color
+163 4 1 verb 0/1 ood_163_verb_color_4_1_verb
+164 4 1 color 1/1 ood_164_verb_color_4_1_color
+165 1 3 verb 1/1 ood_165_verb_color_1_3_verb
+166 1 3 color 0/1 ood_166_verb_color_1_3_color
+167 4 5 verb 0/1 ood_167_verb_color_4_5_verb
+168 4 5 color 0/1 ood_168_verb_color_4_5_color
+169 1 5 verb 1/1 ood_169_verb_color_1_5_verb
+170 1 5 color 0/1 ood_170_verb_color_1_5_color
+171 0 3 verb 0/1 ood_171_verb_color_0_3_verb
+172 0 3 color 0/1 ood_172_verb_color_0_3_color
+173 5 2 verb 0/1 ood_173_verb_color_5_2_verb
+174 5 2 color 0/1 ood_174_verb_color_5_2_color
+175 1 3 verb 0/1 ood_175_verb_color_1_3_verb
+176 1 3 color 0/1 ood_176_verb_color_1_3_color
+177 5 2 verb 0/1 ood_177_verb_color_5_2_verb
+178 5 2 color 0/1 ood_178_verb_color_5_2_color
+179 0 4 verb 0/1 ood_179_verb_color_0_4_verb
+180 0 4 color 1/1 ood_180_verb_color_0_4_color
+181 2 3 verb 0/1 ood_181_verb_color_2_3_verb
+182 2 3 color 0/1 ood_182_verb_color_2_3_color
+183 1 4 verb 0/1 ood_183_verb_color_1_4_verb
+184 1 4 color 0/1 ood_184_verb_color_1_4_color
+185 2 5 verb 0/1 ood_185_verb_color_2_5_verb
+186 2 5 color 0/1 ood_186_verb_color_2_5_color
+187 4 0 verb 0/1 ood_187_verb_color_4_0_verb
+188 4 0 color 0/1 ood_188_verb_color_4_0_color
+189 5 1 verb 0/1 ood_189_verb_color_5_1_verb
+190 5 1 color 1/1 ood_190_verb_color_5_1_color
+191 2 1 verb 0/1 ood_191_verb_color_2_1_verb
+192 2 1 color 0/1 ood_192_verb_color_2_1_color
+193 1 0 verb 0/1 ood_193_verb_color_1_0_verb
+194 1 0 color 0/1 ood_194_verb_color_1_0_color
+195 2 1 verb 0/1 ood_195_verb_color_2_1_verb
+196 2 1 color 1/1 ood_196_verb_color_2_1_color
+197 2 1 verb 0/1 ood_197_verb_color_2_1_verb
+198 2 1 color 1/1 ood_198_verb_color_2_1_color
+199 1 2 verb 1/1 ood_199_verb_color_1_2_verb
+200 1 2 color 0/1 ood_200_verb_color_1_2_color
+201 4 1 verb 0/1 ood_201_verb_color_4_1_verb
+202 4 1 color 1/1 ood_202_verb_color_4_1_color
+203 1 4 verb 1/1 ood_203_verb_color_1_4_verb
+204 1 4 color 0/1 ood_204_verb_color_1_4_color
+205 4 2 verb 0/1 ood_205_verb_color_4_2_verb
+206 4 2 color 0/1 ood_206_verb_color_4_2_color
+207 5 4 verb 0/1 ood_207_verb_color_5_4_verb
+208 5 4 color 0/1 ood_208_verb_color_5_4_color
+209 4 1 verb 0/1 ood_209_verb_color_4_1_verb
+210 4 1 color 0/1 ood_210_verb_color_4_1_color
+211 4 0 verb 1/1 ood_211_verb_color_4_0_verb
+212 4 0 color 0/1 ood_212_verb_color_4_0_color
+213 0 3 verb 0/1 ood_213_verb_color_0_3_verb
+214 0 3 color 0/1 ood_214_verb_color_0_3_color
+215 3 5 verb 0/1 ood_215_verb_color_3_5_verb
+216 3 5 color 1/1 ood_216_verb_color_3_5_color
+217 4 0 verb 0/1 ood_217_verb_color_4_0_verb
+218 4 0 color 0/1 ood_218_verb_color_4_0_color
+219 1 0 verb 0/1 ood_219_verb_color_1_0_verb
+220 1 0 color 1/1 ood_220_verb_color_1_0_color
+221 3 2 verb 0/1 ood_221_verb_color_3_2_verb
+222 3 2 color 0/1 ood_222_verb_color_3_2_color
+223 4 3 verb 0/1 ood_223_verb_color_4_3_verb
+224 4 3 color 0/1 ood_224_verb_color_4_3_color
+225 1 3 verb 0/1 ood_225_verb_color_1_3_verb
+226 1 3 color 0/1 ood_226_verb_color_1_3_color
+227 1 0 verb 0/1 ood_227_verb_color_1_0_verb
+228 1 0 color 0/1 ood_228_verb_color_1_0_color
+229 2 5 verb 0/1 ood_229_verb_color_2_5_verb
+230 2 5 color 0/1 ood_230_verb_color_2_5_color
+231 2 3 verb 0/1 ood_231_verb_color_2_3_verb
+232 2 3 color 0/1 ood_232_verb_color_2_3_color
+233 1 4 verb 0/1 ood_233_verb_color_1_4_verb
+234 1 4 color 0/1 ood_234_verb_color_1_4_color
+235 5 4 verb 1/1 ood_235_verb_color_5_4_verb
+236 5 4 color 0/1 ood_236_verb_color_5_4_color
+237 4 0 verb 0/1 ood_237_verb_color_4_0_verb
+238 4 0 color 0/1 ood_238_verb_color_4_0_color
+239 4 2 verb 1/1 ood_239_verb_color_4_2_verb
+240 4 2 color 0/1 ood_240_verb_color_4_2_color
+241 3 2 verb 0/1 ood_241_verb_color_3_2_verb
+242 3 2 color 0/1 ood_242_verb_color_3_2_color
+243 1 3 verb 1/1 ood_243_verb_color_1_3_verb
+244 1 3 color 0/1 ood_244_verb_color_1_3_color
+245 4 1 verb 0/1 ood_245_verb_color_4_1_verb
+246 4 1 color 0/1 ood_246_verb_color_4_1_color
+247 2 0 verb 0/1 ood_247_verb_color_2_0_verb
+248 2 0 color 0/1 ood_248_verb_color_2_0_color
+249 5 1 verb 1/1 ood_249_verb_color_5_1_verb
+250 5 1 color 0/1 ood_250_verb_color_5_1_color
+251 4 5 verb 0/1 ood_251_verb_color_4_5_verb
+252 4 5 color 0/1 ood_252_verb_color_4_5_color
+253 4 5 verb 0/1 ood_253_verb_color_4_5_verb
+254 4 5 color 0/1 ood_254_verb_color_4_5_color
+255 0 2 verb 0/1 ood_255_verb_color_0_2_verb
+256 0 2 color 0/1 ood_256_verb_color_0_2_color
+257 1 3 verb 1/1 ood_257_verb_color_1_3_verb
+258 1 3 color 0/1 ood_258_verb_color_1_3_color
+259 5 1 verb 0/1 ood_259_verb_color_5_1_verb
+260 5 1 color 0/1 ood_260_verb_color_5_1_color
+261 0 2 verb 0/1 ood_261_verb_color_0_2_verb
+262 0 2 color 0/1 ood_262_verb_color_0_2_color
+263 5 0 verb 0/1 ood_263_verb_color_5_0_verb
+264 5 0 color 0/1 ood_264_verb_color_5_0_color
+265 2 0 verb 0/1 ood_265_verb_color_2_0_verb
+266 2 0 color 0/1 ood_266_verb_color_2_0_color
+267 2 3 verb 0/1 ood_267_verb_color_2_3_verb
+268 2 3 color 0/1 ood_268_verb_color_2_3_color
+269 1 4 verb 1/1 ood_269_verb_color_1_4_verb
+270 1 4 color 0/1 ood_270_verb_color_1_4_color
+271 0 3 verb 0/1 ood_271_verb_color_0_3_verb
+272 0 3 color 0/1 ood_272_verb_color_0_3_color
+273 1 2 verb 0/1 ood_273_verb_color_1_2_verb
+274 1 2 color 0/1 ood_274_verb_color_1_2_color
+275 5 4 verb 0/1 ood_275_verb_color_5_4_verb
+276 5 4 color 0/1 ood_276_verb_color_5_4_color
+277 3 4 verb 0/1 ood_277_verb_color_3_4_verb
+278 3 4 color 1/1 ood_278_verb_color_3_4_color
+279 5 3 verb 0/1 ood_279_verb_color_5_3_verb
+280 5 3 color 0/1 ood_280_verb_color_5_3_color
+281 4 2 verb 1/1 ood_281_verb_color_4_2_verb
+282 4 2 color 0/1 ood_282_verb_color_4_2_color
+283 2 0 verb 0/1 ood_283_verb_color_2_0_verb
+284 2 0 color 0/1 ood_284_verb_color_2_0_color
+285 1 2 verb 0/1 ood_285_verb_color_1_2_verb
+286 1 2 color 0/1 ood_286_verb_color_1_2_color
+287 4 0 verb 0/1 ood_287_verb_color_4_0_verb
+288 4 0 color 0/1 ood_288_verb_color_4_0_color
+289 3 0 verb 0/1 ood_289_verb_color_3_0_verb
+290 3 0 color 0/1 ood_290_verb_color_3_0_color
+291 2 3 verb 0/1 ood_291_verb_color_2_3_verb
+292 2 3 color 0/1 ood_292_verb_color_2_3_color
+293 5 3 verb 0/1 ood_293_verb_color_5_3_verb
+294 5 3 color 0/1 ood_294_verb_color_5_3_color
+295 5 4 verb 0/1 ood_295_verb_color_5_4_verb
+296 5 4 color 0/1 ood_296_verb_color_5_4_color
+297 4 0 verb 0/1 ood_297_verb_color_4_0_verb
+298 4 0 color 0/1 ood_298_verb_color_4_0_color
+299 2 5 verb 0/1 ood_299_verb_color_2_5_verb
+300 2 5 color 1/1 ood_300_verb_color_2_5_color
+301 0 5 verb 0/1 ood_301_verb_color_0_5_verb
+302 0 5 color 0/1 ood_302_verb_color_0_5_color
+303 1 4 verb 0/1 ood_303_verb_color_1_4_verb
+304 1 4 color 0/1 ood_304_verb_color_1_4_color
+305 0 5 verb 0/1 ood_305_verb_color_0_5_verb
+306 0 5 color 1/1 ood_306_verb_color_0_5_color
+307 1 3 verb 0/1 ood_307_verb_color_1_3_verb
+308 1 3 color 0/1 ood_308_verb_color_1_3_color
+309 4 3 verb 0/1 ood_309_verb_color_4_3_verb
+310 4 3 color 0/1 ood_310_verb_color_4_3_color
+311 3 2 verb 0/1 ood_311_verb_color_3_2_verb
+312 3 2 color 0/1 ood_312_verb_color_3_2_color
+313 3 2 verb 0/1 ood_313_verb_color_3_2_verb
+314 3 2 color 0/1 ood_314_verb_color_3_2_color
+315 1 4 verb 0/1 ood_315_verb_color_1_4_verb
+316 1 4 color 0/1 ood_316_verb_color_1_4_color
+317 4 3 verb 0/1 ood_317_verb_color_4_3_verb
+318 4 3 color 0/1 ood_318_verb_color_4_3_color
+319 3 4 verb 0/1 ood_319_verb_color_3_4_verb
+320 3 4 color 0/1 ood_320_verb_color_3_4_color
+321 2 4 verb 0/1 ood_321_verb_color_2_4_verb
+322 2 4 color 0/1 ood_322_verb_color_2_4_color
+323 5 3 verb 0/1 ood_323_verb_color_5_3_verb
+324 5 3 color 0/1 ood_324_verb_color_5_3_color
+325 3 4 verb 0/1 ood_325_verb_color_3_4_verb
+326 3 4 color 0/1 ood_326_verb_color_3_4_color
+327 2 3 verb 0/1 ood_327_verb_color_2_3_verb
+328 2 3 color 0/1 ood_328_verb_color_2_3_color
+329 2 1 verb 0/1 ood_329_verb_color_2_1_verb
+330 2 1 color 0/1 ood_330_verb_color_2_1_color
+331 1 3 verb 0/1 ood_331_verb_color_1_3_verb
+332 1 3 color 0/1 ood_332_verb_color_1_3_color
+333 0 5 verb 0/1 ood_333_verb_color_0_5_verb
+334 0 5 color 0/1 ood_334_verb_color_0_5_color
+335 3 1 verb 0/1 ood_335_verb_color_3_1_verb
+336 3 1 color 0/1 ood_336_verb_color_3_1_color
+337 3 0 verb 0/1 ood_337_verb_color_3_0_verb
+338 3 0 color 1/1 ood_338_verb_color_3_0_color
+339 0 3 verb 1/1 ood_339_verb_color_0_3_verb
+340 0 3 color 0/1 ood_340_verb_color_0_3_color
+341 4 5 verb 0/1 ood_341_verb_color_4_5_verb
+342 4 5 color 0/1 ood_342_verb_color_4_5_color
+343 0 2 verb 0/1 ood_343_verb_color_0_2_verb
+344 0 2 color 0/1 ood_344_verb_color_0_2_color
+345 5 2 verb 0/1 ood_345_verb_color_5_2_verb
+346 5 2 color 0/1 ood_346_verb_color_5_2_color
+347 0 4 verb 0/1 ood_347_verb_color_0_4_verb
+348 0 4 color 0/1 ood_348_verb_color_0_4_color
+349 0 5 verb 0/1 ood_349_verb_color_0_5_verb
+350 0 5 color 0/1 ood_350_verb_color_0_5_color
+351 4 0 verb 0/1 ood_351_verb_color_4_0_verb
+352 4 0 color 0/1 ood_352_verb_color_4_0_color
+353 1 0 verb 1/1 ood_353_verb_color_1_0_verb
+354 1 0 color 0/1 ood_354_verb_color_1_0_color
+355 5 0 verb 0/1 ood_355_verb_color_5_0_verb
+356 5 0 color 0/1 ood_356_verb_color_5_0_color
+357 4 1 verb 0/1 ood_357_verb_color_4_1_verb
+358 4 1 color 0/1 ood_358_verb_color_4_1_color
+359 2 4 verb 0/1 ood_359_verb_color_2_4_verb
+360 2 4 color 0/1 ood_360_verb_color_2_4_color
+361 3 5 verb 0/1 ood_361_verb_color_3_5_verb
+362 3 5 color 1/1 ood_362_verb_color_3_5_color
+363 0 3 verb 0/1 ood_363_verb_color_0_3_verb
+364 0 3 color 0/1 ood_364_verb_color_0_3_color
+365 2 3 verb 1/1 ood_365_verb_color_2_3_verb
+366 2 3 color 0/1 ood_366_verb_color_2_3_color
+367 2 3 verb 0/1 ood_367_verb_color_2_3_verb
+368 2 3 color 0/1 ood_368_verb_color_2_3_color
+369 3 5 verb 0/1 ood_369_verb_color_3_5_verb
+370 3 5 color 0/1 ood_370_verb_color_3_5_color
+371 2 5 verb 0/1 ood_371_verb_color_2_5_verb
+372 2 5 color 0/1 ood_372_verb_color_2_5_color
+373 3 1 verb 0/1 ood_373_verb_color_3_1_verb
+374 3 1 color 0/1 ood_374_verb_color_3_1_color
+375 1 4 verb 1/1 ood_375_verb_color_1_4_verb
+376 1 4 color 0/1 ood_376_verb_color_1_4_color
+377 3 2 verb 0/1 ood_377_verb_color_3_2_verb
+378 3 2 color 0/1 ood_378_verb_color_3_2_color
+379 5 2 verb 0/1 ood_379_verb_color_5_2_verb
+380 5 2 color 0/1 ood_380_verb_color_5_2_color
+381 0 1 verb 0/1 ood_381_verb_color_0_1_verb
+382 0 1 color 1/1 ood_382_verb_color_0_1_color
+383 4 1 verb 1/1 ood_383_verb_color_4_1_verb
+384 4 1 color 0/1 ood_384_verb_color_4_1_color
+385 4 3 verb 0/1 ood_385_verb_color_4_3_verb
+386 4 3 color 0/1 ood_386_verb_color_4_3_color
+387 0 4 verb 0/1 ood_387_verb_color_0_4_verb
+388 0 4 color 0/1 ood_388_verb_color_0_4_color
+389 4 1 verb 0/1 ood_389_verb_color_4_1_verb
+390 4 1 color 0/1 ood_390_verb_color_4_1_color
+391 5 3 verb 0/1 ood_391_verb_color_5_3_verb
+392 5 3 color 0/1 ood_392_verb_color_5_3_color
+393 3 2 verb 0/1 ood_393_verb_color_3_2_verb
+394 3 2 color 0/1 ood_394_verb_color_3_2_color
+395 4 5 verb 0/1 ood_395_verb_color_4_5_verb
+396 4 5 color 0/1 ood_396_verb_color_4_5_color
+397 1 4 verb 0/1 ood_397_verb_color_1_4_verb
+398 1 4 color 0/1 ood_398_verb_color_1_4_color
+399 4 5 verb 0/1 ood_399_verb_color_4_5_verb
+400 4 5 color 1/1 ood_400_verb_color_4_5_color
+
+overall_verb_success=27/200 (13.5%)
+overall_color_success=23/200 (11.5%)

results/genie/conflict_env/genie_verb_object_ood_seed42.txt

@@ -0,0 +1,420 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=verb_object
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/verb_object
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 4 0 verb 0/1 ood_001_verb_object_4_0_verb
+2 4 0 shape 0/1 ood_002_verb_object_4_0_shape
+3 0 4 verb 0/1 ood_003_verb_object_0_4_verb
+4 0 4 shape 0/1 ood_004_verb_object_0_4_shape
+5 0 1 verb 0/1 ood_005_verb_object_0_1_verb
+6 0 1 shape 0/1 ood_006_verb_object_0_1_shape
+7 4 3 verb 0/1 ood_007_verb_object_4_3_verb
+8 4 3 shape 0/1 ood_008_verb_object_4_3_shape
+9 1 4 verb 0/1 ood_009_verb_object_1_4_verb
+10 1 4 shape 0/1 ood_010_verb_object_1_4_shape
+11 1 3 verb 0/1 ood_011_verb_object_1_3_verb
+12 1 3 shape 0/1 ood_012_verb_object_1_3_shape
+13 1 3 verb 0/1 ood_013_verb_object_1_3_verb
+14 1 3 shape 0/1 ood_014_verb_object_1_3_shape
+15 0 5 verb 0/1 ood_015_verb_object_0_5_verb
+16 0 5 shape 0/1 ood_016_verb_object_0_5_shape
+17 4 3 verb 0/1 ood_017_verb_object_4_3_verb
+18 4 3 shape 0/1 ood_018_verb_object_4_3_shape
+19 0 4 verb 0/1 ood_019_verb_object_0_4_verb
+20 0 4 shape 0/1 ood_020_verb_object_0_4_shape
+21 4 1 verb 0/1 ood_021_verb_object_4_1_verb
+22 4 1 shape 1/1 ood_022_verb_object_4_1_shape
+23 4 3 verb 0/1 ood_023_verb_object_4_3_verb
+24 4 3 shape 0/1 ood_024_verb_object_4_3_shape
+25 5 3 verb 0/1 ood_025_verb_object_5_3_verb
+26 5 3 shape 0/1 ood_026_verb_object_5_3_shape
+27 3 2 verb 0/1 ood_027_verb_object_3_2_verb
+28 3 2 shape 0/1 ood_028_verb_object_3_2_shape
+29 0 3 verb 0/1 ood_029_verb_object_0_3_verb
+30 0 3 shape 0/1 ood_030_verb_object_0_3_shape
+31 3 4 verb 0/1 ood_031_verb_object_3_4_verb
+32 3 4 shape 1/1 ood_032_verb_object_3_4_shape
+33 2 4 verb 0/1 ood_033_verb_object_2_4_verb
+34 2 4 shape 0/1 ood_034_verb_object_2_4_shape
+35 0 2 verb 0/1 ood_035_verb_object_0_2_verb
+36 0 2 shape 0/1 ood_036_verb_object_0_2_shape
+37 0 1 verb 0/1 ood_037_verb_object_0_1_verb
+38 0 1 shape 1/1 ood_038_verb_object_0_1_shape
+39 0 3 verb 0/1 ood_039_verb_object_0_3_verb
+40 0 3 shape 0/1 ood_040_verb_object_0_3_shape
+41 1 2 verb 0/1 ood_041_verb_object_1_2_verb
+42 1 2 shape 0/1 ood_042_verb_object_1_2_shape
+43 1 3 verb 0/1 ood_043_verb_object_1_3_verb
+44 1 3 shape 0/1 ood_044_verb_object_1_3_shape
+45 3 1 verb 0/1 ood_045_verb_object_3_1_verb
+46 3 1 shape 0/1 ood_046_verb_object_3_1_shape
+47 3 5 verb 0/1 ood_047_verb_object_3_5_verb
+48 3 5 shape 0/1 ood_048_verb_object_3_5_shape
+49 0 1 verb 0/1 ood_049_verb_object_0_1_verb
+50 0 1 shape 0/1 ood_050_verb_object_0_1_shape
+51 3 2 verb 0/1 ood_051_verb_object_3_2_verb
+52 3 2 shape 0/1 ood_052_verb_object_3_2_shape
+53 1 2 verb 0/1 ood_053_verb_object_1_2_verb
+54 1 2 shape 0/1 ood_054_verb_object_1_2_shape
+55 4 2 verb 0/1 ood_055_verb_object_4_2_verb
+56 4 2 shape 0/1 ood_056_verb_object_4_2_shape
+57 4 0 verb 0/1 ood_057_verb_object_4_0_verb
+58 4 0 shape 0/1 ood_058_verb_object_4_0_shape
+59 4 2 verb 0/1 ood_059_verb_object_4_2_verb
+60 4 2 shape 0/1 ood_060_verb_object_4_2_shape
+61 3 2 verb 0/1 ood_061_verb_object_3_2_verb
+62 3 2 shape 0/1 ood_062_verb_object_3_2_shape
+63 2 4 verb 0/1 ood_063_verb_object_2_4_verb
+64 2 4 shape 0/1 ood_064_verb_object_2_4_shape
+65 1 3 verb 0/1 ood_065_verb_object_1_3_verb
+66 1 3 shape 0/1 ood_066_verb_object_1_3_shape
+67 2 5 verb 0/1 ood_067_verb_object_2_5_verb
+68 2 5 shape 0/1 ood_068_verb_object_2_5_shape
+69 3 4 verb 0/1 ood_069_verb_object_3_4_verb
+70 3 4 shape 0/1 ood_070_verb_object_3_4_shape
+71 1 4 verb 1/1 ood_071_verb_object_1_4_verb
+72 1 4 shape 1/1 ood_072_verb_object_1_4_shape
+73 5 0 verb 0/1 ood_073_verb_object_5_0_verb
+74 5 0 shape 0/1 ood_074_verb_object_5_0_shape
+75 5 2 verb 0/1 ood_075_verb_object_5_2_verb
+76 5 2 shape 0/1 ood_076_verb_object_5_2_shape
+77 0 1 verb 0/1 ood_077_verb_object_0_1_verb
+78 0 1 shape 0/1 ood_078_verb_object_0_1_shape
+79 4 5 verb 0/1 ood_079_verb_object_4_5_verb
+80 4 5 shape 0/1 ood_080_verb_object_4_5_shape
+81 5 0 verb 0/1 ood_081_verb_object_5_0_verb
+82 5 0 shape 0/1 ood_082_verb_object_5_0_shape
+83 1 0 verb 0/1 ood_083_verb_object_1_0_verb
+84 1 0 shape 0/1 ood_084_verb_object_1_0_shape
+85 4 2 verb 0/1 ood_085_verb_object_4_2_verb
+86 4 2 shape 0/1 ood_086_verb_object_4_2_shape
+87 2 4 verb 0/1 ood_087_verb_object_2_4_verb
+88 2 4 shape 0/1 ood_088_verb_object_2_4_shape
+89 2 0 verb 0/1 ood_089_verb_object_2_0_verb
+90 2 0 shape 0/1 ood_090_verb_object_2_0_shape
+91 1 4 verb 0/1 ood_091_verb_object_1_4_verb
+92 1 4 shape 0/1 ood_092_verb_object_1_4_shape
+93 0 5 verb 0/1 ood_093_verb_object_0_5_verb
+94 0 5 shape 0/1 ood_094_verb_object_0_5_shape
+95 1 2 verb 0/1 ood_095_verb_object_1_2_verb
+96 1 2 shape 0/1 ood_096_verb_object_1_2_shape
+97 4 5 verb 0/1 ood_097_verb_object_4_5_verb
+98 4 5 shape 0/1 ood_098_verb_object_4_5_shape
+99 2 0 verb 0/1 ood_099_verb_object_2_0_verb
+100 2 0 shape 1/1 ood_100_verb_object_2_0_shape
+101 0 4 verb 0/1 ood_101_verb_object_0_4_verb
+102 0 4 shape 0/1 ood_102_verb_object_0_4_shape
+103 0 3 verb 0/1 ood_103_verb_object_0_3_verb
+104 0 3 shape 0/1 ood_104_verb_object_0_3_shape
+105 2 3 verb 0/1 ood_105_verb_object_2_3_verb
+106 2 3 shape 0/1 ood_106_verb_object_2_3_shape
+107 0 4 verb 0/1 ood_107_verb_object_0_4_verb
+108 0 4 shape 0/1 ood_108_verb_object_0_4_shape
+109 2 1 verb 0/1 ood_109_verb_object_2_1_verb
+110 2 1 shape 0/1 ood_110_verb_object_2_1_shape
+111 5 2 verb 0/1 ood_111_verb_object_5_2_verb
+112 5 2 shape 0/1 ood_112_verb_object_5_2_shape
+113 2 1 verb 0/1 ood_113_verb_object_2_1_verb
+114 2 1 shape 0/1 ood_114_verb_object_2_1_shape
+115 3 5 verb 0/1 ood_115_verb_object_3_5_verb
+116 3 5 shape 0/1 ood_116_verb_object_3_5_shape
+117 1 4 verb 0/1 ood_117_verb_object_1_4_verb
+118 1 4 shape 0/1 ood_118_verb_object_1_4_shape
+119 5 0 verb 0/1 ood_119_verb_object_5_0_verb
+120 5 0 shape 0/1 ood_120_verb_object_5_0_shape
+121 0 2 verb 0/1 ood_121_verb_object_0_2_verb
+122 0 2 shape 0/1 ood_122_verb_object_0_2_shape
+123 4 3 verb 0/1 ood_123_verb_object_4_3_verb
+124 4 3 shape 0/1 ood_124_verb_object_4_3_shape
+125 2 5 verb 0/1 ood_125_verb_object_2_5_verb
+126 2 5 shape 0/1 ood_126_verb_object_2_5_shape
+127 3 2 verb 0/1 ood_127_verb_object_3_2_verb
+128 3 2 shape 0/1 ood_128_verb_object_3_2_shape
+129 0 4 verb 0/1 ood_129_verb_object_0_4_verb
+130 0 4 shape 0/1 ood_130_verb_object_0_4_shape
+131 5 4 verb 0/1 ood_131_verb_object_5_4_verb
+132 5 4 shape 0/1 ood_132_verb_object_5_4_shape
+133 2 3 verb 0/1 ood_133_verb_object_2_3_verb
+134 2 3 shape 0/1 ood_134_verb_object_2_3_shape
+135 0 3 verb 0/1 ood_135_verb_object_0_3_verb
+136 0 3 shape 0/1 ood_136_verb_object_0_3_shape
+137 3 2 verb 0/1 ood_137_verb_object_3_2_verb
+138 3 2 shape 0/1 ood_138_verb_object_3_2_shape
+139 1 5 verb 0/1 ood_139_verb_object_1_5_verb
+140 1 5 shape 0/1 ood_140_verb_object_1_5_shape
+141 5 1 verb 0/1 ood_141_verb_object_5_1_verb
+142 5 1 shape 0/1 ood_142_verb_object_5_1_shape
+143 4 0 verb 0/1 ood_143_verb_object_4_0_verb
+144 4 0 shape 0/1 ood_144_verb_object_4_0_shape
+145 3 5 verb 0/1 ood_145_verb_object_3_5_verb
+146 3 5 shape 0/1 ood_146_verb_object_3_5_shape
+147 5 3 verb 0/1 ood_147_verb_object_5_3_verb
+148 5 3 shape 0/1 ood_148_verb_object_5_3_shape
+149 5 2 verb 0/1 ood_149_verb_object_5_2_verb
+150 5 2 shape 0/1 ood_150_verb_object_5_2_shape
+151 2 1 verb 0/1 ood_151_verb_object_2_1_verb
+152 2 1 shape 0/1 ood_152_verb_object_2_1_shape
+153 3 4 verb 0/1 ood_153_verb_object_3_4_verb
+154 3 4 shape 0/1 ood_154_verb_object_3_4_shape
+155 1 2 verb 1/1 ood_155_verb_object_1_2_verb
+156 1 2 shape 0/1 ood_156_verb_object_1_2_shape
+157 4 2 verb 0/1 ood_157_verb_object_4_2_verb
+158 4 2 shape 0/1 ood_158_verb_object_4_2_shape
+159 0 3 verb 0/1 ood_159_verb_object_0_3_verb
+160 0 3 shape 0/1 ood_160_verb_object_0_3_shape
+161 0 2 verb 0/1 ood_161_verb_object_0_2_verb
+162 0 2 shape 0/1 ood_162_verb_object_0_2_shape
+163 4 1 verb 0/1 ood_163_verb_object_4_1_verb
+164 4 1 shape 0/1 ood_164_verb_object_4_1_shape
+165 1 3 verb 1/1 ood_165_verb_object_1_3_verb
+166 1 3 shape 0/1 ood_166_verb_object_1_3_shape
+167 4 5 verb 0/1 ood_167_verb_object_4_5_verb
+168 4 5 shape 0/1 ood_168_verb_object_4_5_shape
+169 1 5 verb 0/1 ood_169_verb_object_1_5_verb
+170 1 5 shape 0/1 ood_170_verb_object_1_5_shape
+171 0 3 verb 0/1 ood_171_verb_object_0_3_verb
+172 0 3 shape 0/1 ood_172_verb_object_0_3_shape
+173 5 2 verb 0/1 ood_173_verb_object_5_2_verb
+174 5 2 shape 0/1 ood_174_verb_object_5_2_shape
+175 1 3 verb 0/1 ood_175_verb_object_1_3_verb
+176 1 3 shape 0/1 ood_176_verb_object_1_3_shape
+177 5 2 verb 0/1 ood_177_verb_object_5_2_verb
+178 5 2 shape 0/1 ood_178_verb_object_5_2_shape
+179 0 4 verb 0/1 ood_179_verb_object_0_4_verb
+180 0 4 shape 0/1 ood_180_verb_object_0_4_shape
+181 2 3 verb 0/1 ood_181_verb_object_2_3_verb
+182 2 3 shape 0/1 ood_182_verb_object_2_3_shape
+183 1 4 verb 0/1 ood_183_verb_object_1_4_verb
+184 1 4 shape 0/1 ood_184_verb_object_1_4_shape
+185 2 5 verb 0/1 ood_185_verb_object_2_5_verb
+186 2 5 shape 0/1 ood_186_verb_object_2_5_shape
+187 4 0 verb 0/1 ood_187_verb_object_4_0_verb
+188 4 0 shape 0/1 ood_188_verb_object_4_0_shape
+189 5 1 verb 0/1 ood_189_verb_object_5_1_verb
+190 5 1 shape 0/1 ood_190_verb_object_5_1_shape
+191 2 1 verb 0/1 ood_191_verb_object_2_1_verb
+192 2 1 shape 0/1 ood_192_verb_object_2_1_shape
+193 1 0 verb 0/1 ood_193_verb_object_1_0_verb
+194 1 0 shape 0/1 ood_194_verb_object_1_0_shape
+195 2 1 verb 0/1 ood_195_verb_object_2_1_verb
+196 2 1 shape 0/1 ood_196_verb_object_2_1_shape
+197 2 1 verb 0/1 ood_197_verb_object_2_1_verb
+198 2 1 shape 0/1 ood_198_verb_object_2_1_shape
+199 1 2 verb 0/1 ood_199_verb_object_1_2_verb
+200 1 2 shape 0/1 ood_200_verb_object_1_2_shape
+201 4 1 verb 0/1 ood_201_verb_object_4_1_verb
+202 4 1 shape 1/1 ood_202_verb_object_4_1_shape
+203 1 4 verb 0/1 ood_203_verb_object_1_4_verb
+204 1 4 shape 0/1 ood_204_verb_object_1_4_shape
+205 4 2 verb 0/1 ood_205_verb_object_4_2_verb
+206 4 2 shape 0/1 ood_206_verb_object_4_2_shape
+207 5 4 verb 0/1 ood_207_verb_object_5_4_verb
+208 5 4 shape 0/1 ood_208_verb_object_5_4_shape
+209 4 1 verb 0/1 ood_209_verb_object_4_1_verb
+210 4 1 shape 0/1 ood_210_verb_object_4_1_shape
+211 4 0 verb 0/1 ood_211_verb_object_4_0_verb
+212 4 0 shape 0/1 ood_212_verb_object_4_0_shape
+213 0 3 verb 0/1 ood_213_verb_object_0_3_verb
+214 0 3 shape 0/1 ood_214_verb_object_0_3_shape
+215 3 5 verb 0/1 ood_215_verb_object_3_5_verb
+216 3 5 shape 0/1 ood_216_verb_object_3_5_shape
+217 4 0 verb 0/1 ood_217_verb_object_4_0_verb
+218 4 0 shape 0/1 ood_218_verb_object_4_0_shape
+219 1 0 verb 0/1 ood_219_verb_object_1_0_verb
+220 1 0 shape 0/1 ood_220_verb_object_1_0_shape
+221 3 2 verb 0/1 ood_221_verb_object_3_2_verb
+222 3 2 shape 0/1 ood_222_verb_object_3_2_shape
+223 4 3 verb 0/1 ood_223_verb_object_4_3_verb
+224 4 3 shape 0/1 ood_224_verb_object_4_3_shape
+225 1 3 verb 0/1 ood_225_verb_object_1_3_verb
+226 1 3 shape 0/1 ood_226_verb_object_1_3_shape
+227 1 0 verb 1/1 ood_227_verb_object_1_0_verb
+228 1 0 shape 0/1 ood_228_verb_object_1_0_shape
+229 2 5 verb 0/1 ood_229_verb_object_2_5_verb
+230 2 5 shape 0/1 ood_230_verb_object_2_5_shape
+231 2 3 verb 0/1 ood_231_verb_object_2_3_verb
+232 2 3 shape 0/1 ood_232_verb_object_2_3_shape
+233 1 4 verb 0/1 ood_233_verb_object_1_4_verb
+234 1 4 shape 0/1 ood_234_verb_object_1_4_shape
+235 5 4 verb 0/1 ood_235_verb_object_5_4_verb
+236 5 4 shape 0/1 ood_236_verb_object_5_4_shape
+237 4 0 verb 0/1 ood_237_verb_object_4_0_verb
+238 4 0 shape 0/1 ood_238_verb_object_4_0_shape
+239 4 2 verb 0/1 ood_239_verb_object_4_2_verb
+240 4 2 shape 0/1 ood_240_verb_object_4_2_shape
+241 3 2 verb 0/1 ood_241_verb_object_3_2_verb
+242 3 2 shape 0/1 ood_242_verb_object_3_2_shape
+243 1 3 verb 0/1 ood_243_verb_object_1_3_verb
+244 1 3 shape 0/1 ood_244_verb_object_1_3_shape
+245 4 1 verb 0/1 ood_245_verb_object_4_1_verb
+246 4 1 shape 0/1 ood_246_verb_object_4_1_shape
+247 2 0 verb 0/1 ood_247_verb_object_2_0_verb
+248 2 0 shape 0/1 ood_248_verb_object_2_0_shape
+249 5 1 verb 0/1 ood_249_verb_object_5_1_verb
+250 5 1 shape 0/1 ood_250_verb_object_5_1_shape
+251 4 5 verb 0/1 ood_251_verb_object_4_5_verb
+252 4 5 shape 0/1 ood_252_verb_object_4_5_shape
+253 4 5 verb 0/1 ood_253_verb_object_4_5_verb
+254 4 5 shape 0/1 ood_254_verb_object_4_5_shape
+255 0 2 verb 0/1 ood_255_verb_object_0_2_verb
+256 0 2 shape 0/1 ood_256_verb_object_0_2_shape
+257 1 3 verb 0/1 ood_257_verb_object_1_3_verb
+258 1 3 shape 0/1 ood_258_verb_object_1_3_shape
+259 5 1 verb 0/1 ood_259_verb_object_5_1_verb
+260 5 1 shape 0/1 ood_260_verb_object_5_1_shape
+261 0 2 verb 0/1 ood_261_verb_object_0_2_verb
+262 0 2 shape 0/1 ood_262_verb_object_0_2_shape
+263 5 0 verb 0/1 ood_263_verb_object_5_0_verb
+264 5 0 shape 0/1 ood_264_verb_object_5_0_shape
+265 2 0 verb 0/1 ood_265_verb_object_2_0_verb
+266 2 0 shape 0/1 ood_266_verb_object_2_0_shape
+267 2 3 verb 0/1 ood_267_verb_object_2_3_verb
+268 2 3 shape 0/1 ood_268_verb_object_2_3_shape
+269 1 4 verb 0/1 ood_269_verb_object_1_4_verb
+270 1 4 shape 0/1 ood_270_verb_object_1_4_shape
+271 0 3 verb 0/1 ood_271_verb_object_0_3_verb
+272 0 3 shape 0/1 ood_272_verb_object_0_3_shape
+273 1 2 verb 0/1 ood_273_verb_object_1_2_verb
+274 1 2 shape 0/1 ood_274_verb_object_1_2_shape
+275 5 4 verb 0/1 ood_275_verb_object_5_4_verb
+276 5 4 shape 0/1 ood_276_verb_object_5_4_shape
+278 3 4 shape 0/1 ood_278_verb_object_3_4_shape
+279 5 3 verb 0/1 ood_279_verb_object_5_3_verb
+280 5 3 shape 0/1 ood_280_verb_object_5_3_shape
+281 4 2 verb 0/1 ood_281_verb_object_4_2_verb
+282 4 2 shape 0/1 ood_282_verb_object_4_2_shape
+283 2 0 verb 0/1 ood_283_verb_object_2_0_verb
+284 2 0 shape 0/1 ood_284_verb_object_2_0_shape
+285 1 2 verb 0/1 ood_285_verb_object_1_2_verb
+286 1 2 shape 0/1 ood_286_verb_object_1_2_shape
+287 4 0 verb 0/1 ood_287_verb_object_4_0_verb
+288 4 0 shape 0/1 ood_288_verb_object_4_0_shape
+289 3 0 verb 0/1 ood_289_verb_object_3_0_verb
+290 3 0 shape 0/1 ood_290_verb_object_3_0_shape
+291 2 3 verb 0/1 ood_291_verb_object_2_3_verb
+292 2 3 shape 0/1 ood_292_verb_object_2_3_shape
+293 5 3 verb 0/1 ood_293_verb_object_5_3_verb
+294 5 3 shape 0/1 ood_294_verb_object_5_3_shape
+295 5 4 verb 0/1 ood_295_verb_object_5_4_verb
+296 5 4 shape 0/1 ood_296_verb_object_5_4_shape
+297 4 0 verb 0/1 ood_297_verb_object_4_0_verb
+298 4 0 shape 0/1 ood_298_verb_object_4_0_shape
+299 2 5 verb 0/1 ood_299_verb_object_2_5_verb
+300 2 5 shape 0/1 ood_300_verb_object_2_5_shape
+301 0 5 verb 0/1 ood_301_verb_object_0_5_verb
+302 0 5 shape 0/1 ood_302_verb_object_0_5_shape
+303 1 4 verb 0/1 ood_303_verb_object_1_4_verb
+304 1 4 shape 0/1 ood_304_verb_object_1_4_shape
+305 0 5 verb 0/1 ood_305_verb_object_0_5_verb
+306 0 5 shape 0/1 ood_306_verb_object_0_5_shape
+307 1 3 verb 0/1 ood_307_verb_object_1_3_verb
+308 1 3 shape 0/1 ood_308_verb_object_1_3_shape
+309 4 3 verb 0/1 ood_309_verb_object_4_3_verb
+310 4 3 shape 0/1 ood_310_verb_object_4_3_shape
+311 3 2 verb 0/1 ood_311_verb_object_3_2_verb
+312 3 2 shape 0/1 ood_312_verb_object_3_2_shape
+313 3 2 verb 0/1 ood_313_verb_object_3_2_verb
+314 3 2 shape 0/1 ood_314_verb_object_3_2_shape
+315 1 4 verb 0/1 ood_315_verb_object_1_4_verb
+316 1 4 shape 0/1 ood_316_verb_object_1_4_shape
+317 4 3 verb 0/1 ood_317_verb_object_4_3_verb
+318 4 3 shape 0/1 ood_318_verb_object_4_3_shape
+319 3 4 verb 0/1 ood_319_verb_object_3_4_verb
+320 3 4 shape 0/1 ood_320_verb_object_3_4_shape
+321 2 4 verb 0/1 ood_321_verb_object_2_4_verb
+322 2 4 shape 0/1 ood_322_verb_object_2_4_shape
+323 5 3 verb 0/1 ood_323_verb_object_5_3_verb
+324 5 3 shape 0/1 ood_324_verb_object_5_3_shape
+325 3 4 verb 0/1 ood_325_verb_object_3_4_verb
+326 3 4 shape 0/1 ood_326_verb_object_3_4_shape
+327 2 3 verb 0/1 ood_327_verb_object_2_3_verb
+328 2 3 shape 0/1 ood_328_verb_object_2_3_shape
+329 2 1 verb 0/1 ood_329_verb_object_2_1_verb
+330 2 1 shape 0/1 ood_330_verb_object_2_1_shape
+331 1 3 verb 1/1 ood_331_verb_object_1_3_verb
+332 1 3 shape 0/1 ood_332_verb_object_1_3_shape
+333 0 5 verb 0/1 ood_333_verb_object_0_5_verb
+334 0 5 shape 0/1 ood_334_verb_object_0_5_shape
+335 3 1 verb 0/1 ood_335_verb_object_3_1_verb
+336 3 1 shape 0/1 ood_336_verb_object_3_1_shape
+337 3 0 verb 0/1 ood_337_verb_object_3_0_verb
+338 3 0 shape 0/1 ood_338_verb_object_3_0_shape
+339 0 3 verb 0/1 ood_339_verb_object_0_3_verb
+340 0 3 shape 0/1 ood_340_verb_object_0_3_shape
+341 4 5 verb 0/1 ood_341_verb_object_4_5_verb
+342 4 5 shape 0/1 ood_342_verb_object_4_5_shape
+343 0 2 verb 0/1 ood_343_verb_object_0_2_verb
+344 0 2 shape 0/1 ood_344_verb_object_0_2_shape
+345 5 2 verb 0/1 ood_345_verb_object_5_2_verb
+346 5 2 shape 0/1 ood_346_verb_object_5_2_shape
+347 0 4 verb 0/1 ood_347_verb_object_0_4_verb
+348 0 4 shape 0/1 ood_348_verb_object_0_4_shape
+349 0 5 verb 0/1 ood_349_verb_object_0_5_verb
+350 0 5 shape 0/1 ood_350_verb_object_0_5_shape
+351 4 0 verb 0/1 ood_351_verb_object_4_0_verb
+352 4 0 shape 0/1 ood_352_verb_object_4_0_shape
+353 1 0 verb 0/1 ood_353_verb_object_1_0_verb
+354 1 0 shape 0/1 ood_354_verb_object_1_0_shape
+355 5 0 verb 0/1 ood_355_verb_object_5_0_verb
+356 5 0 shape 0/1 ood_356_verb_object_5_0_shape
+357 4 1 verb 1/1 ood_357_verb_object_4_1_verb
+358 4 1 shape 1/1 ood_358_verb_object_4_1_shape
+359 2 4 verb 0/1 ood_359_verb_object_2_4_verb
+360 2 4 shape 0/1 ood_360_verb_object_2_4_shape
+361 3 5 verb 0/1 ood_361_verb_object_3_5_verb
+362 3 5 shape 0/1 ood_362_verb_object_3_5_shape
+363 0 3 verb 0/1 ood_363_verb_object_0_3_verb
+364 0 3 shape 0/1 ood_364_verb_object_0_3_shape
+365 2 3 verb 0/1 ood_365_verb_object_2_3_verb
+366 2 3 shape 0/1 ood_366_verb_object_2_3_shape
+367 2 3 verb 0/1 ood_367_verb_object_2_3_verb
+368 2 3 shape 0/1 ood_368_verb_object_2_3_shape
+369 3 5 verb 0/1 ood_369_verb_object_3_5_verb
+370 3 5 shape 0/1 ood_370_verb_object_3_5_shape
+371 2 5 verb 0/1 ood_371_verb_object_2_5_verb
+372 2 5 shape 0/1 ood_372_verb_object_2_5_shape
+373 3 1 verb 0/1 ood_373_verb_object_3_1_verb
+374 3 1 shape 0/1 ood_374_verb_object_3_1_shape
+375 1 4 verb 0/1 ood_375_verb_object_1_4_verb
+376 1 4 shape 0/1 ood_376_verb_object_1_4_shape
+377 3 2 verb 0/1 ood_377_verb_object_3_2_verb
+378 3 2 shape 0/1 ood_378_verb_object_3_2_shape
+379 5 2 verb 0/1 ood_379_verb_object_5_2_verb
+380 5 2 shape 0/1 ood_380_verb_object_5_2_shape
+381 0 1 verb 0/1 ood_381_verb_object_0_1_verb
+382 0 1 shape 0/1 ood_382_verb_object_0_1_shape
+383 4 1 verb 0/1 ood_383_verb_object_4_1_verb
+384 4 1 shape 0/1 ood_384_verb_object_4_1_shape
+385 4 3 verb 0/1 ood_385_verb_object_4_3_verb
+386 4 3 shape 0/1 ood_386_verb_object_4_3_shape
+387 0 4 verb 0/1 ood_387_verb_object_0_4_verb
+388 0 4 shape 0/1 ood_388_verb_object_0_4_shape
+389 4 1 verb 0/1 ood_389_verb_object_4_1_verb
+390 4 1 shape 0/1 ood_390_verb_object_4_1_shape
+391 5 3 verb 0/1 ood_391_verb_object_5_3_verb
+392 5 3 shape 0/1 ood_392_verb_object_5_3_shape
+393 3 2 verb 0/1 ood_393_verb_object_3_2_verb
+394 3 2 shape 0/1 ood_394_verb_object_3_2_shape
+395 4 5 verb 0/1 ood_395_verb_object_4_5_verb
+396 4 5 shape 0/1 ood_396_verb_object_4_5_shape
+397 1 4 verb 1/1 ood_397_verb_object_1_4_verb
+398 1 4 shape 1/1 ood_398_verb_object_1_4_shape
+399 4 5 verb 0/1 ood_399_verb_object_4_5_verb
+400 4 5 shape 0/1 ood_400_verb_object_4_5_shape
+
+overall_verb_success=3/61 (4.9%)
+overall_shape_success=2/62 (3.2%)

results/genie/conflict_env/genie_verb_size_ood_seed42.txt

@@ -0,0 +1,421 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=verb_size
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/verb_size
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 5 4 verb 0/1 ood_001_verb_size_5_4_verb
+2 5 4 size 0/1 ood_002_verb_size_5_4_size
+3 0 1 verb 0/1 ood_003_verb_size_0_1_verb
+4 0 1 size 0/1 ood_004_verb_size_0_1_size
+5 0 1 verb 0/1 ood_005_verb_size_0_1_verb
+6 0 1 size 0/1 ood_006_verb_size_0_1_size
+7 5 4 verb 0/1 ood_007_verb_size_5_4_verb
+8 5 4 size 0/1 ood_008_verb_size_5_4_size
+9 2 3 verb 0/1 ood_009_verb_size_2_3_verb
+10 2 3 size 0/1 ood_010_verb_size_2_3_size
+11 1 0 verb 0/1 ood_011_verb_size_1_0_verb
+12 1 0 size 0/1 ood_012_verb_size_1_0_size
+13 1 0 verb 0/1 ood_013_verb_size_1_0_verb
+14 1 0 size 0/1 ood_014_verb_size_1_0_size
+15 1 0 verb 0/1 ood_015_verb_size_1_0_verb
+16 1 0 size 0/1 ood_016_verb_size_1_0_size
+17 5 4 verb 0/1 ood_017_verb_size_5_4_verb
+18 5 4 size 0/1 ood_018_verb_size_5_4_size
+19 0 1 verb 0/1 ood_019_verb_size_0_1_verb
+20 0 1 size 0/1 ood_020_verb_size_0_1_size
+21 5 4 verb 0/1 ood_021_verb_size_5_4_verb
+22 5 4 size 1/1 ood_022_verb_size_5_4_size
+23 5 4 verb 0/1 ood_023_verb_size_5_4_verb
+24 5 4 size 0/1 ood_024_verb_size_5_4_size
+25 4 5 verb 1/1 ood_025_verb_size_4_5_verb
+26 4 5 size 1/1 ood_026_verb_size_4_5_size
+27 0 1 verb 0/1 ood_027_verb_size_0_1_verb
+28 0 1 size 1/1 ood_028_verb_size_0_1_size
+29 4 5 verb 1/1 ood_029_verb_size_4_5_verb
+30 4 5 size 0/1 ood_030_verb_size_4_5_size
+31 3 2 verb 0/1 ood_031_verb_size_3_2_verb
+32 3 2 size 0/1 ood_032_verb_size_3_2_size
+33 0 1 verb 0/1 ood_033_verb_size_0_1_verb
+34 0 1 size 1/1 ood_034_verb_size_0_1_size
+35 0 1 verb 0/1 ood_035_verb_size_0_1_verb
+36 0 1 size 1/1 ood_036_verb_size_0_1_size
+37 0 1 verb 0/1 ood_037_verb_size_0_1_verb
+38 0 1 size 1/1 ood_038_verb_size_0_1_size
+39 1 0 verb 0/1 ood_039_verb_size_1_0_verb
+40 1 0 size 0/1 ood_040_verb_size_1_0_size
+41 1 0 verb 0/1 ood_041_verb_size_1_0_verb
+42 1 0 size 0/1 ood_042_verb_size_1_0_size
+43 4 5 verb 0/1 ood_043_verb_size_4_5_verb
+44 4 5 size 1/1 ood_044_verb_size_4_5_size
+45 4 5 verb 1/1 ood_045_verb_size_4_5_verb
+46 4 5 size 0/1 ood_046_verb_size_4_5_size
+47 0 1 verb 0/1 ood_047_verb_size_0_1_verb
+48 0 1 size 0/1 ood_048_verb_size_0_1_size
+49 4 5 verb 0/1 ood_049_verb_size_4_5_verb
+50 4 5 size 0/1 ood_050_verb_size_4_5_size
+51 1 0 verb 0/1 ood_051_verb_size_1_0_verb
+52 1 0 size 0/1 ood_052_verb_size_1_0_size
+53 5 4 verb 0/1 ood_053_verb_size_5_4_verb
+54 5 4 size 1/1 ood_054_verb_size_5_4_size
+55 5 4 verb 0/1 ood_055_verb_size_5_4_verb
+56 5 4 size 0/1 ood_056_verb_size_5_4_size
+57 5 4 verb 0/1 ood_057_verb_size_5_4_verb
+58 5 4 size 0/1 ood_058_verb_size_5_4_size
+59 4 5 verb 1/1 ood_059_verb_size_4_5_verb
+60 4 5 size 0/1 ood_060_verb_size_4_5_size
+61 3 2 verb 0/1 ood_061_verb_size_3_2_verb
+62 3 2 size 0/1 ood_062_verb_size_3_2_size
+63 1 0 verb 1/1 ood_063_verb_size_1_0_verb
+64 1 0 size 0/1 ood_064_verb_size_1_0_size
+65 3 2 verb 0/1 ood_065_verb_size_3_2_verb
+66 3 2 size 0/1 ood_066_verb_size_3_2_size
+67 4 5 verb 0/1 ood_067_verb_size_4_5_verb
+68 4 5 size 0/1 ood_068_verb_size_4_5_size
+69 2 3 verb 0/1 ood_069_verb_size_2_3_verb
+70 2 3 size 0/1 ood_070_verb_size_2_3_size
+71 0 1 verb 0/1 ood_071_verb_size_0_1_verb
+72 0 1 size 0/1 ood_072_verb_size_0_1_size
+73 1 0 verb 1/1 ood_073_verb_size_1_0_verb
+74 1 0 size 0/1 ood_074_verb_size_1_0_size
+75 5 4 verb 0/1 ood_075_verb_size_5_4_verb
+76 5 4 size 1/1 ood_076_verb_size_5_4_size
+77 3 2 verb 0/1 ood_077_verb_size_3_2_verb
+78 3 2 size 0/1 ood_078_verb_size_3_2_size
+79 2 3 verb 0/1 ood_079_verb_size_2_3_verb
+80 2 3 size 0/1 ood_080_verb_size_2_3_size
+81 2 3 verb 0/1 ood_081_verb_size_2_3_verb
+82 2 3 size 0/1 ood_082_verb_size_2_3_size
+83 1 0 verb 1/1 ood_083_verb_size_1_0_verb
+84 1 0 size 0/1 ood_084_verb_size_1_0_size
+85 1 0 verb 1/1 ood_085_verb_size_1_0_verb
+86 1 0 size 0/1 ood_086_verb_size_1_0_size
+87 2 3 verb 0/1 ood_087_verb_size_2_3_verb
+88 2 3 size 0/1 ood_088_verb_size_2_3_size
+89 0 1 verb 0/1 ood_089_verb_size_0_1_verb
+90 0 1 size 1/1 ood_090_verb_size_0_1_size
+91 0 1 verb 0/1 ood_091_verb_size_0_1_verb
+92 0 1 size 1/1 ood_092_verb_size_0_1_size
+93 3 2 verb 0/1 ood_093_verb_size_3_2_verb
+94 3 2 size 0/1 ood_094_verb_size_3_2_size
+95 0 1 verb 0/1 ood_095_verb_size_0_1_verb
+96 0 1 size 0/1 ood_096_verb_size_0_1_size
+97 2 3 verb 0/1 ood_097_verb_size_2_3_verb
+98 2 3 size 0/1 ood_098_verb_size_2_3_size
+99 2 3 verb 0/1 ood_099_verb_size_2_3_verb
+100 2 3 size 1/1 ood_100_verb_size_2_3_size
+101 4 5 verb 0/1 ood_101_verb_size_4_5_verb
+102 4 5 size 1/1 ood_102_verb_size_4_5_size
+103 2 3 verb 0/1 ood_103_verb_size_2_3_verb
+104 2 3 size 0/1 ood_104_verb_size_2_3_size
+105 0 1 verb 0/1 ood_105_verb_size_0_1_verb
+106 0 1 size 1/1 ood_106_verb_size_0_1_size
+107 5 4 verb 0/1 ood_107_verb_size_5_4_verb
+108 5 4 size 1/1 ood_108_verb_size_5_4_size
+109 3 2 verb 0/1 ood_109_verb_size_3_2_verb
+110 3 2 size 1/1 ood_110_verb_size_3_2_size
+111 4 5 verb 1/1 ood_111_verb_size_4_5_verb
+112 4 5 size 0/1 ood_112_verb_size_4_5_size
+113 0 1 verb 0/1 ood_113_verb_size_0_1_verb
+114 0 1 size 1/1 ood_114_verb_size_0_1_size
+115 3 2 verb 0/1 ood_115_verb_size_3_2_verb
+116 3 2 size 0/1 ood_116_verb_size_3_2_size
+117 0 1 verb 0/1 ood_117_verb_size_0_1_verb
+118 0 1 size 0/1 ood_118_verb_size_0_1_size
+119 4 5 verb 1/1 ood_119_verb_size_4_5_verb
+120 4 5 size 1/1 ood_120_verb_size_4_5_size
+121 2 3 verb 0/1 ood_121_verb_size_2_3_verb
+122 2 3 size 0/1 ood_122_verb_size_2_3_size
+123 5 4 verb 0/1 ood_123_verb_size_5_4_verb
+124 5 4 size 0/1 ood_124_verb_size_5_4_size
+125 4 5 verb 0/1 ood_125_verb_size_4_5_verb
+126 4 5 size 0/1 ood_126_verb_size_4_5_size
+127 2 3 verb 0/1 ood_127_verb_size_2_3_verb
+128 2 3 size 0/1 ood_128_verb_size_2_3_size
+129 4 5 verb 0/1 ood_129_verb_size_4_5_verb
+130 4 5 size 0/1 ood_130_verb_size_4_5_size
+131 1 0 verb 1/1 ood_131_verb_size_1_0_verb
+132 1 0 size 0/1 ood_132_verb_size_1_0_size
+133 5 4 verb 1/1 ood_133_verb_size_5_4_verb
+134 5 4 size 0/1 ood_134_verb_size_5_4_size
+135 0 1 verb 0/1 ood_135_verb_size_0_1_verb
+136 0 1 size 1/1 ood_136_verb_size_0_1_size
+137 0 1 verb 0/1 ood_137_verb_size_0_1_verb
+138 0 1 size 1/1 ood_138_verb_size_0_1_size
+139 5 4 verb 0/1 ood_139_verb_size_5_4_verb
+140 5 4 size 1/1 ood_140_verb_size_5_4_size
+141 1 0 verb 1/1 ood_141_verb_size_1_0_verb
+142 1 0 size 0/1 ood_142_verb_size_1_0_size
+143 2 3 verb 0/1 ood_143_verb_size_2_3_verb
+144 2 3 size 0/1 ood_144_verb_size_2_3_size
+145 0 1 verb 0/1 ood_145_verb_size_0_1_verb
+146 0 1 size 1/1 ood_146_verb_size_0_1_size
+147 1 0 verb 1/1 ood_147_verb_size_1_0_verb
+148 1 0 size 0/1 ood_148_verb_size_1_0_size
+149 0 1 verb 0/1 ood_149_verb_size_0_1_verb
+150 0 1 size 1/1 ood_150_verb_size_0_1_size
+151 3 2 verb 0/1 ood_151_verb_size_3_2_verb
+152 3 2 size 0/1 ood_152_verb_size_3_2_size
+153 2 3 verb 0/1 ood_153_verb_size_2_3_verb
+154 2 3 size 0/1 ood_154_verb_size_2_3_size
+155 3 2 verb 0/1 ood_155_verb_size_3_2_verb
+156 3 2 size 0/1 ood_156_verb_size_3_2_size
+157 5 4 verb 1/1 ood_157_verb_size_5_4_verb
+158 5 4 size 1/1 ood_158_verb_size_5_4_size
+159 2 3 verb 0/1 ood_159_verb_size_2_3_verb
+160 2 3 size 0/1 ood_160_verb_size_2_3_size
+161 1 0 verb 1/1 ood_161_verb_size_1_0_verb
+162 1 0 size 0/1 ood_162_verb_size_1_0_size
+163 2 3 verb 0/1 ood_163_verb_size_2_3_verb
+164 2 3 size 0/1 ood_164_verb_size_2_3_size
+165 2 3 verb 0/1 ood_165_verb_size_2_3_verb
+166 2 3 size 0/1 ood_166_verb_size_2_3_size
+167 1 0 verb 0/1 ood_167_verb_size_1_0_verb
+168 1 0 size 0/1 ood_168_verb_size_1_0_size
+169 5 4 verb 0/1 ood_169_verb_size_5_4_verb
+170 5 4 size 1/1 ood_170_verb_size_5_4_size
+171 2 3 verb 0/1 ood_171_verb_size_2_3_verb
+172 2 3 size 0/1 ood_172_verb_size_2_3_size
+173 5 4 verb 0/1 ood_173_verb_size_5_4_verb
+174 5 4 size 0/1 ood_174_verb_size_5_4_size
+175 5 4 verb 0/1 ood_175_verb_size_5_4_verb
+176 5 4 size 0/1 ood_176_verb_size_5_4_size
+177 5 4 verb 1/1 ood_177_verb_size_5_4_verb
+178 5 4 size 0/1 ood_178_verb_size_5_4_size
+179 0 1 verb 0/1 ood_179_verb_size_0_1_verb
+180 0 1 size 1/1 ood_180_verb_size_0_1_size
+181 4 5 verb 1/1 ood_181_verb_size_4_5_verb
+182 4 5 size 0/1 ood_182_verb_size_4_5_size
+183 5 4 verb 0/1 ood_183_verb_size_5_4_verb
+184 5 4 size 0/1 ood_184_verb_size_5_4_size
+185 1 0 verb 0/1 ood_185_verb_size_1_0_verb
+186 1 0 size 0/1 ood_186_verb_size_1_0_size
+187 4 5 verb 1/1 ood_187_verb_size_4_5_verb
+188 4 5 size 0/1 ood_188_verb_size_4_5_size
+189 5 4 verb 0/1 ood_189_verb_size_5_4_verb
+190 5 4 size 0/1 ood_190_verb_size_5_4_size
+191 1 0 verb 0/1 ood_191_verb_size_1_0_verb
+192 1 0 size 0/1 ood_192_verb_size_1_0_size
+193 1 0 verb 0/1 ood_193_verb_size_1_0_verb
+194 1 0 size 0/1 ood_194_verb_size_1_0_size
+195 3 2 verb 0/1 ood_195_verb_size_3_2_verb
+196 3 2 size 0/1 ood_196_verb_size_3_2_size
+197 3 2 verb 0/1 ood_197_verb_size_3_2_verb
+198 3 2 size 0/1 ood_198_verb_size_3_2_size
+199 2 3 verb 0/1 ood_199_verb_size_2_3_verb
+200 2 3 size 0/1 ood_200_verb_size_2_3_size
+201 5 4 verb 0/1 ood_201_verb_size_5_4_verb
+202 5 4 size 1/1 ood_202_verb_size_5_4_size
+203 5 4 verb 1/1 ood_203_verb_size_5_4_verb
+204 5 4 size 0/1 ood_204_verb_size_5_4_size
+205 4 5 verb 1/1 ood_205_verb_size_4_5_verb
+206 4 5 size 1/1 ood_206_verb_size_4_5_size
+207 1 0 verb 1/1 ood_207_verb_size_1_0_verb
+208 1 0 size 0/1 ood_208_verb_size_1_0_size
+209 5 4 verb 0/1 ood_209_verb_size_5_4_verb
+210 5 4 size 0/1 ood_210_verb_size_5_4_size
+211 2 3 verb 0/1 ood_211_verb_size_2_3_verb
+212 2 3 size 0/1 ood_212_verb_size_2_3_size
+213 0 1 verb 0/1 ood_213_verb_size_0_1_verb
+214 0 1 size 1/1 ood_214_verb_size_0_1_size
+215 1 0 verb 0/1 ood_215_verb_size_1_0_verb
+216 1 0 size 0/1 ood_216_verb_size_1_0_size
+217 0 1 verb 0/1 ood_217_verb_size_0_1_verb
+218 0 1 size 0/1 ood_218_verb_size_0_1_size
+219 2 3 verb 0/1 ood_219_verb_size_2_3_verb
+220 2 3 size 0/1 ood_220_verb_size_2_3_size
+221 3 2 verb 0/1 ood_221_verb_size_3_2_verb
+222 3 2 size 0/1 ood_222_verb_size_3_2_size
+223 2 3 verb 0/1 ood_223_verb_size_2_3_verb
+224 2 3 size 0/1 ood_224_verb_size_2_3_size
+225 0 1 verb 0/1 ood_225_verb_size_0_1_verb
+226 0 1 size 1/1 ood_226_verb_size_0_1_size
+227 1 0 verb 1/1 ood_227_verb_size_1_0_verb
+228 1 0 size 0/1 ood_228_verb_size_1_0_size
+229 4 5 verb 0/1 ood_229_verb_size_4_5_verb
+230 4 5 size 0/1 ood_230_verb_size_4_5_size
+231 5 4 verb 0/1 ood_231_verb_size_5_4_verb
+232 5 4 size 0/1 ood_232_verb_size_5_4_size
+233 2 3 verb 0/1 ood_233_verb_size_2_3_verb
+234 2 3 size 0/1 ood_234_verb_size_2_3_size
+235 1 0 verb 0/1 ood_235_verb_size_1_0_verb
+236 1 0 size 0/1 ood_236_verb_size_1_0_size
+237 5 4 verb 0/1 ood_237_verb_size_5_4_verb
+238 5 4 size 0/1 ood_238_verb_size_5_4_size
+239 3 2 verb 0/1 ood_239_verb_size_3_2_verb
+240 3 2 size 0/1 ood_240_verb_size_3_2_size
+241 3 2 verb 0/1 ood_241_verb_size_3_2_verb
+242 3 2 size 1/1 ood_242_verb_size_3_2_size
+243 5 4 verb 1/1 ood_243_verb_size_5_4_verb
+244 5 4 size 0/1 ood_244_verb_size_5_4_size
+245 3 2 verb 0/1 ood_245_verb_size_3_2_verb
+246 3 2 size 0/1 ood_246_verb_size_3_2_size
+247 1 0 verb 1/1 ood_247_verb_size_1_0_verb
+248 1 0 size 0/1 ood_248_verb_size_1_0_size
+249 2 3 verb 0/1 ood_249_verb_size_2_3_verb
+250 2 3 size 0/1 ood_250_verb_size_2_3_size
+251 1 0 verb 0/1 ood_251_verb_size_1_0_verb
+252 1 0 size 0/1 ood_252_verb_size_1_0_size
+253 1 0 verb 1/1 ood_253_verb_size_1_0_verb
+254 1 0 size 0/1 ood_254_verb_size_1_0_size
+255 5 4 verb 0/1 ood_255_verb_size_5_4_verb
+256 5 4 size 1/1 ood_256_verb_size_5_4_size
+257 4 5 verb 0/1 ood_257_verb_size_4_5_verb
+258 4 5 size 0/1 ood_258_verb_size_4_5_size
+259 4 5 verb 0/1 ood_259_verb_size_4_5_verb
+260 4 5 size 0/1 ood_260_verb_size_4_5_size
+261 2 3 verb 1/1 ood_261_verb_size_2_3_verb
+262 2 3 size 0/1 ood_262_verb_size_2_3_size
+263 5 4 verb 0/1 ood_263_verb_size_5_4_verb
+264 5 4 size 1/1 ood_264_verb_size_5_4_size
+265 4 5 verb 1/1 ood_265_verb_size_4_5_verb
+266 4 5 size 0/1 ood_266_verb_size_4_5_size
+267 3 2 verb 0/1 ood_267_verb_size_3_2_verb
+268 3 2 size 0/1 ood_268_verb_size_3_2_size
+269 4 5 verb 1/1 ood_269_verb_size_4_5_verb
+270 4 5 size 0/1 ood_270_verb_size_4_5_size
+271 3 2 verb 0/1 ood_271_verb_size_3_2_verb
+272 3 2 size 1/1 ood_272_verb_size_3_2_size
+273 2 3 verb 0/1 ood_273_verb_size_2_3_verb
+274 2 3 size 0/1 ood_274_verb_size_2_3_size
+275 1 0 verb 1/1 ood_275_verb_size_1_0_verb
+276 1 0 size 0/1 ood_276_verb_size_1_0_size
+277 1 0 verb 0/1 ood_277_verb_size_1_0_verb
+278 1 0 size 0/1 ood_278_verb_size_1_0_size
+279 4 5 verb 0/1 ood_279_verb_size_4_5_verb
+280 4 5 size 0/1 ood_280_verb_size_4_5_size
+281 3 2 verb 0/1 ood_281_verb_size_3_2_verb
+282 3 2 size 0/1 ood_282_verb_size_3_2_size
+283 0 1 verb 0/1 ood_283_verb_size_0_1_verb
+284 0 1 size 1/1 ood_284_verb_size_0_1_size
+285 0 1 verb 0/1 ood_285_verb_size_0_1_verb
+286 0 1 size 0/1 ood_286_verb_size_0_1_size
+287 0 1 verb 0/1 ood_287_verb_size_0_1_verb
+288 0 1 size 0/1 ood_288_verb_size_0_1_size
+289 1 0 verb 0/1 ood_289_verb_size_1_0_verb
+290 1 0 size 0/1 ood_290_verb_size_1_0_size
+291 5 4 verb 0/1 ood_291_verb_size_5_4_verb
+292 5 4 size 0/1 ood_292_verb_size_5_4_size
+293 1 0 verb 1/1 ood_293_verb_size_1_0_verb
+294 1 0 size 0/1 ood_294_verb_size_1_0_size
+295 5 4 verb 1/1 ood_295_verb_size_5_4_verb
+296 5 4 size 1/1 ood_296_verb_size_5_4_size
+297 3 2 verb 0/1 ood_297_verb_size_3_2_verb
+298 3 2 size 0/1 ood_298_verb_size_3_2_size
+299 4 5 verb 1/1 ood_299_verb_size_4_5_verb
+300 4 5 size 0/1 ood_300_verb_size_4_5_size
+301 0 1 verb 0/1 ood_301_verb_size_0_1_verb
+302 0 1 size 1/1 ood_302_verb_size_0_1_size
+303 3 2 verb 0/1 ood_303_verb_size_3_2_verb
+304 3 2 size 0/1 ood_304_verb_size_3_2_size
+305 3 2 verb 0/1 ood_305_verb_size_3_2_verb
+306 3 2 size 0/1 ood_306_verb_size_3_2_size
+307 4 5 verb 0/1 ood_307_verb_size_4_5_verb
+308 4 5 size 0/1 ood_308_verb_size_4_5_size
+309 3 2 verb 0/1 ood_309_verb_size_3_2_verb
+310 3 2 size 0/1 ood_310_verb_size_3_2_size
+311 4 5 verb 0/1 ood_311_verb_size_4_5_verb
+312 4 5 size 0/1 ood_312_verb_size_4_5_size
+313 2 3 verb 0/1 ood_313_verb_size_2_3_verb
+314 2 3 size 0/1 ood_314_verb_size_2_3_size
+315 4 5 verb 0/1 ood_315_verb_size_4_5_verb
+316 4 5 size 0/1 ood_316_verb_size_4_5_size
+317 0 1 verb 0/1 ood_317_verb_size_0_1_verb
+318 0 1 size 1/1 ood_318_verb_size_0_1_size
+319 5 4 verb 0/1 ood_319_verb_size_5_4_verb
+320 5 4 size 1/1 ood_320_verb_size_5_4_size
+321 5 4 verb 0/1 ood_321_verb_size_5_4_verb
+322 5 4 size 0/1 ood_322_verb_size_5_4_size
+323 0 1 verb 0/1 ood_323_verb_size_0_1_verb
+324 0 1 size 0/1 ood_324_verb_size_0_1_size
+325 5 4 verb 1/1 ood_325_verb_size_5_4_verb
+326 5 4 size 0/1 ood_326_verb_size_5_4_size
+327 4 5 verb 0/1 ood_327_verb_size_4_5_verb
+328 4 5 size 0/1 ood_328_verb_size_4_5_size
+329 2 3 verb 0/1 ood_329_verb_size_2_3_verb
+330 2 3 size 0/1 ood_330_verb_size_2_3_size
+331 5 4 verb 1/1 ood_331_verb_size_5_4_verb
+332 5 4 size 0/1 ood_332_verb_size_5_4_size
+333 2 3 verb 0/1 ood_333_verb_size_2_3_verb
+334 2 3 size 0/1 ood_334_verb_size_2_3_size
+335 0 1 verb 0/1 ood_335_verb_size_0_1_verb
+336 0 1 size 1/1 ood_336_verb_size_0_1_size
+337 2 3 verb 0/1 ood_337_verb_size_2_3_verb
+338 2 3 size 0/1 ood_338_verb_size_2_3_size
+339 3 2 verb 0/1 ood_339_verb_size_3_2_verb
+340 3 2 size 0/1 ood_340_verb_size_3_2_size
+341 1 0 verb 0/1 ood_341_verb_size_1_0_verb
+342 1 0 size 0/1 ood_342_verb_size_1_0_size
+343 3 2 verb 0/1 ood_343_verb_size_3_2_verb
+344 3 2 size 1/1 ood_344_verb_size_3_2_size
+345 0 1 verb 0/1 ood_345_verb_size_0_1_verb
+346 0 1 size 1/1 ood_346_verb_size_0_1_size
+347 5 4 verb 0/1 ood_347_verb_size_5_4_verb
+348 5 4 size 0/1 ood_348_verb_size_5_4_size
+349 5 4 verb 1/1 ood_349_verb_size_5_4_verb
+350 5 4 size 0/1 ood_350_verb_size_5_4_size
+351 2 3 verb 0/1 ood_351_verb_size_2_3_verb
+352 2 3 size 0/1 ood_352_verb_size_2_3_size
+353 4 5 verb 0/1 ood_353_verb_size_4_5_verb
+354 4 5 size 0/1 ood_354_verb_size_4_5_size
+355 1 0 verb 0/1 ood_355_verb_size_1_0_verb
+356 1 0 size 0/1 ood_356_verb_size_1_0_size
+357 4 5 verb 0/1 ood_357_verb_size_4_5_verb
+358 4 5 size 0/1 ood_358_verb_size_4_5_size
+359 0 1 verb 0/1 ood_359_verb_size_0_1_verb
+360 0 1 size 0/1 ood_360_verb_size_0_1_size
+361 5 4 verb 0/1 ood_361_verb_size_5_4_verb
+362 5 4 size 0/1 ood_362_verb_size_5_4_size
+363 2 3 verb 0/1 ood_363_verb_size_2_3_verb
+364 2 3 size 0/1 ood_364_verb_size_2_3_size
+365 5 4 verb 0/1 ood_365_verb_size_5_4_verb
+366 5 4 size 1/1 ood_366_verb_size_5_4_size
+367 4 5 verb 0/1 ood_367_verb_size_4_5_verb
+368 4 5 size 0/1 ood_368_verb_size_4_5_size
+369 4 5 verb 0/1 ood_369_verb_size_4_5_verb
+370 4 5 size 0/1 ood_370_verb_size_4_5_size
+371 1 0 verb 0/1 ood_371_verb_size_1_0_verb
+372 1 0 size 0/1 ood_372_verb_size_1_0_size
+373 1 0 verb 1/1 ood_373_verb_size_1_0_verb
+374 1 0 size 0/1 ood_374_verb_size_1_0_size
+375 2 3 verb 0/1 ood_375_verb_size_2_3_verb
+376 2 3 size 0/1 ood_376_verb_size_2_3_size
+377 1 0 verb 1/1 ood_377_verb_size_1_0_verb
+378 1 0 size 0/1 ood_378_verb_size_1_0_size
+379 4 5 verb 0/1 ood_379_verb_size_4_5_verb
+380 4 5 size 0/1 ood_380_verb_size_4_5_size
+381 4 5 verb 1/1 ood_381_verb_size_4_5_verb
+382 4 5 size 0/1 ood_382_verb_size_4_5_size
+383 0 1 verb 0/1 ood_383_verb_size_0_1_verb
+384 0 1 size 0/1 ood_384_verb_size_0_1_size
+385 4 5 verb 0/1 ood_385_verb_size_4_5_verb
+386 4 5 size 0/1 ood_386_verb_size_4_5_size
+387 2 3 verb 0/1 ood_387_verb_size_2_3_verb
+388 2 3 size 0/1 ood_388_verb_size_2_3_size
+389 3 2 verb 0/1 ood_389_verb_size_3_2_verb
+390 3 2 size 0/1 ood_390_verb_size_3_2_size
+391 0 1 verb 0/1 ood_391_verb_size_0_1_verb
+392 0 1 size 1/1 ood_392_verb_size_0_1_size
+393 0 1 verb 0/1 ood_393_verb_size_0_1_verb
+394 0 1 size 1/1 ood_394_verb_size_0_1_size
+395 2 3 verb 0/1 ood_395_verb_size_2_3_verb
+396 2 3 size 0/1 ood_396_verb_size_2_3_size
+397 2 3 verb 0/1 ood_397_verb_size_2_3_verb
+398 2 3 size 0/1 ood_398_verb_size_2_3_size
+399 1 0 verb 0/1 ood_399_verb_size_1_0_verb
+400 1 0 size 0/1 ood_400_verb_size_1_0_size
+
+overall_verb_success=39/200 (19.5%)
+overall_size_success=45/200 (22.5%)

results/genie/conflict_env/genie_verb_spatial_ood_seed42.txt

@@ -0,0 +1,421 @@
+# OOD pairwise inference summary (Genie-Envisioner)
+experiment=verb_spatial
+seed=42
+total_episodes_target=200
+num_episodes_per_run=1
+total_runs=400
+total_episodes_actual=400
+third_seed=42
+weight=/workspace/groot_eval/genie_ckpts/verb_spatial
+ltx_model=/workspace/groot_eval/LTX-Video
+domain_name=conflict
+sim_backend=cpu
+max_episode_steps=300
+replan_steps=5
+num_inference_steps=5
+seed_base=0
+
+index pair_i pair_j run_type success run_name
+1 0 4 verb 0/1 ood_001_verb_spatial_0_4_verb
+2 0 4 spatial 0/1 ood_002_verb_spatial_0_4_spatial
+3 0 1 verb 0/1 ood_003_verb_spatial_0_1_verb
+4 0 1 spatial 1/1 ood_004_verb_spatial_0_1_spatial
+5 2 0 verb 0/1 ood_005_verb_spatial_2_0_verb
+6 2 0 spatial 0/1 ood_006_verb_spatial_2_0_spatial
+7 1 4 verb 0/1 ood_007_verb_spatial_1_4_verb
+8 1 4 spatial 0/1 ood_008_verb_spatial_1_4_spatial
+9 1 4 verb 1/1 ood_009_verb_spatial_1_4_verb
+10 1 4 spatial 0/1 ood_010_verb_spatial_1_4_spatial
+11 1 0 verb 0/1 ood_011_verb_spatial_1_0_verb
+12 1 0 spatial 0/1 ood_012_verb_spatial_1_0_spatial
+13 0 4 verb 0/1 ood_013_verb_spatial_0_4_verb
+14 0 4 spatial 0/1 ood_014_verb_spatial_0_4_spatial
+15 4 1 verb 0/1 ood_015_verb_spatial_4_1_verb
+16 4 1 spatial 1/1 ood_016_verb_spatial_4_1_spatial
+17 0 3 verb 0/1 ood_017_verb_spatial_0_3_verb
+18 0 3 spatial 0/1 ood_018_verb_spatial_0_3_spatial
+19 4 2 verb 0/1 ood_019_verb_spatial_4_2_verb
+20 4 2 spatial 0/1 ood_020_verb_spatial_4_2_spatial
+21 3 1 verb 0/1 ood_021_verb_spatial_3_1_verb
+22 3 1 spatial 0/1 ood_022_verb_spatial_3_1_spatial
+23 0 2 verb 0/1 ood_023_verb_spatial_0_2_verb
+24 0 2 spatial 0/1 ood_024_verb_spatial_0_2_spatial
+25 0 1 verb 0/1 ood_025_verb_spatial_0_1_verb
+26 0 1 spatial 1/1 ood_026_verb_spatial_0_1_spatial
+27 0 3 verb 0/1 ood_027_verb_spatial_0_3_verb
+28 0 3 spatial 0/1 ood_028_verb_spatial_0_3_spatial
+29 1 3 verb 0/1 ood_029_verb_spatial_1_3_verb
+30 1 3 spatial 0/1 ood_030_verb_spatial_1_3_spatial
+31 1 4 verb 1/1 ood_031_verb_spatial_1_4_verb
+32 1 4 spatial 0/1 ood_032_verb_spatial_1_4_spatial
+33 4 0 verb 0/1 ood_033_verb_spatial_4_0_verb
+34 4 0 spatial 0/1 ood_034_verb_spatial_4_0_spatial
+35 4 3 verb 0/1 ood_035_verb_spatial_4_3_verb
+36 4 3 spatial 0/1 ood_036_verb_spatial_4_3_spatial
+37 0 1 verb 0/1 ood_037_verb_spatial_0_1_verb
+38 0 1 spatial 1/1 ood_038_verb_spatial_0_1_spatial
+39 4 1 verb 0/1 ood_039_verb_spatial_4_1_verb
+40 4 1 spatial 1/1 ood_040_verb_spatial_4_1_spatial
+41 1 3 verb 0/1 ood_041_verb_spatial_1_3_verb
+42 1 3 spatial 0/1 ood_042_verb_spatial_1_3_spatial
+43 4 1 verb 0/1 ood_043_verb_spatial_4_1_verb
+44 4 1 spatial 1/1 ood_044_verb_spatial_4_1_spatial
+45 3 1 verb 0/1 ood_045_verb_spatial_3_1_verb
+46 3 1 spatial 0/1 ood_046_verb_spatial_3_1_spatial
+47 1 4 verb 1/1 ood_047_verb_spatial_1_4_verb
+48 1 4 spatial 0/1 ood_048_verb_spatial_1_4_spatial
+49 3 2 verb 0/1 ood_049_verb_spatial_3_2_verb
+50 3 2 spatial 0/1 ood_050_verb_spatial_3_2_spatial
+51 4 2 verb 0/1 ood_051_verb_spatial_4_2_verb
+52 4 2 spatial 0/1 ood_052_verb_spatial_4_2_spatial
+53 2 0 verb 0/1 ood_053_verb_spatial_2_0_verb
+54 2 0 spatial 0/1 ood_054_verb_spatial_2_0_spatial
+55 0 1 verb 0/1 ood_055_verb_spatial_0_1_verb
+56 0 1 spatial 1/1 ood_056_verb_spatial_0_1_spatial
+57 1 2 verb 0/1 ood_057_verb_spatial_1_2_verb
+58 1 2 spatial 0/1 ood_058_verb_spatial_1_2_spatial
+59 3 1 verb 0/1 ood_059_verb_spatial_3_1_verb
+60 3 1 spatial 0/1 ood_060_verb_spatial_3_1_spatial
+61 2 3 verb 0/1 ood_061_verb_spatial_2_3_verb
+62 2 3 spatial 0/1 ood_062_verb_spatial_2_3_spatial
+63 2 0 verb 0/1 ood_063_verb_spatial_2_0_verb
+64 2 0 spatial 1/1 ood_064_verb_spatial_2_0_spatial
+65 1 0 verb 0/1 ood_065_verb_spatial_1_0_verb
+66 1 0 spatial 0/1 ood_066_verb_spatial_1_0_spatial
+67 1 3 verb 0/1 ood_067_verb_spatial_1_3_verb
+68 1 3 spatial 0/1 ood_068_verb_spatial_1_3_spatial
+69 2 3 verb 0/1 ood_069_verb_spatial_2_3_verb
+70 2 3 spatial 0/1 ood_070_verb_spatial_2_3_spatial
+71 0 4 verb 0/1 ood_071_verb_spatial_0_4_verb
+72 0 4 spatial 0/1 ood_072_verb_spatial_0_4_spatial
+73 0 3 verb 0/1 ood_073_verb_spatial_0_3_verb
+74 0 3 spatial 0/1 ood_074_verb_spatial_0_3_spatial
+75 3 0 verb 0/1 ood_075_verb_spatial_3_0_verb
+76 3 0 spatial 0/1 ood_076_verb_spatial_3_0_spatial
+77 0 4 verb 0/1 ood_077_verb_spatial_0_4_verb
+78 0 4 spatial 0/1 ood_078_verb_spatial_0_4_spatial
+79 2 4 verb 0/1 ood_079_verb_spatial_2_4_verb
+80 2 4 spatial 0/1 ood_080_verb_spatial_2_4_spatial
+81 2 4 verb 0/1 ood_081_verb_spatial_2_4_verb
+82 2 4 spatial 0/1 ood_082_verb_spatial_2_4_spatial
+83 4 3 verb 0/1 ood_083_verb_spatial_4_3_verb
+84 4 3 spatial 0/1 ood_084_verb_spatial_4_3_spatial
+85 2 0 verb 0/1 ood_085_verb_spatial_2_0_verb
+86 2 0 spatial 0/1 ood_086_verb_spatial_2_0_spatial
+87 0 2 verb 0/1 ood_087_verb_spatial_0_2_verb
+88 0 2 spatial 1/1 ood_088_verb_spatial_0_2_spatial
+89 3 2 verb 0/1 ood_089_verb_spatial_3_2_verb
+90 3 2 spatial 0/1 ood_090_verb_spatial_3_2_spatial
+91 4 1 verb 0/1 ood_091_verb_spatial_4_1_verb
+92 4 1 spatial 0/1 ood_092_verb_spatial_4_1_spatial
+93 0 4 verb 0/1 ood_093_verb_spatial_0_4_verb
+94 0 4 spatial 0/1 ood_094_verb_spatial_0_4_spatial
+95 3 0 verb 0/1 ood_095_verb_spatial_3_0_verb
+96 3 0 spatial 0/1 ood_096_verb_spatial_3_0_spatial
+97 0 3 verb 0/1 ood_097_verb_spatial_0_3_verb
+98 0 3 spatial 0/1 ood_098_verb_spatial_0_3_spatial
+99 4 1 verb 0/1 ood_099_verb_spatial_4_1_verb
+100 4 1 spatial 1/1 ood_100_verb_spatial_4_1_spatial
+101 2 1 verb 0/1 ood_101_verb_spatial_2_1_verb
+102 2 1 spatial 0/1 ood_102_verb_spatial_2_1_spatial
+103 4 3 verb 0/1 ood_103_verb_spatial_4_3_verb
+104 4 3 spatial 0/1 ood_104_verb_spatial_4_3_spatial
+105 2 4 verb 0/1 ood_105_verb_spatial_2_4_verb
+106 2 4 spatial 0/1 ood_106_verb_spatial_2_4_spatial
+107 4 2 verb 1/1 ood_107_verb_spatial_4_2_verb
+108 4 2 spatial 0/1 ood_108_verb_spatial_4_2_spatial
+109 1 3 verb 0/1 ood_109_verb_spatial_1_3_verb
+110 1 3 spatial 0/1 ood_110_verb_spatial_1_3_spatial
+111 0 3 verb 0/1 ood_111_verb_spatial_0_3_verb
+112 0 3 spatial 0/1 ood_112_verb_spatial_0_3_spatial
+113 0 2 verb 0/1 ood_113_verb_spatial_0_2_verb
+114 0 2 spatial 0/1 ood_114_verb_spatial_0_2_spatial
+115 1 4 verb 1/1 ood_115_verb_spatial_1_4_verb
+116 1 4 spatial 0/1 ood_116_verb_spatial_1_4_spatial
+117 2 1 verb 0/1 ood_117_verb_spatial_2_1_verb
+118 2 1 spatial 1/1 ood_118_verb_spatial_2_1_spatial
+119 0 3 verb 0/1 ood_119_verb_spatial_0_3_verb
+120 0 3 spatial 0/1 ood_120_verb_spatial_0_3_spatial
+121 1 4 verb 1/1 ood_121_verb_spatial_1_4_verb
+122 1 4 spatial 0/1 ood_122_verb_spatial_1_4_spatial
+123 0 4 verb 0/1 ood_123_verb_spatial_0_4_verb
+124 0 4 spatial 0/1 ood_124_verb_spatial_0_4_spatial
+125 3 0 verb 0/1 ood_125_verb_spatial_3_0_verb
+126 3 0 spatial 0/1 ood_126_verb_spatial_3_0_spatial
+127 2 0 verb 0/1 ood_127_verb_spatial_2_0_verb
+128 2 0 spatial 0/1 ood_128_verb_spatial_2_0_spatial
+129 3 2 verb 0/1 ood_129_verb_spatial_3_2_verb
+130 3 2 spatial 0/1 ood_130_verb_spatial_3_2_spatial
+131 2 4 verb 0/1 ood_131_verb_spatial_2_4_verb
+132 2 4 spatial 0/1 ood_132_verb_spatial_2_4_spatial
+133 1 2 verb 0/1 ood_133_verb_spatial_1_2_verb
+134 1 2 spatial 0/1 ood_134_verb_spatial_1_2_spatial
+135 2 4 verb 0/1 ood_135_verb_spatial_2_4_verb
+136 2 4 spatial 0/1 ood_136_verb_spatial_2_4_spatial
+137 2 4 verb 0/1 ood_137_verb_spatial_2_4_verb
+138 2 4 spatial 0/1 ood_138_verb_spatial_2_4_spatial
+139 1 3 verb 0/1 ood_139_verb_spatial_1_3_verb
+140 1 3 spatial 0/1 ood_140_verb_spatial_1_3_spatial
+141 2 0 verb 0/1 ood_141_verb_spatial_2_0_verb
+142 2 0 spatial 0/1 ood_142_verb_spatial_2_0_spatial
+143 0 3 verb 0/1 ood_143_verb_spatial_0_3_verb
+144 0 3 spatial 0/1 ood_144_verb_spatial_0_3_spatial
+145 4 3 verb 0/1 ood_145_verb_spatial_4_3_verb
+146 4 3 spatial 0/1 ood_146_verb_spatial_4_3_spatial
+147 1 2 verb 0/1 ood_147_verb_spatial_1_2_verb
+148 1 2 spatial 0/1 ood_148_verb_spatial_1_2_spatial
+149 4 1 verb 0/1 ood_149_verb_spatial_4_1_verb
+150 4 1 spatial 1/1 ood_150_verb_spatial_4_1_spatial
+151 1 4 verb 1/1 ood_151_verb_spatial_1_4_verb
+152 1 4 spatial 0/1 ood_152_verb_spatial_1_4_spatial
+153 1 2 verb 0/1 ood_153_verb_spatial_1_2_verb
+154 1 2 spatial 0/1 ood_154_verb_spatial_1_2_spatial
+155 3 2 verb 0/1 ood_155_verb_spatial_3_2_verb
+156 3 2 spatial 0/1 ood_156_verb_spatial_3_2_spatial
+157 3 0 verb 0/1 ood_157_verb_spatial_3_0_verb
+158 3 0 spatial 0/1 ood_158_verb_spatial_3_0_spatial
+159 2 0 verb 0/1 ood_159_verb_spatial_2_0_verb
+160 2 0 spatial 0/1 ood_160_verb_spatial_2_0_spatial
+161 4 1 verb 0/1 ood_161_verb_spatial_4_1_verb
+162 4 1 spatial 1/1 ood_162_verb_spatial_4_1_spatial
+163 1 4 verb 0/1 ood_163_verb_spatial_1_4_verb
+164 1 4 spatial 0/1 ood_164_verb_spatial_1_4_spatial
+165 2 3 verb 0/1 ood_165_verb_spatial_2_3_verb
+166 2 3 spatial 0/1 ood_166_verb_spatial_2_3_spatial
+167 0 2 verb 0/1 ood_167_verb_spatial_0_2_verb
+168 0 2 spatial 1/1 ood_168_verb_spatial_0_2_spatial
+169 1 4 verb 0/1 ood_169_verb_spatial_1_4_verb
+170 1 4 spatial 0/1 ood_170_verb_spatial_1_4_spatial
+171 0 2 verb 0/1 ood_171_verb_spatial_0_2_verb
+172 0 2 spatial 0/1 ood_172_verb_spatial_0_2_spatial
+173 2 3 verb 0/1 ood_173_verb_spatial_2_3_verb
+174 2 3 spatial 0/1 ood_174_verb_spatial_2_3_spatial
+175 3 0 verb 0/1 ood_175_verb_spatial_3_0_verb
+176 3 0 spatial 0/1 ood_176_verb_spatial_3_0_spatial
+177 2 0 verb 0/1 ood_177_verb_spatial_2_0_verb
+178 2 0 spatial 0/1 ood_178_verb_spatial_2_0_spatial
+179 0 3 verb 0/1 ood_179_verb_spatial_0_3_verb
+180 0 3 spatial 0/1 ood_180_verb_spatial_0_3_spatial
+181 1 3 verb 0/1 ood_181_verb_spatial_1_3_verb
+182 1 3 spatial 0/1 ood_182_verb_spatial_1_3_spatial
+183 4 2 verb 0/1 ood_183_verb_spatial_4_2_verb
+184 4 2 spatial 0/1 ood_184_verb_spatial_4_2_spatial
+185 2 3 verb 0/1 ood_185_verb_spatial_2_3_verb
+186 2 3 spatial 0/1 ood_186_verb_spatial_2_3_spatial
+187 1 3 verb 0/1 ood_187_verb_spatial_1_3_verb
+188 1 3 spatial 0/1 ood_188_verb_spatial_1_3_spatial
+189 3 4 verb 0/1 ood_189_verb_spatial_3_4_verb
+190 3 4 spatial 0/1 ood_190_verb_spatial_3_4_spatial
+191 3 0 verb 0/1 ood_191_verb_spatial_3_0_verb
+192 3 0 spatial 0/1 ood_192_verb_spatial_3_0_spatial
+193 3 2 verb 0/1 ood_193_verb_spatial_3_2_verb
+194 3 2 spatial 0/1 ood_194_verb_spatial_3_2_spatial
+195 1 0 verb 0/1 ood_195_verb_spatial_1_0_verb
+196 1 0 spatial 0/1 ood_196_verb_spatial_1_0_spatial
+197 2 0 verb 0/1 ood_197_verb_spatial_2_0_verb
+198 2 0 spatial 0/1 ood_198_verb_spatial_2_0_spatial
+199 1 0 verb 0/1 ood_199_verb_spatial_1_0_verb
+200 1 0 spatial 0/1 ood_200_verb_spatial_1_0_spatial
+201 1 4 verb 0/1 ood_201_verb_spatial_1_4_verb
+202 1 4 spatial 1/1 ood_202_verb_spatial_1_4_spatial
+203 4 1 verb 0/1 ood_203_verb_spatial_4_1_verb
+204 4 1 spatial 1/1 ood_204_verb_spatial_4_1_spatial
+205 4 1 verb 0/1 ood_205_verb_spatial_4_1_verb
+206 4 1 spatial 1/1 ood_206_verb_spatial_4_1_spatial
+207 2 0 verb 0/1 ood_207_verb_spatial_2_0_verb
+208 2 0 spatial 0/1 ood_208_verb_spatial_2_0_spatial
+209 4 2 verb 0/1 ood_209_verb_spatial_4_2_verb
+210 4 2 spatial 0/1 ood_210_verb_spatial_4_2_spatial
+211 3 1 verb 0/1 ood_211_verb_spatial_3_1_verb
+212 3 1 spatial 0/1 ood_212_verb_spatial_3_1_spatial
+213 4 2 verb 0/1 ood_213_verb_spatial_4_2_verb
+214 4 2 spatial 0/1 ood_214_verb_spatial_4_2_spatial
+215 3 0 verb 0/1 ood_215_verb_spatial_3_0_verb
+216 3 0 spatial 0/1 ood_216_verb_spatial_3_0_spatial
+217 2 4 verb 0/1 ood_217_verb_spatial_2_4_verb
+218 2 4 spatial 0/1 ood_218_verb_spatial_2_4_spatial
+219 1 4 verb 0/1 ood_219_verb_spatial_1_4_verb
+220 1 4 spatial 0/1 ood_220_verb_spatial_1_4_spatial
+221 1 0 verb 0/1 ood_221_verb_spatial_1_0_verb
+222 1 0 spatial 0/1 ood_222_verb_spatial_1_0_spatial
+223 4 0 verb 0/1 ood_223_verb_spatial_4_0_verb
+224 4 0 spatial 0/1 ood_224_verb_spatial_4_0_spatial
+225 3 4 verb 0/1 ood_225_verb_spatial_3_4_verb
+226 3 4 spatial 0/1 ood_226_verb_spatial_3_4_spatial
+227 0 3 verb 0/1 ood_227_verb_spatial_0_3_verb
+228 0 3 spatial 0/1 ood_228_verb_spatial_0_3_spatial
+229 0 2 verb 0/1 ood_229_verb_spatial_0_2_verb
+230 0 2 spatial 0/1 ood_230_verb_spatial_0_2_spatial
+231 0 4 verb 0/1 ood_231_verb_spatial_0_4_verb
+232 0 4 spatial 0/1 ood_232_verb_spatial_0_4_spatial
+233 1 0 verb 0/1 ood_233_verb_spatial_1_0_verb
+234 1 0 spatial 0/1 ood_234_verb_spatial_1_0_spatial
+235 1 2 verb 0/1 ood_235_verb_spatial_1_2_verb
+236 1 2 spatial 0/1 ood_236_verb_spatial_1_2_spatial
+237 3 1 verb 0/1 ood_237_verb_spatial_3_1_verb
+238 3 1 spatial 0/1 ood_238_verb_spatial_3_1_spatial
+239 4 3 verb 0/1 ood_239_verb_spatial_4_3_verb
+240 4 3 spatial 0/1 ood_240_verb_spatial_4_3_spatial
+241 0 3 verb 0/1 ood_241_verb_spatial_0_3_verb
+242 0 3 spatial 0/1 ood_242_verb_spatial_0_3_spatial
+243 3 0 verb 0/1 ood_243_verb_spatial_3_0_verb
+244 3 0 spatial 0/1 ood_244_verb_spatial_3_0_spatial
+245 3 0 verb 0/1 ood_245_verb_spatial_3_0_verb
+246 3 0 spatial 0/1 ood_246_verb_spatial_3_0_spatial
+247 4 3 verb 0/1 ood_247_verb_spatial_4_3_verb
+248 4 3 spatial 0/1 ood_248_verb_spatial_4_3_spatial
+249 3 2 verb 0/1 ood_249_verb_spatial_3_2_verb
+250 3 2 spatial 0/1 ood_250_verb_spatial_3_2_spatial
+251 4 0 verb 0/1 ood_251_verb_spatial_4_0_verb
+252 4 0 spatial 0/1 ood_252_verb_spatial_4_0_spatial
+253 2 0 verb 0/1 ood_253_verb_spatial_2_0_verb
+254 2 0 spatial 0/1 ood_254_verb_spatial_2_0_spatial
+255 4 1 verb 0/1 ood_255_verb_spatial_4_1_verb
+256 4 1 spatial 1/1 ood_256_verb_spatial_4_1_spatial
+257 0 1 verb 0/1 ood_257_verb_spatial_0_1_verb
+258 0 1 spatial 1/1 ood_258_verb_spatial_0_1_spatial
+259 0 4 verb 0/1 ood_259_verb_spatial_0_4_verb
+260 0 4 spatial 0/1 ood_260_verb_spatial_0_4_spatial
+261 4 1 verb 0/1 ood_261_verb_spatial_4_1_verb
+262 4 1 spatial 1/1 ood_262_verb_spatial_4_1_spatial
+263 2 0 verb 0/1 ood_263_verb_spatial_2_0_verb
+264 2 0 spatial 0/1 ood_264_verb_spatial_2_0_spatial
+265 2 3 verb 0/1 ood_265_verb_spatial_2_3_verb
+266 2 3 spatial 0/1 ood_266_verb_spatial_2_3_spatial
+267 0 4 verb 0/1 ood_267_verb_spatial_0_4_verb
+268 0 4 spatial 0/1 ood_268_verb_spatial_0_4_spatial
+269 2 1 verb 0/1 ood_269_verb_spatial_2_1_verb
+270 2 1 spatial 0/1 ood_270_verb_spatial_2_1_spatial
+271 3 1 verb 0/1 ood_271_verb_spatial_3_1_verb
+272 3 1 spatial 0/1 ood_272_verb_spatial_3_1_spatial
+273 1 2 verb 0/1 ood_273_verb_spatial_1_2_verb
+274 1 2 spatial 0/1 ood_274_verb_spatial_1_2_spatial
+275 3 2 verb 0/1 ood_275_verb_spatial_3_2_verb
+276 3 2 spatial 0/1 ood_276_verb_spatial_3_2_spatial
+277 0 1 verb 0/1 ood_277_verb_spatial_0_1_verb
+278 0 1 spatial 1/1 ood_278_verb_spatial_0_1_spatial
+279 2 0 verb 1/1 ood_279_verb_spatial_2_0_verb
+280 2 0 spatial 0/1 ood_280_verb_spatial_2_0_spatial
+281 4 0 verb 0/1 ood_281_verb_spatial_4_0_verb
+282 4 0 spatial 0/1 ood_282_verb_spatial_4_0_spatial
+283 1 2 verb 0/1 ood_283_verb_spatial_1_2_verb
+284 1 2 spatial 0/1 ood_284_verb_spatial_1_2_spatial
+285 4 0 verb 0/1 ood_285_verb_spatial_4_0_verb
+286 4 0 spatial 0/1 ood_286_verb_spatial_4_0_spatial
+287 0 4 verb 0/1 ood_287_verb_spatial_0_4_verb
+288 0 4 spatial 0/1 ood_288_verb_spatial_0_4_spatial
+289 2 1 verb 0/1 ood_289_verb_spatial_2_1_verb
+290 2 1 spatial 0/1 ood_290_verb_spatial_2_1_spatial
+291 4 0 verb 0/1 ood_291_verb_spatial_4_0_verb
+292 4 0 spatial 0/1 ood_292_verb_spatial_4_0_spatial
+293 4 3 verb 0/1 ood_293_verb_spatial_4_3_verb
+294 4 3 spatial 0/1 ood_294_verb_spatial_4_3_spatial
+295 1 3 verb 0/1 ood_295_verb_spatial_1_3_verb
+296 1 3 spatial 0/1 ood_296_verb_spatial_1_3_spatial
+297 1 0 verb 1/1 ood_297_verb_spatial_1_0_verb
+298 1 0 spatial 0/1 ood_298_verb_spatial_1_0_spatial
+299 2 4 verb 0/1 ood_299_verb_spatial_2_4_verb
+300 2 4 spatial 0/1 ood_300_verb_spatial_2_4_spatial
+301 1 2 verb 0/1 ood_301_verb_spatial_1_2_verb
+302 1 2 spatial 0/1 ood_302_verb_spatial_1_2_spatial
+303 4 1 verb 0/1 ood_303_verb_spatial_4_1_verb
+304 4 1 spatial 0/1 ood_304_verb_spatial_4_1_spatial
+305 4 0 verb 0/1 ood_305_verb_spatial_4_0_verb
+306 4 0 spatial 0/1 ood_306_verb_spatial_4_0_spatial
+307 0 1 verb 0/1 ood_307_verb_spatial_0_1_verb
+308 0 1 spatial 1/1 ood_308_verb_spatial_0_1_spatial
+309 4 3 verb 0/1 ood_309_verb_spatial_4_3_verb
+310 4 3 spatial 0/1 ood_310_verb_spatial_4_3_spatial
+311 2 3 verb 0/1 ood_311_verb_spatial_2_3_verb
+312 2 3 spatial 0/1 ood_312_verb_spatial_2_3_spatial
+313 3 4 verb 0/1 ood_313_verb_spatial_3_4_verb
+314 3 4 spatial 0/1 ood_314_verb_spatial_3_4_spatial
+315 0 1 verb 0/1 ood_315_verb_spatial_0_1_verb
+316 0 1 spatial 1/1 ood_316_verb_spatial_0_1_spatial
+317 0 4 verb 0/1 ood_317_verb_spatial_0_4_verb
+318 0 4 spatial 0/1 ood_318_verb_spatial_0_4_spatial
+319 2 4 verb 0/1 ood_319_verb_spatial_2_4_verb
+320 2 4 spatial 0/1 ood_320_verb_spatial_2_4_spatial
+321 2 1 verb 0/1 ood_321_verb_spatial_2_1_verb
+322 2 1 spatial 0/1 ood_322_verb_spatial_2_1_spatial
+323 1 4 verb 1/1 ood_323_verb_spatial_1_4_verb
+324 1 4 spatial 0/1 ood_324_verb_spatial_1_4_spatial
+325 0 2 verb 0/1 ood_325_verb_spatial_0_2_verb
+326 0 2 spatial 1/1 ood_326_verb_spatial_0_2_spatial
+327 1 4 verb 0/1 ood_327_verb_spatial_1_4_verb
+328 1 4 spatial 0/1 ood_328_verb_spatial_1_4_spatial
+329 4 2 verb 0/1 ood_329_verb_spatial_4_2_verb
+330 4 2 spatial 0/1 ood_330_verb_spatial_4_2_spatial
+331 0 3 verb 0/1 ood_331_verb_spatial_0_3_verb
+332 0 3 spatial 0/1 ood_332_verb_spatial_0_3_spatial
+333 0 3 verb 0/1 ood_333_verb_spatial_0_3_verb
+334 0 3 spatial 0/1 ood_334_verb_spatial_0_3_spatial
+335 3 4 verb 0/1 ood_335_verb_spatial_3_4_verb
+336 3 4 spatial 0/1 ood_336_verb_spatial_3_4_spatial
+337 0 3 verb 0/1 ood_337_verb_spatial_0_3_verb
+338 0 3 spatial 0/1 ood_338_verb_spatial_0_3_spatial
+339 4 1 verb 0/1 ood_339_verb_spatial_4_1_verb
+340 4 1 spatial 0/1 ood_340_verb_spatial_4_1_spatial
+341 1 0 verb 0/1 ood_341_verb_spatial_1_0_verb
+342 1 0 spatial 0/1 ood_342_verb_spatial_1_0_spatial
+343 1 0 verb 0/1 ood_343_verb_spatial_1_0_verb
+344 1 0 spatial 0/1 ood_344_verb_spatial_1_0_spatial
+345 3 4 verb 0/1 ood_345_verb_spatial_3_4_verb
+346 3 4 spatial 0/1 ood_346_verb_spatial_3_4_spatial
+347 4 1 verb 0/1 ood_347_verb_spatial_4_1_verb
+348 4 1 spatial 1/1 ood_348_verb_spatial_4_1_spatial
+349 1 2 verb 0/1 ood_349_verb_spatial_1_2_verb
+350 1 2 spatial 0/1 ood_350_verb_spatial_1_2_spatial
+351 2 0 verb 0/1 ood_351_verb_spatial_2_0_verb
+352 2 0 spatial 0/1 ood_352_verb_spatial_2_0_spatial
+353 4 0 verb 0/1 ood_353_verb_spatial_4_0_verb
+354 4 0 spatial 0/1 ood_354_verb_spatial_4_0_spatial
+355 4 3 verb 0/1 ood_355_verb_spatial_4_3_verb
+356 4 3 spatial 0/1 ood_356_verb_spatial_4_3_spatial
+357 3 1 verb 0/1 ood_357_verb_spatial_3_1_verb
+358 3 1 spatial 1/1 ood_358_verb_spatial_3_1_spatial
+359 1 3 verb 0/1 ood_359_verb_spatial_1_3_verb
+360 1 3 spatial 0/1 ood_360_verb_spatial_1_3_spatial
+361 4 1 verb 0/1 ood_361_verb_spatial_4_1_verb
+362 4 1 spatial 1/1 ood_362_verb_spatial_4_1_spatial
+363 1 3 verb 0/1 ood_363_verb_spatial_1_3_verb
+364 1 3 spatial 0/1 ood_364_verb_spatial_1_3_spatial
+365 2 1 verb 0/1 ood_365_verb_spatial_2_1_verb
+366 2 1 spatial 0/1 ood_366_verb_spatial_2_1_spatial
+367 3 0 verb 0/1 ood_367_verb_spatial_3_0_verb
+368 3 0 spatial 0/1 ood_368_verb_spatial_3_0_spatial
+369 2 4 verb 0/1 ood_369_verb_spatial_2_4_verb
+370 2 4 spatial 0/1 ood_370_verb_spatial_2_4_spatial
+371 3 2 verb 0/1 ood_371_verb_spatial_3_2_verb
+372 3 2 spatial 0/1 ood_372_verb_spatial_3_2_spatial
+373 4 0 verb 0/1 ood_373_verb_spatial_4_0_verb
+374 4 0 spatial 0/1 ood_374_verb_spatial_4_0_spatial
+375 3 2 verb 0/1 ood_375_verb_spatial_3_2_verb
+376 3 2 spatial 0/1 ood_376_verb_spatial_3_2_spatial
+377 0 4 verb 0/1 ood_377_verb_spatial_0_4_verb
+378 0 4 spatial 0/1 ood_378_verb_spatial_0_4_spatial
+379 1 4 verb 1/1 ood_379_verb_spatial_1_4_verb
+380 1 4 spatial 0/1 ood_380_verb_spatial_1_4_spatial
+381 1 4 verb 1/1 ood_381_verb_spatial_1_4_verb
+382 1 4 spatial 0/1 ood_382_verb_spatial_1_4_spatial
+383 0 3 verb 0/1 ood_383_verb_spatial_0_3_verb
+384 0 3 spatial 0/1 ood_384_verb_spatial_0_3_spatial
+385 2 3 verb 0/1 ood_385_verb_spatial_2_3_verb
+386 2 3 spatial 0/1 ood_386_verb_spatial_2_3_spatial
+387 0 1 verb 0/1 ood_387_verb_spatial_0_1_verb
+388 0 1 spatial 0/1 ood_388_verb_spatial_0_1_spatial
+389 4 2 verb 0/1 ood_389_verb_spatial_4_2_verb
+390 4 2 spatial 0/1 ood_390_verb_spatial_4_2_spatial
+391 4 1 verb 0/1 ood_391_verb_spatial_4_1_verb
+392 4 1 spatial 1/1 ood_392_verb_spatial_4_1_spatial
+393 1 4 verb 1/1 ood_393_verb_spatial_1_4_verb
+394 1 4 spatial 0/1 ood_394_verb_spatial_1_4_spatial
+395 4 2 verb 0/1 ood_395_verb_spatial_4_2_verb
+396 4 2 spatial 0/1 ood_396_verb_spatial_4_2_spatial
+397 1 4 verb 1/1 ood_397_verb_spatial_1_4_verb
+398 1 4 spatial 0/1 ood_398_verb_spatial_1_4_spatial
+399 0 1 verb 0/1 ood_399_verb_spatial_0_1_verb
+400 0 1 spatial 1/1 ood_400_verb_spatial_0_1_spatial
+
+overall_verb_success=14/200 (7.0%)
+overall_spatial_success=29/200 (14.5%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n100/SUMMARY.txt

@@ -0,0 +1,5 @@
+# all_factor_Lrandom_f50_n100 — full-factor TASK-aligned, single-process batch (sample_n=200 seed=42, 200 eps, max_steps=500, no_distractor=0.70, default difficulty, sim=gpu)
+seed40: 87/200 (43.5%)
+seed41: 99/200 (49.5%)
+seed42: 93/200 (46.5%)
+AVG over 3 seed(s): 46.5%

results/gr00t/all_factor/all_factor_Lrandom_f50_n100/full_factor_all_factor_Lrandom_f50_n100_seed40.txt

@@ -0,0 +1,212 @@
+# Full-factor inference (GR00T N1.7) [single-process batch]
+sample_n=200  sample_seed=42  total_cells=200
+total_episodes_target=200  num_episodes_per_cell=1
+total_episodes_actual=200
+host=127.0.0.1  port=5704
+sim_backend=gpu  render_backend=gpu
+max_episode_steps=500  seed_base=40
+no_distractor_prob=0.7  replan_steps=5
+
+index verb color shape spatial size prompt successes/total
+1 pull black car right small "Pull the small black car on the right." 0/1
+2 pull red cube right larger "Pull the larger red cube on the right." 0/1
+3 lift yellow sphere middle small "Lift the small yellow sphere in the middle." 1/1
+4 push red car behind small "Push the small red car at the back." 0/1
+5 grasp red cube middle smaller "Grasp the smaller red cube in the middle." 1/1
+6 grasp black cube front large "Grasp the large black cube in front." 1/1
+7 grasp orange sphere left smaller "Grasp the smaller orange sphere on the left." 1/1
+8 grasp black car middle large "Grasp the large black car in the middle." 0/1
+9 slide orange cup middle small "Slide the small orange cup in the middle." 1/1
+10 pull red car left large "Pull the large red car on the left." 0/1
+11 grasp blue cup middle smaller "Grasp the smaller blue cup in the middle." 1/1
+12 slide green star behind smaller "Slide the smaller green star at the back." 0/1
+13 lift orange cube left large "Lift the large orange cube on the left." 1/1
+14 rotate green sphere behind smaller "Rotate the smaller green sphere at the back." 0/1
+15 rotate green star behind larger "Rotate the larger green star at the back." 1/1
+16 pull blue cube middle large "Pull the large blue cube in the middle." 0/1
+17 pull red star right small "Pull the small red star on the right." 0/1
+18 lift red star middle small "Lift the small red star in the middle." 1/1
+19 push black cube left larger "Push the larger black cube on the left." 0/1
+20 slide red car right smaller "Slide the smaller red car on the right." 1/1
+21 rotate red star right large "Rotate the large red star on the right." 0/1
+22 slide orange car right small "Slide the small orange car on the right." 1/1
+23 slide black cube behind larger "Slide the larger black cube at the back." 1/1
+24 lift black pyramid behind larger "Lift the larger black pyramid at the back." 0/1
+25 pull green cube middle small "Pull the small green cube in the middle." 0/1
+26 pull orange cube right large "Pull the large orange cube on the right." 0/1
+27 grasp yellow cup right large "Grasp the large yellow cup on the right." 0/1
+28 rotate green cube front larger "Rotate the larger green cube in front." 1/1
+29 grasp yellow cup front smaller "Grasp the smaller yellow cup in front." 1/1
+30 lift orange cube middle larger "Lift the larger orange cube in the middle." 0/1
+31 rotate green cube right small "Rotate the small green cube on the right." 1/1
+32 slide blue sphere middle large "Slide the large blue sphere in the middle." 1/1
+33 rotate yellow car left smaller "Rotate the smaller yellow car on the left." 0/1
+34 grasp blue star middle larger "Grasp the larger blue star in the middle." 0/1
+35 rotate orange star behind smaller "Rotate the smaller orange star at the back." 0/1
+36 push yellow cup middle large "Push the large yellow cup in the middle." 1/1
+37 slide yellow car right small "Slide the small yellow car on the right." 1/1
+38 grasp yellow car left smaller "Grasp the smaller yellow car on the left." 0/1
+39 slide black cube front smaller "Slide the smaller black cube in front." 0/1
+40 push green cube front large "Push the large green cube in front." 1/1
+41 lift green car behind large "Lift the large green car at the back." 1/1
+42 slide yellow car middle smaller "Slide the smaller yellow car in the middle." 1/1
+43 pull green car middle larger "Pull the larger green car in the middle." 0/1
+44 grasp blue sphere front small "Grasp the small blue sphere in front." 0/1
+45 grasp red star right small "Grasp the small red star on the right." 0/1
+46 pull blue star behind smaller "Pull the smaller blue star at the back." 0/1
+47 grasp red sphere left large "Grasp the large red sphere on the left." 1/1
+48 push green cup right large "Push the large green cup on the right." 1/1
+49 push orange car front large "Push the large orange car in front." 0/1
+50 lift black cube left small "Lift the small black cube on the left." 0/1
+51 lift black star left smaller "Lift the smaller black star on the left." 0/1
+52 lift blue pyramid behind large "Lift the large blue pyramid at the back." 0/1
+53 lift black car front small "Lift the small black car in front." 0/1
+54 grasp yellow car middle small "Grasp the small yellow car in the middle." 1/1
+55 slide orange pyramid left small "Slide the small orange pyramid on the left." 1/1
+56 lift black sphere left smaller "Lift the smaller black sphere on the left." 1/1
+57 lift orange cup front larger "Lift the larger orange cup in front." 0/1
+58 push blue sphere front large "Push the large blue sphere in front." 0/1
+59 pull blue car front smaller "Pull the smaller blue car in front." 1/1
+60 push orange sphere right large "Push the large orange sphere on the right." 0/1
+61 rotate blue sphere left small "Rotate the small blue sphere on the left." 1/1
+62 grasp blue star front larger "Grasp the larger blue star in front." 1/1
+63 slide black cube right smaller "Slide the smaller black cube on the right." 0/1
+64 pull blue car middle small "Pull the small blue car in the middle." 0/1
+65 push green pyramid middle large "Push the large green pyramid in the middle." 0/1
+66 pull black pyramid left small "Pull the small black pyramid on the left." 0/1
+67 grasp blue pyramid behind large "Grasp the large blue pyramid at the back." 1/1
+68 lift orange car right smaller "Lift the smaller orange car on the right." 0/1
+69 pull green sphere behind large "Pull the large green sphere at the back." 0/1
+70 grasp yellow sphere front small "Grasp the small yellow sphere in front." 0/1
+71 slide orange pyramid front smaller "Slide the smaller orange pyramid in front." 0/1
+72 lift yellow cube middle larger "Lift the larger yellow cube in the middle." 0/1
+73 lift yellow star front larger "Lift the larger yellow star in front." 0/1
+74 push green cube front small "Push the small green cube in front." 0/1
+75 push orange sphere left small "Push the small orange sphere on the left." 1/1
+76 rotate orange pyramid behind small "Rotate the small orange pyramid at the back." 0/1
+77 slide red pyramid behind small "Slide the small red pyramid at the back." 1/1
+78 lift red star front small "Lift the small red star in front." 0/1
+79 push green car middle small "Push the small green car in the middle." 1/1
+80 grasp red cube front small "Grasp the small red cube in front." 1/1
+81 pull red cup front larger "Pull the larger red cup in front." 0/1
+82 pull red sphere middle large "Pull the large red sphere in the middle." 0/1
+83 lift green sphere front smaller "Lift the smaller green sphere in front." 0/1
+84 push yellow car left smaller "Push the smaller yellow car on the left." 0/1
+85 rotate blue cup behind small "Rotate the small blue cup at the back." 1/1
+86 lift yellow cup right smaller "Lift the smaller yellow cup on the right." 0/1
+87 pull red star left large "Pull the large red star on the left." 1/1
+88 lift red car front large "Lift the large red car in front." 0/1
+89 slide yellow sphere right larger "Slide the larger yellow sphere on the right." 0/1
+90 lift orange cup behind larger "Lift the larger orange cup at the back." 1/1
+91 lift red pyramid behind smaller "Lift the smaller red pyramid at the back." 0/1
+92 push red pyramid behind larger "Push the larger red pyramid at the back." 0/1
+93 pull blue cup behind small "Pull the small blue cup at the back." 0/1
+94 lift yellow car front smaller "Lift the smaller yellow car in front." 0/1
+95 grasp red cup front larger "Grasp the larger red cup in front." 0/1
+96 rotate red cube front smaller "Rotate the smaller red cube in front." 1/1
+97 lift black car right smaller "Lift the smaller black car on the right." 1/1
+98 lift orange cup right small "Lift the small orange cup on the right." 1/1
+99 pull green pyramid right smaller "Pull the smaller green pyramid on the right." 0/1
+100 pull blue sphere left large "Pull the large blue sphere on the left." 1/1
+101 push green star front larger "Push the larger green star in front." 0/1
+102 rotate red pyramid left larger "Rotate the larger red pyramid on the left." 1/1
+103 rotate red car right larger "Rotate the larger red car on the right." 0/1
+104 pull black pyramid right small "Pull the small black pyramid on the right." 0/1
+105 slide black cup behind smaller "Slide the smaller black cup at the back." 0/1
+106 lift orange cube left smaller "Lift the smaller orange cube on the left." 1/1
+107 grasp red car middle large "Grasp the large red car in the middle." 1/1
+108 rotate blue pyramid behind larger "Rotate the larger blue pyramid at the back." 1/1
+109 push blue sphere right larger "Push the larger blue sphere on the right." 0/1
+110 lift orange sphere behind larger "Lift the larger orange sphere at the back." 0/1
+111 pull red cup right larger "Pull the larger red cup on the right." 0/1
+112 push orange star right smaller "Push the smaller orange star on the right." 0/1
+113 grasp green sphere middle large "Grasp the large green sphere in the middle." 1/1
+114 rotate green pyramid left large "Rotate the large green pyramid on the left." 1/1
+115 grasp green star front smaller "Grasp the smaller green star in front." 0/1
+116 rotate blue pyramid right larger "Rotate the larger blue pyramid on the right." 0/1
+117 grasp orange car front large "Grasp the large orange car in front." 1/1
+118 lift orange star behind smaller "Lift the smaller orange star at the back." 0/1
+119 grasp black cup behind smaller "Grasp the smaller black cup at the back." 1/1
+120 push blue cup left large "Push the large blue cup on the left." 0/1
+121 grasp black car left larger "Grasp the larger black car on the left." 0/1
+122 rotate red star middle large "Rotate the large red star in the middle." 1/1
+123 lift blue star behind large "Lift the large blue star at the back." 1/1
+124 lift black cup behind larger "Lift the larger black cup at the back." 1/1
+125 slide black star front larger "Slide the larger black star in front." 0/1
+126 lift green star front large "Lift the large green star in front." 1/1
+127 grasp orange sphere front large "Grasp the large orange sphere in front." 0/1
+128 grasp yellow car front larger "Grasp the larger yellow car in front." 0/1
+129 push yellow cup right larger "Push the larger yellow cup on the right." 1/1
+130 slide yellow car behind large "Slide the large yellow car at the back." 1/1
+131 slide yellow cup right small "Slide the small yellow cup on the right." 1/1
+132 grasp red sphere middle smaller "Grasp the smaller red sphere in the middle." 1/1
+133 pull black car left smaller "Pull the smaller black car on the left." 0/1
+134 grasp yellow cube behind larger "Grasp the larger yellow cube at the back." 1/1
+135 push orange pyramid right larger "Push the larger orange pyramid on the right." 0/1
+136 pull green star left small "Pull the small green star on the left." 1/1
+137 pull orange car middle large "Pull the large orange car in the middle." 0/1
+138 lift blue cube right larger "Lift the larger blue cube on the right." 0/1
+139 lift black sphere behind larger "Lift the larger black sphere at the back." 0/1
+140 slide red pyramid middle small "Slide the small red pyramid in the middle." 0/1
+141 push black pyramid right larger "Push the larger black pyramid on the right." 0/1
+142 pull blue car front large "Pull the large blue car in front." 1/1
+143 rotate red cup left small "Rotate the small red cup on the left." 1/1
+144 pull green cup front larger "Pull the larger green cup in front." 1/1
+145 rotate black star middle smaller "Rotate the smaller black star in the middle." 1/1
+146 push orange sphere middle smaller "Push the smaller orange sphere in the middle." 1/1
+147 grasp green sphere middle larger "Grasp the larger green sphere in the middle." 1/1
+148 rotate yellow cup middle large "Rotate the large yellow cup in the middle." 1/1
+149 slide blue star middle smaller "Slide the smaller blue star in the middle." 0/1
+150 slide red cube behind smaller "Slide the smaller red cube at the back." 0/1
+151 slide blue cube left small "Slide the small blue cube on the left." 1/1
+152 grasp yellow cup front large "Grasp the large yellow cup in front." 0/1
+153 grasp blue cube front larger "Grasp the larger blue cube in front." 0/1
+154 grasp green cube front smaller "Grasp the smaller green cube in front." 1/1
+155 push orange star middle large "Push the large orange star in the middle." 0/1
+156 pull green cup left larger "Pull the larger green cup on the left." 0/1
+157 grasp yellow cube middle large "Grasp the large yellow cube in the middle." 1/1
+158 grasp green cube left small "Grasp the small green cube on the left." 1/1
+159 push green car left large "Push the large green car on the left." 1/1
+160 slide red cup right large "Slide the large red cup on the right." 1/1
+161 push black cup middle larger "Push the larger black cup in the middle." 1/1
+162 pull red car behind larger "Pull the larger red car at the back." 0/1
+163 slide green pyramid left larger "Slide the larger green pyramid on the left." 0/1
+164 slide orange pyramid behind small "Slide the small orange pyramid at the back." 0/1
+165 grasp black pyramid left small "Grasp the small black pyramid on the left." 1/1
+166 push yellow pyramid middle larger "Push the larger yellow pyramid in the middle." 0/1
+167 slide orange star left large "Slide the large orange star on the left." 0/1
+168 slide orange star front large "Slide the large orange star in front." 0/1
+169 grasp orange pyramid middle small "Grasp the small orange pyramid in the middle." 1/1
+170 lift red sphere behind large "Lift the large red sphere at the back." 1/1
+171 grasp red sphere behind large "Grasp the large red sphere at the back." 0/1
+172 slide black cup right smaller "Slide the smaller black cup on the right." 0/1
+173 lift green pyramid right larger "Lift the larger green pyramid on the right." 0/1
+174 slide blue cube middle small "Slide the small blue cube in the middle." 1/1
+175 push green cup front smaller "Push the smaller green cup in front." 0/1
+176 grasp blue cup behind larger "Grasp the larger blue cup at the back." 0/1
+177 slide green cup left smaller "Slide the smaller green cup on the left." 1/1
+178 grasp blue pyramid behind larger "Grasp the larger blue pyramid at the back." 1/1
+179 pull red star behind larger "Pull the larger red star at the back." 0/1
+180 pull red sphere behind larger "Pull the larger red sphere at the back." 0/1
+181 slide orange cup front larger "Slide the larger orange cup in front." 1/1
+182 pull black cup left small "Pull the small black cup on the left." 1/1
+183 lift red cube right large "Lift the large red cube on the right." 1/1
+184 push green star right larger "Push the larger green star on the right." 0/1
+185 pull yellow pyramid behind larger "Pull the larger yellow pyramid at the back." 0/1
+186 lift blue star front small "Lift the small blue star in front." 0/1
+187 rotate green cup front small "Rotate the small green cup in front." 0/1
+188 slide blue star left small "Slide the small blue star on the left." 0/1
+189 lift blue cube behind larger "Lift the larger blue cube at the back." 0/1
+190 rotate red cube right larger "Rotate the larger red cube on the right." 1/1
+191 lift yellow pyramid front small "Lift the small yellow pyramid in front." 0/1
+192 push blue cup middle smaller "Push the smaller blue cup in the middle." 1/1
+193 slide yellow star front larger "Slide the larger yellow star in front." 0/1
+194 rotate red car behind large "Rotate the large red car at the back." 1/1
+195 lift green car right small "Lift the small green car on the right." 1/1
+196 slide black cube left smaller "Slide the smaller black cube on the left." 0/1
+197 lift red sphere behind small "Lift the small red sphere at the back." 0/1
+198 lift orange car middle larger "Lift the larger orange car in the middle." 1/1
+199 push yellow sphere front smaller "Push the smaller yellow sphere in front." 0/1
+200 slide blue star right small "Slide the small blue star on the right." 0/1
+
+overall_success=87/200 (43.5%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n100/full_factor_all_factor_Lrandom_f50_n100_seed41.txt

@@ -0,0 +1,212 @@
+# Full-factor inference (GR00T N1.7) [single-process batch]
+sample_n=200  sample_seed=42  total_cells=200
+total_episodes_target=200  num_episodes_per_cell=1
+total_episodes_actual=200
+host=127.0.0.1  port=5704
+sim_backend=gpu  render_backend=gpu
+max_episode_steps=500  seed_base=41
+no_distractor_prob=0.7  replan_steps=5
+
+index verb color shape spatial size prompt successes/total
+1 pull black car right small "Pull the small black car on the right." 0/1
+2 pull red cube right larger "Pull the larger red cube on the right." 0/1
+3 lift yellow sphere middle small "Lift the small yellow sphere in the middle." 1/1
+4 push red car behind small "Push the small red car at the back." 0/1
+5 grasp red cube middle smaller "Grasp the smaller red cube in the middle." 1/1
+6 grasp black cube front large "Grasp the large black cube in front." 0/1
+7 grasp orange sphere left smaller "Grasp the smaller orange sphere on the left." 1/1
+8 grasp black car middle large "Grasp the large black car in the middle." 0/1
+9 slide orange cup middle small "Slide the small orange cup in the middle." 1/1
+10 pull red car left large "Pull the large red car on the left." 1/1
+11 grasp blue cup middle smaller "Grasp the smaller blue cup in the middle." 1/1
+12 slide green star behind smaller "Slide the smaller green star at the back." 1/1
+13 lift orange cube left large "Lift the large orange cube on the left." 1/1
+14 rotate green sphere behind smaller "Rotate the smaller green sphere at the back." 1/1
+15 rotate green star behind larger "Rotate the larger green star at the back." 0/1
+16 pull blue cube middle large "Pull the large blue cube in the middle." 0/1
+17 pull red star right small "Pull the small red star on the right." 0/1
+18 lift red star middle small "Lift the small red star in the middle." 0/1
+19 push black cube left larger "Push the larger black cube on the left." 1/1
+20 slide red car right smaller "Slide the smaller red car on the right." 1/1
+21 rotate red star right large "Rotate the large red star on the right." 0/1
+22 slide orange car right small "Slide the small orange car on the right." 1/1
+23 slide black cube behind larger "Slide the larger black cube at the back." 1/1
+24 lift black pyramid behind larger "Lift the larger black pyramid at the back." 0/1
+25 pull green cube middle small "Pull the small green cube in the middle." 0/1
+26 pull orange cube right large "Pull the large orange cube on the right." 0/1
+27 grasp yellow cup right large "Grasp the large yellow cup on the right." 0/1
+28 rotate green cube front larger "Rotate the larger green cube in front." 1/1
+29 grasp yellow cup front smaller "Grasp the smaller yellow cup in front." 1/1
+30 lift orange cube middle larger "Lift the larger orange cube in the middle." 1/1
+31 rotate green cube right small "Rotate the small green cube on the right." 1/1
+32 slide blue sphere middle large "Slide the large blue sphere in the middle." 1/1
+33 rotate yellow car left smaller "Rotate the smaller yellow car on the left." 0/1
+34 grasp blue star middle larger "Grasp the larger blue star in the middle." 0/1
+35 rotate orange star behind smaller "Rotate the smaller orange star at the back." 0/1
+36 push yellow cup middle large "Push the large yellow cup in the middle." 1/1
+37 slide yellow car right small "Slide the small yellow car on the right." 1/1
+38 grasp yellow car left smaller "Grasp the smaller yellow car on the left." 0/1
+39 slide black cube front smaller "Slide the smaller black cube in front." 1/1
+40 push green cube front large "Push the large green cube in front." 0/1
+41 lift green car behind large "Lift the large green car at the back." 1/1
+42 slide yellow car middle smaller "Slide the smaller yellow car in the middle." 0/1
+43 pull green car middle larger "Pull the larger green car in the middle." 0/1
+44 grasp blue sphere front small "Grasp the small blue sphere in front." 0/1
+45 grasp red star right small "Grasp the small red star on the right." 0/1
+46 pull blue star behind smaller "Pull the smaller blue star at the back." 0/1
+47 grasp red sphere left large "Grasp the large red sphere on the left." 1/1
+48 push green cup right large "Push the large green cup on the right." 0/1
+49 push orange car front large "Push the large orange car in front." 0/1
+50 lift black cube left small "Lift the small black cube on the left." 1/1
+51 lift black star left smaller "Lift the smaller black star on the left." 1/1
+52 lift blue pyramid behind large "Lift the large blue pyramid at the back." 0/1
+53 lift black car front small "Lift the small black car in front." 1/1
+54 grasp yellow car middle small "Grasp the small yellow car in the middle." 0/1
+55 slide orange pyramid left small "Slide the small orange pyramid on the left." 1/1
+56 lift black sphere left smaller "Lift the smaller black sphere on the left." 0/1
+57 lift orange cup front larger "Lift the larger orange cup in front." 1/1
+58 push blue sphere front large "Push the large blue sphere in front." 0/1
+59 pull blue car front smaller "Pull the smaller blue car in front." 0/1
+60 push orange sphere right large "Push the large orange sphere on the right." 1/1
+61 rotate blue sphere left small "Rotate the small blue sphere on the left." 1/1
+62 grasp blue star front larger "Grasp the larger blue star in front." 0/1
+63 slide black cube right smaller "Slide the smaller black cube on the right." 0/1
+64 pull blue car middle small "Pull the small blue car in the middle." 0/1
+65 push green pyramid middle large "Push the large green pyramid in the middle." 1/1
+66 pull black pyramid left small "Pull the small black pyramid on the left." 0/1
+67 grasp blue pyramid behind large "Grasp the large blue pyramid at the back." 1/1
+68 lift orange car right smaller "Lift the smaller orange car on the right." 1/1
+69 pull green sphere behind large "Pull the large green sphere at the back." 0/1
+70 grasp yellow sphere front small "Grasp the small yellow sphere in front." 0/1
+71 slide orange pyramid front smaller "Slide the smaller orange pyramid in front." 0/1
+72 lift yellow cube middle larger "Lift the larger yellow cube in the middle." 1/1
+73 lift yellow star front larger "Lift the larger yellow star in front." 0/1
+74 push green cube front small "Push the small green cube in front." 0/1
+75 push orange sphere left small "Push the small orange sphere on the left." 1/1
+76 rotate orange pyramid behind small "Rotate the small orange pyramid at the back." 0/1
+77 slide red pyramid behind small "Slide the small red pyramid at the back." 1/1
+78 lift red star front small "Lift the small red star in front." 1/1
+79 push green car middle small "Push the small green car in the middle." 1/1
+80 grasp red cube front small "Grasp the small red cube in front." 0/1
+81 pull red cup front larger "Pull the larger red cup in front." 1/1
+82 pull red sphere middle large "Pull the large red sphere in the middle." 1/1
+83 lift green sphere front smaller "Lift the smaller green sphere in front." 1/1
+84 push yellow car left smaller "Push the smaller yellow car on the left." 0/1
+85 rotate blue cup behind small "Rotate the small blue cup at the back." 0/1
+86 lift yellow cup right smaller "Lift the smaller yellow cup on the right." 1/1
+87 pull red star left large "Pull the large red star on the left." 1/1
+88 lift red car front large "Lift the large red car in front." 0/1
+89 slide yellow sphere right larger "Slide the larger yellow sphere on the right." 0/1
+90 lift orange cup behind larger "Lift the larger orange cup at the back." 0/1
+91 lift red pyramid behind smaller "Lift the smaller red pyramid at the back." 0/1
+92 push red pyramid behind larger "Push the larger red pyramid at the back." 0/1
+93 pull blue cup behind small "Pull the small blue cup at the back." 0/1
+94 lift yellow car front smaller "Lift the smaller yellow car in front." 0/1
+95 grasp red cup front larger "Grasp the larger red cup in front." 0/1
+96 rotate red cube front smaller "Rotate the smaller red cube in front." 1/1
+97 lift black car right smaller "Lift the smaller black car on the right." 1/1
+98 lift orange cup right small "Lift the small orange cup on the right." 1/1
+99 pull green pyramid right smaller "Pull the smaller green pyramid on the right." 0/1
+100 pull blue sphere left large "Pull the large blue sphere on the left." 1/1
+101 push green star front larger "Push the larger green star in front." 0/1
+102 rotate red pyramid left larger "Rotate the larger red pyramid on the left." 0/1
+103 rotate red car right larger "Rotate the larger red car on the right." 1/1
+104 pull black pyramid right small "Pull the small black pyramid on the right." 0/1
+105 slide black cup behind smaller "Slide the smaller black cup at the back." 0/1
+106 lift orange cube left smaller "Lift the smaller orange cube on the left." 1/1
+107 grasp red car middle large "Grasp the large red car in the middle." 0/1
+108 rotate blue pyramid behind larger "Rotate the larger blue pyramid at the back." 0/1
+109 push blue sphere right larger "Push the larger blue sphere on the right." 1/1
+110 lift orange sphere behind larger "Lift the larger orange sphere at the back." 0/1
+111 pull red cup right larger "Pull the larger red cup on the right." 0/1
+112 push orange star right smaller "Push the smaller orange star on the right." 1/1
+113 grasp green sphere middle large "Grasp the large green sphere in the middle." 1/1
+114 rotate green pyramid left large "Rotate the large green pyramid on the left." 1/1
+115 grasp green star front smaller "Grasp the smaller green star in front." 1/1
+116 rotate blue pyramid right larger "Rotate the larger blue pyramid on the right." 0/1
+117 grasp orange car front large "Grasp the large orange car in front." 1/1
+118 lift orange star behind smaller "Lift the smaller orange star at the back." 0/1
+119 grasp black cup behind smaller "Grasp the smaller black cup at the back." 0/1
+120 push blue cup left large "Push the large blue cup on the left." 1/1
+121 grasp black car left larger "Grasp the larger black car on the left." 0/1
+122 rotate red star middle large "Rotate the large red star in the middle." 0/1
+123 lift blue star behind large "Lift the large blue star at the back." 0/1
+124 lift black cup behind larger "Lift the larger black cup at the back." 0/1
+125 slide black star front larger "Slide the larger black star in front." 1/1
+126 lift green star front large "Lift the large green star in front." 1/1
+127 grasp orange sphere front large "Grasp the large orange sphere in front." 0/1
+128 grasp yellow car front larger "Grasp the larger yellow car in front." 0/1
+129 push yellow cup right larger "Push the larger yellow cup on the right." 1/1
+130 slide yellow car behind large "Slide the large yellow car at the back." 1/1
+131 slide yellow cup right small "Slide the small yellow cup on the right." 1/1
+132 grasp red sphere middle smaller "Grasp the smaller red sphere in the middle." 1/1
+133 pull black car left smaller "Pull the smaller black car on the left." 0/1
+134 grasp yellow cube behind larger "Grasp the larger yellow cube at the back." 1/1
+135 push orange pyramid right larger "Push the larger orange pyramid on the right." 1/1
+136 pull green star left small "Pull the small green star on the left." 1/1
+137 pull orange car middle large "Pull the large orange car in the middle." 0/1
+138 lift blue cube right larger "Lift the larger blue cube on the right." 0/1
+139 lift black sphere behind larger "Lift the larger black sphere at the back." 1/1
+140 slide red pyramid middle small "Slide the small red pyramid in the middle." 1/1
+141 push black pyramid right larger "Push the larger black pyramid on the right." 0/1
+142 pull blue car front large "Pull the large blue car in front." 0/1
+143 rotate red cup left small "Rotate the small red cup on the left." 1/1
+144 pull green cup front larger "Pull the larger green cup in front." 1/1
+145 rotate black star middle smaller "Rotate the smaller black star in the middle." 1/1
+146 push orange sphere middle smaller "Push the smaller orange sphere in the middle." 1/1
+147 grasp green sphere middle larger "Grasp the larger green sphere in the middle." 0/1
+148 rotate yellow cup middle large "Rotate the large yellow cup in the middle." 1/1
+149 slide blue star middle smaller "Slide the smaller blue star in the middle." 1/1
+150 slide red cube behind smaller "Slide the smaller red cube at the back." 0/1
+151 slide blue cube left small "Slide the small blue cube on the left." 1/1
+152 grasp yellow cup front large "Grasp the large yellow cup in front." 0/1
+153 grasp blue cube front larger "Grasp the larger blue cube in front." 1/1
+154 grasp green cube front smaller "Grasp the smaller green cube in front." 0/1
+155 push orange star middle large "Push the large orange star in the middle." 0/1
+156 pull green cup left larger "Pull the larger green cup on the left." 1/1
+157 grasp yellow cube middle large "Grasp the large yellow cube in the middle." 1/1
+158 grasp green cube left small "Grasp the small green cube on the left." 1/1
+159 push green car left large "Push the large green car on the left." 1/1
+160 slide red cup right large "Slide the large red cup on the right." 1/1
+161 push black cup middle larger "Push the larger black cup in the middle." 0/1
+162 pull red car behind larger "Pull the larger red car at the back." 0/1
+163 slide green pyramid left larger "Slide the larger green pyramid on the left." 1/1
+164 slide orange pyramid behind small "Slide the small orange pyramid at the back." 1/1
+165 grasp black pyramid left small "Grasp the small black pyramid on the left." 0/1
+166 push yellow pyramid middle larger "Push the larger yellow pyramid in the middle." 0/1
+167 slide orange star left large "Slide the large orange star on the left." 1/1
+168 slide orange star front large "Slide the large orange star in front." 0/1
+169 grasp orange pyramid middle small "Grasp the small orange pyramid in the middle." 1/1
+170 lift red sphere behind large "Lift the large red sphere at the back." 0/1
+171 grasp red sphere behind large "Grasp the large red sphere at the back." 0/1
+172 slide black cup right smaller "Slide the smaller black cup on the right." 1/1
+173 lift green pyramid right larger "Lift the larger green pyramid on the right." 0/1
+174 slide blue cube middle small "Slide the small blue cube in the middle." 1/1
+175 push green cup front smaller "Push the smaller green cup in front." 1/1
+176 grasp blue cup behind larger "Grasp the larger blue cup at the back." 0/1
+177 slide green cup left smaller "Slide the smaller green cup on the left." 1/1
+178 grasp blue pyramid behind larger "Grasp the larger blue pyramid at the back." 1/1
+179 pull red star behind larger "Pull the larger red star at the back." 0/1
+180 pull red sphere behind larger "Pull the larger red sphere at the back." 0/1
+181 slide orange cup front larger "Slide the larger orange cup in front." 1/1
+182 pull black cup left small "Pull the small black cup on the left." 0/1
+183 lift red cube right large "Lift the large red cube on the right." 1/1
+184 push green star right larger "Push the larger green star on the right." 0/1
+185 pull yellow pyramid behind larger "Pull the larger yellow pyramid at the back." 0/1
+186 lift blue star front small "Lift the small blue star in front." 0/1
+187 rotate green cup front small "Rotate the small green cup in front." 1/1
+188 slide blue star left small "Slide the small blue star on the left." 0/1
+189 lift blue cube behind larger "Lift the larger blue cube at the back." 1/1
+190 rotate red cube right larger "Rotate the larger red cube on the right." 1/1
+191 lift yellow pyramid front small "Lift the small yellow pyramid in front." 1/1
+192 push blue cup middle smaller "Push the smaller blue cup in the middle." 0/1
+193 slide yellow star front larger "Slide the larger yellow star in front." 0/1
+194 rotate red car behind large "Rotate the large red car at the back." 1/1
+195 lift green car right small "Lift the small green car on the right." 0/1
+196 slide black cube left smaller "Slide the smaller black cube on the left." 1/1
+197 lift red sphere behind small "Lift the small red sphere at the back." 0/1
+198 lift orange car middle larger "Lift the larger orange car in the middle." 1/1
+199 push yellow sphere front smaller "Push the smaller yellow sphere in front." 0/1
+200 slide blue star right small "Slide the small blue star on the right." 1/1
+
+overall_success=99/200 (49.5%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n100/full_factor_all_factor_Lrandom_f50_n100_seed42.txt

@@ -0,0 +1,212 @@
+# Full-factor inference (GR00T N1.7) [single-process batch]
+sample_n=200  sample_seed=42  total_cells=200
+total_episodes_target=200  num_episodes_per_cell=1
+total_episodes_actual=200
+host=127.0.0.1  port=5704
+sim_backend=gpu  render_backend=gpu
+max_episode_steps=500  seed_base=42
+no_distractor_prob=0.7  replan_steps=5
+
+index verb color shape spatial size prompt successes/total
+1 pull black car right small "Pull the small black car on the right." 0/1
+2 pull red cube right larger "Pull the larger red cube on the right." 0/1
+3 lift yellow sphere middle small "Lift the small yellow sphere in the middle." 1/1
+4 push red car behind small "Push the small red car at the back." 0/1
+5 grasp red cube middle smaller "Grasp the smaller red cube in the middle." 1/1
+6 grasp black cube front large "Grasp the large black cube in front." 0/1
+7 grasp orange sphere left smaller "Grasp the smaller orange sphere on the left." 1/1
+8 grasp black car middle large "Grasp the large black car in the middle." 0/1
+9 slide orange cup middle small "Slide the small orange cup in the middle." 0/1
+10 pull red car left large "Pull the large red car on the left." 1/1
+11 grasp blue cup middle smaller "Grasp the smaller blue cup in the middle." 1/1
+12 slide green star behind smaller "Slide the smaller green star at the back." 0/1
+13 lift orange cube left large "Lift the large orange cube on the left." 0/1
+14 rotate green sphere behind smaller "Rotate the smaller green sphere at the back." 1/1
+15 rotate green star behind larger "Rotate the larger green star at the back." 1/1
+16 pull blue cube middle large "Pull the large blue cube in the middle." 0/1
+17 pull red star right small "Pull the small red star on the right." 0/1
+18 lift red star middle small "Lift the small red star in the middle." 1/1
+19 push black cube left larger "Push the larger black cube on the left." 0/1
+20 slide red car right smaller "Slide the smaller red car on the right." 1/1
+21 rotate red star right large "Rotate the large red star on the right." 0/1
+22 slide orange car right small "Slide the small orange car on the right." 1/1
+23 slide black cube behind larger "Slide the larger black cube at the back." 0/1
+24 lift black pyramid behind larger "Lift the larger black pyramid at the back." 0/1
+25 pull green cube middle small "Pull the small green cube in the middle." 0/1
+26 pull orange cube right large "Pull the large orange cube on the right." 0/1
+27 grasp yellow cup right large "Grasp the large yellow cup on the right." 0/1
+28 rotate green cube front larger "Rotate the larger green cube in front." 1/1
+29 grasp yellow cup front smaller "Grasp the smaller yellow cup in front." 1/1
+30 lift orange cube middle larger "Lift the larger orange cube in the middle." 0/1
+31 rotate green cube right small "Rotate the small green cube on the right." 0/1
+32 slide blue sphere middle large "Slide the large blue sphere in the middle." 1/1
+33 rotate yellow car left smaller "Rotate the smaller yellow car on the left." 0/1
+34 grasp blue star middle larger "Grasp the larger blue star in the middle." 1/1
+35 rotate orange star behind smaller "Rotate the smaller orange star at the back." 1/1
+36 push yellow cup middle large "Push the large yellow cup in the middle." 1/1
+37 slide yellow car right small "Slide the small yellow car on the right." 1/1
+38 grasp yellow car left smaller "Grasp the smaller yellow car on the left." 0/1
+39 slide black cube front smaller "Slide the smaller black cube in front." 1/1
+40 push green cube front large "Push the large green cube in front." 0/1
+41 lift green car behind large "Lift the large green car at the back." 0/1
+42 slide yellow car middle smaller "Slide the smaller yellow car in the middle." 0/1
+43 pull green car middle larger "Pull the larger green car in the middle." 0/1
+44 grasp blue sphere front small "Grasp the small blue sphere in front." 0/1
+45 grasp red star right small "Grasp the small red star on the right." 0/1
+46 pull blue star behind smaller "Pull the smaller blue star at the back." 0/1
+47 grasp red sphere left large "Grasp the large red sphere on the left." 1/1
+48 push green cup right large "Push the large green cup on the right." 0/1
+49 push orange car front large "Push the large orange car in front." 0/1
+50 lift black cube left small "Lift the small black cube on the left." 0/1
+51 lift black star left smaller "Lift the smaller black star on the left." 0/1
+52 lift blue pyramid behind large "Lift the large blue pyramid at the back." 1/1
+53 lift black car front small "Lift the small black car in front." 0/1
+54 grasp yellow car middle small "Grasp the small yellow car in the middle." 1/1
+55 slide orange pyramid left small "Slide the small orange pyramid on the left." 1/1
+56 lift black sphere left smaller "Lift the smaller black sphere on the left." 1/1
+57 lift orange cup front larger "Lift the larger orange cup in front." 1/1
+58 push blue sphere front large "Push the large blue sphere in front." 0/1
+59 pull blue car front smaller "Pull the smaller blue car in front." 0/1
+60 push orange sphere right large "Push the large orange sphere on the right." 0/1
+61 rotate blue sphere left small "Rotate the small blue sphere on the left." 1/1
+62 grasp blue star front larger "Grasp the larger blue star in front." 0/1
+63 slide black cube right smaller "Slide the smaller black cube on the right." 1/1
+64 pull blue car middle small "Pull the small blue car in the middle." 0/1
+65 push green pyramid middle large "Push the large green pyramid in the middle." 0/1
+66 pull black pyramid left small "Pull the small black pyramid on the left." 0/1
+67 grasp blue pyramid behind large "Grasp the large blue pyramid at the back." 1/1
+68 lift orange car right smaller "Lift the smaller orange car on the right." 1/1
+69 pull green sphere behind large "Pull the large green sphere at the back." 0/1
+70 grasp yellow sphere front small "Grasp the small yellow sphere in front." 1/1
+71 slide orange pyramid front smaller "Slide the smaller orange pyramid in front." 1/1
+72 lift yellow cube middle larger "Lift the larger yellow cube in the middle." 1/1
+73 lift yellow star front larger "Lift the larger yellow star in front." 1/1
+74 push green cube front small "Push the small green cube in front." 0/1
+75 push orange sphere left small "Push the small orange sphere on the left." 1/1
+76 rotate orange pyramid behind small "Rotate the small orange pyramid at the back." 0/1
+77 slide red pyramid behind small "Slide the small red pyramid at the back." 0/1
+78 lift red star front small "Lift the small red star in front." 1/1
+79 push green car middle small "Push the small green car in the middle." 0/1
+80 grasp red cube front small "Grasp the small red cube in front." 1/1
+81 pull red cup front larger "Pull the larger red cup in front." 0/1
+82 pull red sphere middle large "Pull the large red sphere in the middle." 1/1
+83 lift green sphere front smaller "Lift the smaller green sphere in front." 0/1
+84 push yellow car left smaller "Push the smaller yellow car on the left." 0/1
+85 rotate blue cup behind small "Rotate the small blue cup at the back." 0/1
+86 lift yellow cup right smaller "Lift the smaller yellow cup on the right." 1/1
+87 pull red star left large "Pull the large red star on the left." 0/1
+88 lift red car front large "Lift the large red car in front." 0/1
+89 slide yellow sphere right larger "Slide the larger yellow sphere on the right." 0/1
+90 lift orange cup behind larger "Lift the larger orange cup at the back." 1/1
+91 lift red pyramid behind smaller "Lift the smaller red pyramid at the back." 0/1
+92 push red pyramid behind larger "Push the larger red pyramid at the back." 0/1
+93 pull blue cup behind small "Pull the small blue cup at the back." 0/1
+94 lift yellow car front smaller "Lift the smaller yellow car in front." 0/1
+95 grasp red cup front larger "Grasp the larger red cup in front." 1/1
+96 rotate red cube front smaller "Rotate the smaller red cube in front." 1/1
+97 lift black car right smaller "Lift the smaller black car on the right." 1/1
+98 lift orange cup right small "Lift the small orange cup on the right." 1/1
+99 pull green pyramid right smaller "Pull the smaller green pyramid on the right." 0/1
+100 pull blue sphere left large "Pull the large blue sphere on the left." 1/1
+101 push green star front larger "Push the larger green star in front." 0/1
+102 rotate red pyramid left larger "Rotate the larger red pyramid on the left." 0/1
+103 rotate red car right larger "Rotate the larger red car on the right." 1/1
+104 pull black pyramid right small "Pull the small black pyramid on the right." 0/1
+105 slide black cup behind smaller "Slide the smaller black cup at the back." 0/1
+106 lift orange cube left smaller "Lift the smaller orange cube on the left." 1/1
+107 grasp red car middle large "Grasp the large red car in the middle." 0/1
+108 rotate blue pyramid behind larger "Rotate the larger blue pyramid at the back." 0/1
+109 push blue sphere right larger "Push the larger blue sphere on the right." 1/1
+110 lift orange sphere behind larger "Lift the larger orange sphere at the back." 0/1
+111 pull red cup right larger "Pull the larger red cup on the right." 0/1
+112 push orange star right smaller "Push the smaller orange star on the right." 0/1
+113 grasp green sphere middle large "Grasp the large green sphere in the middle." 1/1
+114 rotate green pyramid left large "Rotate the large green pyramid on the left." 0/1
+115 grasp green star front smaller "Grasp the smaller green star in front." 0/1
+116 rotate blue pyramid right larger "Rotate the larger blue pyramid on the right." 1/1
+117 grasp orange car front large "Grasp the large orange car in front." 0/1
+118 lift orange star behind smaller "Lift the smaller orange star at the back." 1/1
+119 grasp black cup behind smaller "Grasp the smaller black cup at the back." 0/1
+120 push blue cup left large "Push the large blue cup on the left." 1/1
+121 grasp black car left larger "Grasp the larger black car on the left." 0/1
+122 rotate red star middle large "Rotate the large red star in the middle." 1/1
+123 lift blue star behind large "Lift the large blue star at the back." 1/1
+124 lift black cup behind larger "Lift the larger black cup at the back." 0/1
+125 slide black star front larger "Slide the larger black star in front." 0/1
+126 lift green star front large "Lift the large green star in front." 0/1
+127 grasp orange sphere front large "Grasp the large orange sphere in front." 0/1
+128 grasp yellow car front larger "Grasp the larger yellow car in front." 0/1
+129 push yellow cup right larger "Push the larger yellow cup on the right." 1/1
+130 slide yellow car behind large "Slide the large yellow car at the back." 1/1
+131 slide yellow cup right small "Slide the small yellow cup on the right." 0/1
+132 grasp red sphere middle smaller "Grasp the smaller red sphere in the middle." 1/1
+133 pull black car left smaller "Pull the smaller black car on the left." 0/1
+134 grasp yellow cube behind larger "Grasp the larger yellow cube at the back." 0/1
+135 push orange pyramid right larger "Push the larger orange pyramid on the right." 1/1
+136 pull green star left small "Pull the small green star on the left." 0/1
+137 pull orange car middle large "Pull the large orange car in the middle." 0/1
+138 lift blue cube right larger "Lift the larger blue cube on the right." 1/1
+139 lift black sphere behind larger "Lift the larger black sphere at the back." 1/1
+140 slide red pyramid middle small "Slide the small red pyramid in the middle." 1/1
+141 push black pyramid right larger "Push the larger black pyramid on the right." 0/1
+142 pull blue car front large "Pull the large blue car in front." 0/1
+143 rotate red cup left small "Rotate the small red cup on the left." 1/1
+144 pull green cup front larger "Pull the larger green cup in front." 1/1
+145 rotate black star middle smaller "Rotate the smaller black star in the middle." 1/1
+146 push orange sphere middle smaller "Push the smaller orange sphere in the middle." 1/1
+147 grasp green sphere middle larger "Grasp the larger green sphere in the middle." 1/1
+148 rotate yellow cup middle large "Rotate the large yellow cup in the middle." 1/1
+149 slide blue star middle smaller "Slide the smaller blue star in the middle." 0/1
+150 slide red cube behind smaller "Slide the smaller red cube at the back." 1/1
+151 slide blue cube left small "Slide the small blue cube on the left." 1/1
+152 grasp yellow cup front large "Grasp the large yellow cup in front." 1/1
+153 grasp blue cube front larger "Grasp the larger blue cube in front." 1/1
+154 grasp green cube front smaller "Grasp the smaller green cube in front." 1/1
+155 push orange star middle large "Push the large orange star in the middle." 1/1
+156 pull green cup left larger "Pull the larger green cup on the left." 1/1
+157 grasp yellow cube middle large "Grasp the large yellow cube in the middle." 0/1
+158 grasp green cube left small "Grasp the small green cube on the left." 1/1
+159 push green car left large "Push the large green car on the left." 1/1
+160 slide red cup right large "Slide the large red cup on the right." 1/1
+161 push black cup middle larger "Push the larger black cup in the middle." 1/1
+162 pull red car behind larger "Pull the larger red car at the back." 0/1
+163 slide green pyramid left larger "Slide the larger green pyramid on the left." 0/1
+164 slide orange pyramid behind small "Slide the small orange pyramid at the back." 0/1
+165 grasp black pyramid left small "Grasp the small black pyramid on the left." 0/1
+166 push yellow pyramid middle larger "Push the larger yellow pyramid in the middle." 1/1
+167 slide orange star left large "Slide the large orange star on the left." 1/1
+168 slide orange star front large "Slide the large orange star in front." 0/1
+169 grasp orange pyramid middle small "Grasp the small orange pyramid in the middle." 1/1
+170 lift red sphere behind large "Lift the large red sphere at the back." 0/1
+171 grasp red sphere behind large "Grasp the large red sphere at the back." 0/1
+172 slide black cup right smaller "Slide the smaller black cup on the right." 1/1
+173 lift green pyramid right larger "Lift the larger green pyramid on the right." 1/1
+174 slide blue cube middle small "Slide the small blue cube in the middle." 1/1
+175 push green cup front smaller "Push the smaller green cup in front." 0/1
+176 grasp blue cup behind larger "Grasp the larger blue cup at the back." 1/1
+177 slide green cup left smaller "Slide the smaller green cup on the left." 1/1
+178 grasp blue pyramid behind larger "Grasp the larger blue pyramid at the back." 0/1
+179 pull red star behind larger "Pull the larger red star at the back." 0/1
+180 pull red sphere behind larger "Pull the larger red sphere at the back." 0/1
+181 slide orange cup front larger "Slide the larger orange cup in front." 1/1
+182 pull black cup left small "Pull the small black cup on the left." 0/1
+183 lift red cube right large "Lift the large red cube on the right." 1/1
+184 push green star right larger "Push the larger green star on the right." 1/1
+185 pull yellow pyramid behind larger "Pull the larger yellow pyramid at the back." 0/1
+186 lift blue star front small "Lift the small blue star in front." 0/1
+187 rotate green cup front small "Rotate the small green cup in front." 0/1
+188 slide blue star left small "Slide the small blue star on the left." 1/1
+189 lift blue cube behind larger "Lift the larger blue cube at the back." 0/1
+190 rotate red cube right larger "Rotate the larger red cube on the right." 1/1
+191 lift yellow pyramid front small "Lift the small yellow pyramid in front." 0/1
+192 push blue cup middle smaller "Push the smaller blue cup in the middle." 1/1
+193 slide yellow star front larger "Slide the larger yellow star in front." 0/1
+194 rotate red car behind large "Rotate the large red car at the back." 0/1
+195 lift green car right small "Lift the small green car on the right." 0/1
+196 slide black cube left smaller "Slide the smaller black cube on the left." 1/1
+197 lift red sphere behind small "Lift the small red sphere at the back." 1/1
+198 lift orange car middle larger "Lift the larger orange car in the middle." 1/1
+199 push yellow sphere front smaller "Push the smaller yellow sphere in front." 0/1
+200 slide blue star right small "Slide the small blue star on the right." 0/1
+
+overall_success=93/200 (46.5%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n200/SUMMARY.txt

@@ -0,0 +1,5 @@
+# all_factor_Lrandom_f50_n200 — full-factor TASK-aligned, single-process batch (sample_n=200 seed=42, 200 eps, max_steps=500, no_distractor=0.70, default difficulty, sim=gpu)
+seed40: 86/200 (43.0%)
+seed41: 72/200 (36.0%)
+seed42: 88/200 (44.0%)
+AVG over 3 seed(s): 41.0%

results/gr00t/all_factor/all_factor_Lrandom_f50_n200/full_factor_all_factor_Lrandom_f50_n200_seed40.txt

@@ -0,0 +1,212 @@
+# Full-factor inference (GR00T N1.7) [single-process batch]
+sample_n=200  sample_seed=42  total_cells=200
+total_episodes_target=200  num_episodes_per_cell=1
+total_episodes_actual=200
+host=127.0.0.1  port=5707
+sim_backend=gpu  render_backend=gpu
+max_episode_steps=500  seed_base=40
+no_distractor_prob=0.7  replan_steps=5
+
+index verb color shape spatial size prompt successes/total
+1 pull black car right small "Pull the small black car on the right." 0/1
+2 pull red cube right larger "Pull the larger red cube on the right." 0/1
+3 lift yellow sphere middle small "Lift the small yellow sphere in the middle." 0/1
+4 push red car behind small "Push the small red car at the back." 0/1
+5 grasp red cube middle smaller "Grasp the smaller red cube in the middle." 1/1
+6 grasp black cube front large "Grasp the large black cube in front." 0/1
+7 grasp orange sphere left smaller "Grasp the smaller orange sphere on the left." 1/1
+8 grasp black car middle large "Grasp the large black car in the middle." 1/1
+9 slide orange cup middle small "Slide the small orange cup in the middle." 0/1
+10 pull red car left large "Pull the large red car on the left." 0/1
+11 grasp blue cup middle smaller "Grasp the smaller blue cup in the middle." 1/1
+12 slide green star behind smaller "Slide the smaller green star at the back." 1/1
+13 lift orange cube left large "Lift the large orange cube on the left." 1/1
+14 rotate green sphere behind smaller "Rotate the smaller green sphere at the back." 1/1
+15 rotate green star behind larger "Rotate the larger green star at the back." 0/1
+16 pull blue cube middle large "Pull the large blue cube in the middle." 0/1
+17 pull red star right small "Pull the small red star on the right." 0/1
+18 lift red star middle small "Lift the small red star in the middle." 0/1
+19 push black cube left larger "Push the larger black cube on the left." 0/1
+20 slide red car right smaller "Slide the smaller red car on the right." 0/1
+21 rotate red star right large "Rotate the large red star on the right." 0/1
+22 slide orange car right small "Slide the small orange car on the right." 1/1
+23 slide black cube behind larger "Slide the larger black cube at the back." 1/1
+24 lift black pyramid behind larger "Lift the larger black pyramid at the back." 0/1
+25 pull green cube middle small "Pull the small green cube in the middle." 0/1
+26 pull orange cube right large "Pull the large orange cube on the right." 0/1
+27 grasp yellow cup right large "Grasp the large yellow cup on the right." 0/1
+28 rotate green cube front larger "Rotate the larger green cube in front." 1/1
+29 grasp yellow cup front smaller "Grasp the smaller yellow cup in front." 0/1
+30 lift orange cube middle larger "Lift the larger orange cube in the middle." 0/1
+31 rotate green cube right small "Rotate the small green cube on the right." 1/1
+32 slide blue sphere middle large "Slide the large blue sphere in the middle." 0/1
+33 rotate yellow car left smaller "Rotate the smaller yellow car on the left." 1/1
+34 grasp blue star middle larger "Grasp the larger blue star in the middle." 0/1
+35 rotate orange star behind smaller "Rotate the smaller orange star at the back." 0/1
+36 push yellow cup middle large "Push the large yellow cup in the middle." 0/1
+37 slide yellow car right small "Slide the small yellow car on the right." 1/1
+38 grasp yellow car left smaller "Grasp the smaller yellow car on the left." 0/1
+39 slide black cube front smaller "Slide the smaller black cube in front." 1/1
+40 push green cube front large "Push the large green cube in front." 1/1
+41 lift green car behind large "Lift the large green car at the back." 1/1
+42 slide yellow car middle smaller "Slide the smaller yellow car in the middle." 1/1
+43 pull green car middle larger "Pull the larger green car in the middle." 0/1
+44 grasp blue sphere front small "Grasp the small blue sphere in front." 1/1
+45 grasp red star right small "Grasp the small red star on the right." 0/1
+46 pull blue star behind smaller "Pull the smaller blue star at the back." 0/1
+47 grasp red sphere left large "Grasp the large red sphere on the left." 1/1
+48 push green cup right large "Push the large green cup on the right." 1/1
+49 push orange car front large "Push the large orange car in front." 0/1
+50 lift black cube left small "Lift the small black cube on the left." 1/1
+51 lift black star left smaller "Lift the smaller black star on the left." 0/1
+52 lift blue pyramid behind large "Lift the large blue pyramid at the back." 0/1
+53 lift black car front small "Lift the small black car in front." 0/1
+54 grasp yellow car middle small "Grasp the small yellow car in the middle." 1/1
+55 slide orange pyramid left small "Slide the small orange pyramid on the left." 0/1
+56 lift black sphere left smaller "Lift the smaller black sphere on the left." 1/1
+57 lift orange cup front larger "Lift the larger orange cup in front." 0/1
+58 push blue sphere front large "Push the large blue sphere in front." 0/1
+59 pull blue car front smaller "Pull the smaller blue car in front." 0/1
+60 push orange sphere right large "Push the large orange sphere on the right." 0/1
+61 rotate blue sphere left small "Rotate the small blue sphere on the left." 1/1
+62 grasp blue star front larger "Grasp the larger blue star in front." 1/1
+63 slide black cube right smaller "Slide the smaller black cube on the right." 1/1
+64 pull blue car middle small "Pull the small blue car in the middle." 0/1
+65 push green pyramid middle large "Push the large green pyramid in the middle." 0/1
+66 pull black pyramid left small "Pull the small black pyramid on the left." 0/1
+67 grasp blue pyramid behind large "Grasp the large blue pyramid at the back." 0/1
+68 lift orange car right smaller "Lift the smaller orange car on the right." 0/1
+69 pull green sphere behind large "Pull the large green sphere at the back." 0/1
+70 grasp yellow sphere front small "Grasp the small yellow sphere in front." 1/1
+71 slide orange pyramid front smaller "Slide the smaller orange pyramid in front." 1/1
+72 lift yellow cube middle larger "Lift the larger yellow cube in the middle." 0/1
+73 lift yellow star front larger "Lift the larger yellow star in front." 1/1
+74 push green cube front small "Push the small green cube in front." 1/1
+75 push orange sphere left small "Push the small orange sphere on the left." 1/1
+76 rotate orange pyramid behind small "Rotate the small orange pyramid at the back." 0/1
+77 slide red pyramid behind small "Slide the small red pyramid at the back." 1/1
+78 lift red star front small "Lift the small red star in front." 1/1
+79 push green car middle small "Push the small green car in the middle." 0/1
+80 grasp red cube front small "Grasp the small red cube in front." 1/1
+81 pull red cup front larger "Pull the larger red cup in front." 1/1
+82 pull red sphere middle large "Pull the large red sphere in the middle." 1/1
+83 lift green sphere front smaller "Lift the smaller green sphere in front." 0/1
+84 push yellow car left smaller "Push the smaller yellow car on the left." 0/1
+85 rotate blue cup behind small "Rotate the small blue cup at the back." 1/1
+86 lift yellow cup right smaller "Lift the smaller yellow cup on the right." 1/1
+87 pull red star left large "Pull the large red star on the left." 0/1
+88 lift red car front large "Lift the large red car in front." 0/1
+89 slide yellow sphere right larger "Slide the larger yellow sphere on the right." 0/1
+90 lift orange cup behind larger "Lift the larger orange cup at the back." 1/1
+91 lift red pyramid behind smaller "Lift the smaller red pyramid at the back." 0/1
+92 push red pyramid behind larger "Push the larger red pyramid at the back." 1/1
+93 pull blue cup behind small "Pull the small blue cup at the back." 0/1
+94 lift yellow car front smaller "Lift the smaller yellow car in front." 0/1
+95 grasp red cup front larger "Grasp the larger red cup in front." 0/1
+96 rotate red cube front smaller "Rotate the smaller red cube in front." 1/1
+97 lift black car right smaller "Lift the smaller black car on the right." 0/1
+98 lift orange cup right small "Lift the small orange cup on the right." 1/1
+99 pull green pyramid right smaller "Pull the smaller green pyramid on the right." 0/1
+100 pull blue sphere left large "Pull the large blue sphere on the left." 1/1
+101 push green star front larger "Push the larger green star in front." 0/1
+102 rotate red pyramid left larger "Rotate the larger red pyramid on the left." 0/1
+103 rotate red car right larger "Rotate the larger red car on the right." 1/1
+104 pull black pyramid right small "Pull the small black pyramid on the right." 0/1
+105 slide black cup behind smaller "Slide the smaller black cup at the back." 1/1
+106 lift orange cube left smaller "Lift the smaller orange cube on the left." 1/1
+107 grasp red car middle large "Grasp the large red car in the middle." 0/1
+108 rotate blue pyramid behind larger "Rotate the larger blue pyramid at the back." 1/1
+109 push blue sphere right larger "Push the larger blue sphere on the right." 0/1
+110 lift orange sphere behind larger "Lift the larger orange sphere at the back." 0/1
+111 pull red cup right larger "Pull the larger red cup on the right." 0/1
+112 push orange star right smaller "Push the smaller orange star on the right." 0/1
+113 grasp green sphere middle large "Grasp the large green sphere in the middle." 0/1
+114 rotate green pyramid left large "Rotate the large green pyramid on the left." 1/1
+115 grasp green star front smaller "Grasp the smaller green star in front." 0/1
+116 rotate blue pyramid right larger "Rotate the larger blue pyramid on the right." 1/1
+117 grasp orange car front large "Grasp the large orange car in front." 0/1
+118 lift orange star behind smaller "Lift the smaller orange star at the back." 0/1
+119 grasp black cup behind smaller "Grasp the smaller black cup at the back." 0/1
+120 push blue cup left large "Push the large blue cup on the left." 0/1
+121 grasp black car left larger "Grasp the larger black car on the left." 0/1
+122 rotate red star middle large "Rotate the large red star in the middle." 0/1
+123 lift blue star behind large "Lift the large blue star at the back." 1/1
+124 lift black cup behind larger "Lift the larger black cup at the back." 0/1
+125 slide black star front larger "Slide the larger black star in front." 1/1
+126 lift green star front large "Lift the large green star in front." 0/1
+127 grasp orange sphere front large "Grasp the large orange sphere in front." 1/1
+128 grasp yellow car front larger "Grasp the larger yellow car in front." 0/1
+129 push yellow cup right larger "Push the larger yellow cup on the right." 0/1
+130 slide yellow car behind large "Slide the large yellow car at the back." 1/1
+131 slide yellow cup right small "Slide the small yellow cup on the right." 1/1
+132 grasp red sphere middle smaller "Grasp the smaller red sphere in the middle." 1/1
+133 pull black car left smaller "Pull the smaller black car on the left." 0/1
+134 grasp yellow cube behind larger "Grasp the larger yellow cube at the back." 0/1
+135 push orange pyramid right larger "Push the larger orange pyramid on the right." 0/1
+136 pull green star left small "Pull the small green star on the left." 0/1
+137 pull orange car middle large "Pull the large orange car in the middle." 0/1
+138 lift blue cube right larger "Lift the larger blue cube on the right." 0/1
+139 lift black sphere behind larger "Lift the larger black sphere at the back." 1/1
+140 slide red pyramid middle small "Slide the small red pyramid in the middle." 0/1
+141 push black pyramid right larger "Push the larger black pyramid on the right." 0/1
+142 pull blue car front large "Pull the large blue car in front." 0/1
+143 rotate red cup left small "Rotate the small red cup on the left." 1/1
+144 pull green cup front larger "Pull the larger green cup in front." 1/1
+145 rotate black star middle smaller "Rotate the smaller black star in the middle." 0/1
+146 push orange sphere middle smaller "Push the smaller orange sphere in the middle." 1/1
+147 grasp green sphere middle larger "Grasp the larger green sphere in the middle." 1/1
+148 rotate yellow cup middle large "Rotate the large yellow cup in the middle." 1/1
+149 slide blue star middle smaller "Slide the smaller blue star in the middle." 0/1
+150 slide red cube behind smaller "Slide the smaller red cube at the back." 1/1
+151 slide blue cube left small "Slide the small blue cube on the left." 1/1
+152 grasp yellow cup front large "Grasp the large yellow cup in front." 0/1
+153 grasp blue cube front larger "Grasp the larger blue cube in front." 1/1
+154 grasp green cube front smaller "Grasp the smaller green cube in front." 1/1
+155 push orange star middle large "Push the large orange star in the middle." 0/1
+156 pull green cup left larger "Pull the larger green cup on the left." 0/1
+157 grasp yellow cube middle large "Grasp the large yellow cube in the middle." 1/1
+158 grasp green cube left small "Grasp the small green cube on the left." 1/1
+159 push green car left large "Push the large green car on the left." 1/1
+160 slide red cup right large "Slide the large red cup on the right." 1/1
+161 push black cup middle larger "Push the larger black cup in the middle." 1/1
+162 pull red car behind larger "Pull the larger red car at the back." 0/1
+163 slide green pyramid left larger "Slide the larger green pyramid on the left." 1/1
+164 slide orange pyramid behind small "Slide the small orange pyramid at the back." 0/1
+165 grasp black pyramid left small "Grasp the small black pyramid on the left." 0/1
+166 push yellow pyramid middle larger "Push the larger yellow pyramid in the middle." 0/1
+167 slide orange star left large "Slide the large orange star on the left." 1/1
+168 slide orange star front large "Slide the large orange star in front." 0/1
+169 grasp orange pyramid middle small "Grasp the small orange pyramid in the middle." 0/1
+170 lift red sphere behind large "Lift the large red sphere at the back." 0/1
+171 grasp red sphere behind large "Grasp the large red sphere at the back." 0/1
+172 slide black cup right smaller "Slide the smaller black cup on the right." 1/1
+173 lift green pyramid right larger "Lift the larger green pyramid on the right." 1/1
+174 slide blue cube middle small "Slide the small blue cube in the middle." 1/1
+175 push green cup front smaller "Push the smaller green cup in front." 1/1
+176 grasp blue cup behind larger "Grasp the larger blue cup at the back." 0/1
+177 slide green cup left smaller "Slide the smaller green cup on the left." 1/1
+178 grasp blue pyramid behind larger "Grasp the larger blue pyramid at the back." 0/1
+179 pull red star behind larger "Pull the larger red star at the back." 0/1
+180 pull red sphere behind larger "Pull the larger red sphere at the back." 0/1
+181 slide orange cup front larger "Slide the larger orange cup in front." 1/1
+182 pull black cup left small "Pull the small black cup on the left." 1/1
+183 lift red cube right large "Lift the large red cube on the right." 0/1
+184 push green star right larger "Push the larger green star on the right." 0/1
+185 pull yellow pyramid behind larger "Pull the larger yellow pyramid at the back." 0/1
+186 lift blue star front small "Lift the small blue star in front." 1/1
+187 rotate green cup front small "Rotate the small green cup in front." 1/1
+188 slide blue star left small "Slide the small blue star on the left." 0/1
+189 lift blue cube behind larger "Lift the larger blue cube at the back." 0/1
+190 rotate red cube right larger "Rotate the larger red cube on the right." 1/1
+191 lift yellow pyramid front small "Lift the small yellow pyramid in front." 0/1
+192 push blue cup middle smaller "Push the smaller blue cup in the middle." 0/1
+193 slide yellow star front larger "Slide the larger yellow star in front." 1/1
+194 rotate red car behind large "Rotate the large red car at the back." 1/1
+195 lift green car right small "Lift the small green car on the right." 0/1
+196 slide black cube left smaller "Slide the smaller black cube on the left." 1/1
+197 lift red sphere behind small "Lift the small red sphere at the back." 0/1
+198 lift orange car middle larger "Lift the larger orange car in the middle." 1/1
+199 push yellow sphere front smaller "Push the smaller yellow sphere in front." 0/1
+200 slide blue star right small "Slide the small blue star on the right." 0/1
+
+overall_success=86/200 (43.0%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n200/full_factor_all_factor_Lrandom_f50_n200_seed41.txt

@@ -0,0 +1,212 @@
+# Full-factor inference (GR00T N1.7) [single-process batch]
+sample_n=200  sample_seed=42  total_cells=200
+total_episodes_target=200  num_episodes_per_cell=1
+total_episodes_actual=200
+host=127.0.0.1  port=5707
+sim_backend=gpu  render_backend=gpu
+max_episode_steps=500  seed_base=41
+no_distractor_prob=0.7  replan_steps=5
+
+index verb color shape spatial size prompt successes/total
+1 pull black car right small "Pull the small black car on the right." 0/1
+2 pull red cube right larger "Pull the larger red cube on the right." 0/1
+3 lift yellow sphere middle small "Lift the small yellow sphere in the middle." 1/1
+4 push red car behind small "Push the small red car at the back." 0/1
+5 grasp red cube middle smaller "Grasp the smaller red cube in the middle." 0/1
+6 grasp black cube front large "Grasp the large black cube in front." 1/1
+7 grasp orange sphere left smaller "Grasp the smaller orange sphere on the left." 0/1
+8 grasp black car middle large "Grasp the large black car in the middle." 0/1
+9 slide orange cup middle small "Slide the small orange cup in the middle." 1/1
+10 pull red car left large "Pull the large red car on the left." 0/1
+11 grasp blue cup middle smaller "Grasp the smaller blue cup in the middle." 0/1
+12 slide green star behind smaller "Slide the smaller green star at the back." 0/1
+13 lift orange cube left large "Lift the large orange cube on the left." 0/1
+14 rotate green sphere behind smaller "Rotate the smaller green sphere at the back." 0/1
+15 rotate green star behind larger "Rotate the larger green star at the back." 0/1
+16 pull blue cube middle large "Pull the large blue cube in the middle." 0/1
+17 pull red star right small "Pull the small red star on the right." 0/1
+18 lift red star middle small "Lift the small red star in the middle." 0/1
+19 push black cube left larger "Push the larger black cube on the left." 0/1
+20 slide red car right smaller "Slide the smaller red car on the right." 0/1
+21 rotate red star right large "Rotate the large red star on the right." 0/1
+22 slide orange car right small "Slide the small orange car on the right." 1/1
+23 slide black cube behind larger "Slide the larger black cube at the back." 1/1
+24 lift black pyramid behind larger "Lift the larger black pyramid at the back." 0/1
+25 pull green cube middle small "Pull the small green cube in the middle." 0/1
+26 pull orange cube right large "Pull the large orange cube on the right." 0/1
+27 grasp yellow cup right large "Grasp the large yellow cup on the right." 0/1
+28 rotate green cube front larger "Rotate the larger green cube in front." 1/1
+29 grasp yellow cup front smaller "Grasp the smaller yellow cup in front." 1/1
+30 lift orange cube middle larger "Lift the larger orange cube in the middle." 1/1
+31 rotate green cube right small "Rotate the small green cube on the right." 1/1
+32 slide blue sphere middle large "Slide the large blue sphere in the middle." 0/1
+33 rotate yellow car left smaller "Rotate the smaller yellow car on the left." 1/1
+34 grasp blue star middle larger "Grasp the larger blue star in the middle." 0/1
+35 rotate orange star behind smaller "Rotate the smaller orange star at the back." 0/1
+36 push yellow cup middle large "Push the large yellow cup in the middle." 0/1
+37 slide yellow car right small "Slide the small yellow car on the right." 1/1
+38 grasp yellow car left smaller "Grasp the smaller yellow car on the left." 0/1
+39 slide black cube front smaller "Slide the smaller black cube in front." 1/1
+40 push green cube front large "Push the large green cube in front." 1/1
+41 lift green car behind large "Lift the large green car at the back." 1/1
+42 slide yellow car middle smaller "Slide the smaller yellow car in the middle." 0/1
+43 pull green car middle larger "Pull the larger green car in the middle." 0/1
+44 grasp blue sphere front small "Grasp the small blue sphere in front." 0/1
+45 grasp red star right small "Grasp the small red star on the right." 0/1
+46 pull blue star behind smaller "Pull the smaller blue star at the back." 0/1
+47 grasp red sphere left large "Grasp the large red sphere on the left." 1/1
+48 push green cup right large "Push the large green cup on the right." 0/1
+49 push orange car front large "Push the large orange car in front." 0/1
+50 lift black cube left small "Lift the small black cube on the left." 1/1
+51 lift black star left smaller "Lift the smaller black star on the left." 1/1
+52 lift blue pyramid behind large "Lift the large blue pyramid at the back." 0/1
+53 lift black car front small "Lift the small black car in front." 1/1
+54 grasp yellow car middle small "Grasp the small yellow car in the middle." 1/1
+55 slide orange pyramid left small "Slide the small orange pyramid on the left." 1/1
+56 lift black sphere left smaller "Lift the smaller black sphere on the left." 0/1
+57 lift orange cup front larger "Lift the larger orange cup in front." 1/1
+58 push blue sphere front large "Push the large blue sphere in front." 0/1
+59 pull blue car front smaller "Pull the smaller blue car in front." 0/1
+60 push orange sphere right large "Push the large orange sphere on the right." 1/1
+61 rotate blue sphere left small "Rotate the small blue sphere on the left." 1/1
+62 grasp blue star front larger "Grasp the larger blue star in front." 0/1
+63 slide black cube right smaller "Slide the smaller black cube on the right." 1/1
+64 pull blue car middle small "Pull the small blue car in the middle." 0/1
+65 push green pyramid middle large "Push the large green pyramid in the middle." 0/1
+66 pull black pyramid left small "Pull the small black pyramid on the left." 0/1
+67 grasp blue pyramid behind large "Grasp the large blue pyramid at the back." 0/1
+68 lift orange car right smaller "Lift the smaller orange car on the right." 1/1
+69 pull green sphere behind large "Pull the large green sphere at the back." 0/1
+70 grasp yellow sphere front small "Grasp the small yellow sphere in front." 0/1
+71 slide orange pyramid front smaller "Slide the smaller orange pyramid in front." 0/1
+72 lift yellow cube middle larger "Lift the larger yellow cube in the middle." 0/1
+73 lift yellow star front larger "Lift the larger yellow star in front." 1/1
+74 push green cube front small "Push the small green cube in front." 1/1
+75 push orange sphere left small "Push the small orange sphere on the left." 1/1
+76 rotate orange pyramid behind small "Rotate the small orange pyramid at the back." 0/1
+77 slide red pyramid behind small "Slide the small red pyramid at the back." 1/1
+78 lift red star front small "Lift the small red star in front." 0/1
+79 push green car middle small "Push the small green car in the middle." 0/1
+80 grasp red cube front small "Grasp the small red cube in front." 1/1
+81 pull red cup front larger "Pull the larger red cup in front." 0/1
+82 pull red sphere middle large "Pull the large red sphere in the middle." 1/1
+83 lift green sphere front smaller "Lift the smaller green sphere in front." 0/1
+84 push yellow car left smaller "Push the smaller yellow car on the left." 0/1
+85 rotate blue cup behind small "Rotate the small blue cup at the back." 1/1
+86 lift yellow cup right smaller "Lift the smaller yellow cup on the right." 1/1
+87 pull red star left large "Pull the large red star on the left." 0/1
+88 lift red car front large "Lift the large red car in front." 0/1
+89 slide yellow sphere right larger "Slide the larger yellow sphere on the right." 1/1
+90 lift orange cup behind larger "Lift the larger orange cup at the back." 0/1
+91 lift red pyramid behind smaller "Lift the smaller red pyramid at the back." 1/1
+92 push red pyramid behind larger "Push the larger red pyramid at the back." 0/1
+93 pull blue cup behind small "Pull the small blue cup at the back." 0/1
+94 lift yellow car front smaller "Lift the smaller yellow car in front." 1/1
+95 grasp red cup front larger "Grasp the larger red cup in front." 1/1
+96 rotate red cube front smaller "Rotate the smaller red cube in front." 1/1
+97 lift black car right smaller "Lift the smaller black car on the right." 1/1
+98 lift orange cup right small "Lift the small orange cup on the right." 1/1
+99 pull green pyramid right smaller "Pull the smaller green pyramid on the right." 0/1
+100 pull blue sphere left large "Pull the large blue sphere on the left." 1/1
+101 push green star front larger "Push the larger green star in front." 0/1
+102 rotate red pyramid left larger "Rotate the larger red pyramid on the left." 0/1
+103 rotate red car right larger "Rotate the larger red car on the right." 1/1
+104 pull black pyramid right small "Pull the small black pyramid on the right." 0/1
+105 slide black cup behind smaller "Slide the smaller black cup at the back." 1/1
+106 lift orange cube left smaller "Lift the smaller orange cube on the left." 0/1
+107 grasp red car middle large "Grasp the large red car in the middle." 0/1
+108 rotate blue pyramid behind larger "Rotate the larger blue pyramid at the back." 0/1
+109 push blue sphere right larger "Push the larger blue sphere on the right." 0/1
+110 lift orange sphere behind larger "Lift the larger orange sphere at the back." 0/1
+111 pull red cup right larger "Pull the larger red cup on the right." 0/1
+112 push orange star right smaller "Push the smaller orange star on the right." 0/1
+113 grasp green sphere middle large "Grasp the large green sphere in the middle." 1/1
+114 rotate green pyramid left large "Rotate the large green pyramid on the left." 1/1
+115 grasp green star front smaller "Grasp the smaller green star in front." 0/1
+116 rotate blue pyramid right larger "Rotate the larger blue pyramid on the right." 1/1
+117 grasp orange car front large "Grasp the large orange car in front." 1/1
+118 lift orange star behind smaller "Lift the smaller orange star at the back." 0/1
+119 grasp black cup behind smaller "Grasp the smaller black cup at the back." 0/1
+120 push blue cup left large "Push the large blue cup on the left." 0/1
+121 grasp black car left larger "Grasp the larger black car on the left." 0/1
+122 rotate red star middle large "Rotate the large red star in the middle." 1/1
+123 lift blue star behind large "Lift the large blue star at the back." 0/1
+124 lift black cup behind larger "Lift the larger black cup at the back." 0/1
+125 slide black star front larger "Slide the larger black star in front." 0/1
+126 lift green star front large "Lift the large green star in front." 0/1
+127 grasp orange sphere front large "Grasp the large orange sphere in front." 0/1
+128 grasp yellow car front larger "Grasp the larger yellow car in front." 0/1
+129 push yellow cup right larger "Push the larger yellow cup on the right." 0/1
+130 slide yellow car behind large "Slide the large yellow car at the back." 1/1
+131 slide yellow cup right small "Slide the small yellow cup on the right." 0/1
+132 grasp red sphere middle smaller "Grasp the smaller red sphere in the middle." 1/1
+133 pull black car left smaller "Pull the smaller black car on the left." 0/1
+134 grasp yellow cube behind larger "Grasp the larger yellow cube at the back." 0/1
+135 push orange pyramid right larger "Push the larger orange pyramid on the right." 0/1
+136 pull green star left small "Pull the small green star on the left." 0/1
+137 pull orange car middle large "Pull the large orange car in the middle." 0/1
+138 lift blue cube right larger "Lift the larger blue cube on the right." 1/1
+139 lift black sphere behind larger "Lift the larger black sphere at the back." 1/1
+140 slide red pyramid middle small "Slide the small red pyramid in the middle." 0/1
+141 push black pyramid right larger "Push the larger black pyramid on the right." 0/1
+142 pull blue car front large "Pull the large blue car in front." 0/1
+143 rotate red cup left small "Rotate the small red cup on the left." 0/1
+144 pull green cup front larger "Pull the larger green cup in front." 1/1
+145 rotate black star middle smaller "Rotate the smaller black star in the middle." 0/1
+146 push orange sphere middle smaller "Push the smaller orange sphere in the middle." 1/1
+147 grasp green sphere middle larger "Grasp the larger green sphere in the middle." 1/1
+148 rotate yellow cup middle large "Rotate the large yellow cup in the middle." 0/1
+149 slide blue star middle smaller "Slide the smaller blue star in the middle." 0/1
+150 slide red cube behind smaller "Slide the smaller red cube at the back." 1/1
+151 slide blue cube left small "Slide the small blue cube on the left." 1/1
+152 grasp yellow cup front large "Grasp the large yellow cup in front." 1/1
+153 grasp blue cube front larger "Grasp the larger blue cube in front." 1/1
+154 grasp green cube front smaller "Grasp the smaller green cube in front." 0/1
+155 push orange star middle large "Push the large orange star in the middle." 0/1
+156 pull green cup left larger "Pull the larger green cup on the left." 1/1
+157 grasp yellow cube middle large "Grasp the large yellow cube in the middle." 1/1
+158 grasp green cube left small "Grasp the small green cube on the left." 1/1
+159 push green car left large "Push the large green car on the left." 0/1
+160 slide red cup right large "Slide the large red cup on the right." 0/1
+161 push black cup middle larger "Push the larger black cup in the middle." 0/1
+162 pull red car behind larger "Pull the larger red car at the back." 0/1
+163 slide green pyramid left larger "Slide the larger green pyramid on the left." 1/1
+164 slide orange pyramid behind small "Slide the small orange pyramid at the back." 0/1
+165 grasp black pyramid left small "Grasp the small black pyramid on the left." 1/1
+166 push yellow pyramid middle larger "Push the larger yellow pyramid in the middle." 0/1
+167 slide orange star left large "Slide the large orange star on the left." 0/1
+168 slide orange star front large "Slide the large orange star in front." 1/1
+169 grasp orange pyramid middle small "Grasp the small orange pyramid in the middle." 0/1
+170 lift red sphere behind large "Lift the large red sphere at the back." 0/1
+171 grasp red sphere behind large "Grasp the large red sphere at the back." 0/1
+172 slide black cup right smaller "Slide the smaller black cup on the right." 1/1
+173 lift green pyramid right larger "Lift the larger green pyramid on the right." 1/1
+174 slide blue cube middle small "Slide the small blue cube in the middle." 0/1
+175 push green cup front smaller "Push the smaller green cup in front." 0/1
+176 grasp blue cup behind larger "Grasp the larger blue cup at the back." 0/1
+177 slide green cup left smaller "Slide the smaller green cup on the left." 0/1
+178 grasp blue pyramid behind larger "Grasp the larger blue pyramid at the back." 0/1
+179 pull red star behind larger "Pull the larger red star at the back." 0/1
+180 pull red sphere behind larger "Pull the larger red sphere at the back." 0/1
+181 slide orange cup front larger "Slide the larger orange cup in front." 1/1
+182 pull black cup left small "Pull the small black cup on the left." 0/1
+183 lift red cube right large "Lift the large red cube on the right." 1/1
+184 push green star right larger "Push the larger green star on the right." 0/1
+185 pull yellow pyramid behind larger "Pull the larger yellow pyramid at the back." 0/1
+186 lift blue star front small "Lift the small blue star in front." 0/1
+187 rotate green cup front small "Rotate the small green cup in front." 0/1
+188 slide blue star left small "Slide the small blue star on the left." 1/1
+189 lift blue cube behind larger "Lift the larger blue cube at the back." 1/1
+190 rotate red cube right larger "Rotate the larger red cube on the right." 0/1
+191 lift yellow pyramid front small "Lift the small yellow pyramid in front." 0/1
+192 push blue cup middle smaller "Push the smaller blue cup in the middle." 0/1
+193 slide yellow star front larger "Slide the larger yellow star in front." 0/1
+194 rotate red car behind large "Rotate the large red car at the back." 0/1
+195 lift green car right small "Lift the small green car on the right." 0/1
+196 slide black cube left smaller "Slide the smaller black cube on the left." 0/1
+197 lift red sphere behind small "Lift the small red sphere at the back." 1/1
+198 lift orange car middle larger "Lift the larger orange car in the middle." 0/1
+199 push yellow sphere front smaller "Push the smaller yellow sphere in front." 0/1
+200 slide blue star right small "Slide the small blue star on the right." 0/1
+
+overall_success=72/200 (36.0%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n200/full_factor_all_factor_Lrandom_f50_n200_seed42.txt

@@ -0,0 +1,212 @@
+# Full-factor inference (GR00T N1.7) [single-process batch]
+sample_n=200  sample_seed=42  total_cells=200
+total_episodes_target=200  num_episodes_per_cell=1
+total_episodes_actual=200
+host=127.0.0.1  port=5707
+sim_backend=gpu  render_backend=gpu
+max_episode_steps=500  seed_base=42
+no_distractor_prob=0.7  replan_steps=5
+
+index verb color shape spatial size prompt successes/total
+1 pull black car right small "Pull the small black car on the right." 0/1
+2 pull red cube right larger "Pull the larger red cube on the right." 0/1
+3 lift yellow sphere middle small "Lift the small yellow sphere in the middle." 1/1
+4 push red car behind small "Push the small red car at the back." 0/1
+5 grasp red cube middle smaller "Grasp the smaller red cube in the middle." 1/1
+6 grasp black cube front large "Grasp the large black cube in front." 0/1
+7 grasp orange sphere left smaller "Grasp the smaller orange sphere on the left." 0/1
+8 grasp black car middle large "Grasp the large black car in the middle." 0/1
+9 slide orange cup middle small "Slide the small orange cup in the middle." 1/1
+10 pull red car left large "Pull the large red car on the left." 0/1
+11 grasp blue cup middle smaller "Grasp the smaller blue cup in the middle." 1/1
+12 slide green star behind smaller "Slide the smaller green star at the back." 0/1
+13 lift orange cube left large "Lift the large orange cube on the left." 0/1
+14 rotate green sphere behind smaller "Rotate the smaller green sphere at the back." 1/1
+15 rotate green star behind larger "Rotate the larger green star at the back." 0/1
+16 pull blue cube middle large "Pull the large blue cube in the middle." 0/1
+17 pull red star right small "Pull the small red star on the right." 0/1
+18 lift red star middle small "Lift the small red star in the middle." 0/1
+19 push black cube left larger "Push the larger black cube on the left." 1/1
+20 slide red car right smaller "Slide the smaller red car on the right." 0/1
+21 rotate red star right large "Rotate the large red star on the right." 0/1
+22 slide orange car right small "Slide the small orange car on the right." 0/1
+23 slide black cube behind larger "Slide the larger black cube at the back." 1/1
+24 lift black pyramid behind larger "Lift the larger black pyramid at the back." 0/1
+25 pull green cube middle small "Pull the small green cube in the middle." 0/1
+26 pull orange cube right large "Pull the large orange cube on the right." 0/1
+27 grasp yellow cup right large "Grasp the large yellow cup on the right." 0/1
+28 rotate green cube front larger "Rotate the larger green cube in front." 1/1
+29 grasp yellow cup front smaller "Grasp the smaller yellow cup in front." 0/1
+30 lift orange cube middle larger "Lift the larger orange cube in the middle." 1/1
+31 rotate green cube right small "Rotate the small green cube on the right." 1/1
+32 slide blue sphere middle large "Slide the large blue sphere in the middle." 1/1
+33 rotate yellow car left smaller "Rotate the smaller yellow car on the left." 1/1
+34 grasp blue star middle larger "Grasp the larger blue star in the middle." 0/1
+35 rotate orange star behind smaller "Rotate the smaller orange star at the back." 0/1
+36 push yellow cup middle large "Push the large yellow cup in the middle." 0/1
+37 slide yellow car right small "Slide the small yellow car on the right." 1/1
+38 grasp yellow car left smaller "Grasp the smaller yellow car on the left." 0/1
+39 slide black cube front smaller "Slide the smaller black cube in front." 1/1
+40 push green cube front large "Push the large green cube in front." 1/1
+41 lift green car behind large "Lift the large green car at the back." 0/1
+42 slide yellow car middle smaller "Slide the smaller yellow car in the middle." 1/1
+43 pull green car middle larger "Pull the larger green car in the middle." 0/1
+44 grasp blue sphere front small "Grasp the small blue sphere in front." 1/1
+45 grasp red star right small "Grasp the small red star on the right." 0/1
+46 pull blue star behind smaller "Pull the smaller blue star at the back." 0/1
+47 grasp red sphere left large "Grasp the large red sphere on the left." 0/1
+48 push green cup right large "Push the large green cup on the right." 0/1
+49 push orange car front large "Push the large orange car in front." 0/1
+50 lift black cube left small "Lift the small black cube on the left." 1/1
+51 lift black star left smaller "Lift the smaller black star on the left." 0/1
+52 lift blue pyramid behind large "Lift the large blue pyramid at the back." 0/1
+53 lift black car front small "Lift the small black car in front." 0/1
+54 grasp yellow car middle small "Grasp the small yellow car in the middle." 1/1
+55 slide orange pyramid left small "Slide the small orange pyramid on the left." 0/1
+56 lift black sphere left smaller "Lift the smaller black sphere on the left." 1/1
+57 lift orange cup front larger "Lift the larger orange cup in front." 1/1
+58 push blue sphere front large "Push the large blue sphere in front." 0/1
+59 pull blue car front smaller "Pull the smaller blue car in front." 0/1
+60 push orange sphere right large "Push the large orange sphere on the right." 0/1
+61 rotate blue sphere left small "Rotate the small blue sphere on the left." 1/1
+62 grasp blue star front larger "Grasp the larger blue star in front." 0/1
+63 slide black cube right smaller "Slide the smaller black cube on the right." 1/1
+64 pull blue car middle small "Pull the small blue car in the middle." 0/1
+65 push green pyramid middle large "Push the large green pyramid in the middle." 0/1
+66 pull black pyramid left small "Pull the small black pyramid on the left." 0/1
+67 grasp blue pyramid behind large "Grasp the large blue pyramid at the back." 0/1
+68 lift orange car right smaller "Lift the smaller orange car on the right." 1/1
+69 pull green sphere behind large "Pull the large green sphere at the back." 0/1
+70 grasp yellow sphere front small "Grasp the small yellow sphere in front." 1/1
+71 slide orange pyramid front smaller "Slide the smaller orange pyramid in front." 1/1
+72 lift yellow cube middle larger "Lift the larger yellow cube in the middle." 1/1
+73 lift yellow star front larger "Lift the larger yellow star in front." 0/1
+74 push green cube front small "Push the small green cube in front." 1/1
+75 push orange sphere left small "Push the small orange sphere on the left." 1/1
+76 rotate orange pyramid behind small "Rotate the small orange pyramid at the back." 1/1
+77 slide red pyramid behind small "Slide the small red pyramid at the back." 0/1
+78 lift red star front small "Lift the small red star in front." 0/1
+79 push green car middle small "Push the small green car in the middle." 0/1
+80 grasp red cube front small "Grasp the small red cube in front." 1/1
+81 pull red cup front larger "Pull the larger red cup in front." 1/1
+82 pull red sphere middle large "Pull the large red sphere in the middle." 1/1
+83 lift green sphere front smaller "Lift the smaller green sphere in front." 0/1
+84 push yellow car left smaller "Push the smaller yellow car on the left." 0/1
+85 rotate blue cup behind small "Rotate the small blue cup at the back." 0/1
+86 lift yellow cup right smaller "Lift the smaller yellow cup on the right." 1/1
+87 pull red star left large "Pull the large red star on the left." 0/1
+88 lift red car front large "Lift the large red car in front." 1/1
+89 slide yellow sphere right larger "Slide the larger yellow sphere on the right." 0/1
+90 lift orange cup behind larger "Lift the larger orange cup at the back." 1/1
+91 lift red pyramid behind smaller "Lift the smaller red pyramid at the back." 0/1
+92 push red pyramid behind larger "Push the larger red pyramid at the back." 1/1
+93 pull blue cup behind small "Pull the small blue cup at the back." 0/1
+94 lift yellow car front smaller "Lift the smaller yellow car in front." 0/1
+95 grasp red cup front larger "Grasp the larger red cup in front." 1/1
+96 rotate red cube front smaller "Rotate the smaller red cube in front." 1/1
+97 lift black car right smaller "Lift the smaller black car on the right." 1/1
+98 lift orange cup right small "Lift the small orange cup on the right." 0/1
+99 pull green pyramid right smaller "Pull the smaller green pyramid on the right." 0/1
+100 pull blue sphere left large "Pull the large blue sphere on the left." 1/1
+101 push green star front larger "Push the larger green star in front." 0/1
+102 rotate red pyramid left larger "Rotate the larger red pyramid on the left." 0/1
+103 rotate red car right larger "Rotate the larger red car on the right." 1/1
+104 pull black pyramid right small "Pull the small black pyramid on the right." 0/1
+105 slide black cup behind smaller "Slide the smaller black cup at the back." 1/1
+106 lift orange cube left smaller "Lift the smaller orange cube on the left." 1/1
+107 grasp red car middle large "Grasp the large red car in the middle." 0/1
+108 rotate blue pyramid behind larger "Rotate the larger blue pyramid at the back." 0/1
+109 push blue sphere right larger "Push the larger blue sphere on the right." 0/1
+110 lift orange sphere behind larger "Lift the larger orange sphere at the back." 0/1
+111 pull red cup right larger "Pull the larger red cup on the right." 0/1
+112 push orange star right smaller "Push the smaller orange star on the right." 0/1
+113 grasp green sphere middle large "Grasp the large green sphere in the middle." 1/1
+114 rotate green pyramid left large "Rotate the large green pyramid on the left." 0/1
+115 grasp green star front smaller "Grasp the smaller green star in front." 0/1
+116 rotate blue pyramid right larger "Rotate the larger blue pyramid on the right." 1/1
+117 grasp orange car front large "Grasp the large orange car in front." 1/1
+118 lift orange star behind smaller "Lift the smaller orange star at the back." 1/1
+119 grasp black cup behind smaller "Grasp the smaller black cup at the back." 0/1
+120 push blue cup left large "Push the large blue cup on the left." 1/1
+121 grasp black car left larger "Grasp the larger black car on the left." 0/1
+122 rotate red star middle large "Rotate the large red star in the middle." 1/1
+123 lift blue star behind large "Lift the large blue star at the back." 0/1
+124 lift black cup behind larger "Lift the larger black cup at the back." 0/1
+125 slide black star front larger "Slide the larger black star in front." 0/1
+126 lift green star front large "Lift the large green star in front." 1/1
+127 grasp orange sphere front large "Grasp the large orange sphere in front." 1/1
+128 grasp yellow car front larger "Grasp the larger yellow car in front." 0/1
+129 push yellow cup right larger "Push the larger yellow cup on the right." 1/1
+130 slide yellow car behind large "Slide the large yellow car at the back." 1/1
+131 slide yellow cup right small "Slide the small yellow cup on the right." 1/1
+132 grasp red sphere middle smaller "Grasp the smaller red sphere in the middle." 1/1
+133 pull black car left smaller "Pull the smaller black car on the left." 1/1
+134 grasp yellow cube behind larger "Grasp the larger yellow cube at the back." 0/1
+135 push orange pyramid right larger "Push the larger orange pyramid on the right." 0/1
+136 pull green star left small "Pull the small green star on the left." 0/1
+137 pull orange car middle large "Pull the large orange car in the middle." 0/1
+138 lift blue cube right larger "Lift the larger blue cube on the right." 1/1
+139 lift black sphere behind larger "Lift the larger black sphere at the back." 0/1
+140 slide red pyramid middle small "Slide the small red pyramid in the middle." 1/1
+141 push black pyramid right larger "Push the larger black pyramid on the right." 1/1
+142 pull blue car front large "Pull the large blue car in front." 1/1
+143 rotate red cup left small "Rotate the small red cup on the left." 1/1
+144 pull green cup front larger "Pull the larger green cup in front." 1/1
+145 rotate black star middle smaller "Rotate the smaller black star in the middle." 0/1
+146 push orange sphere middle smaller "Push the smaller orange sphere in the middle." 1/1
+147 grasp green sphere middle larger "Grasp the larger green sphere in the middle." 1/1
+148 rotate yellow cup middle large "Rotate the large yellow cup in the middle." 1/1
+149 slide blue star middle smaller "Slide the smaller blue star in the middle." 0/1
+150 slide red cube behind smaller "Slide the smaller red cube at the back." 1/1
+151 slide blue cube left small "Slide the small blue cube on the left." 1/1
+152 grasp yellow cup front large "Grasp the large yellow cup in front." 1/1
+153 grasp blue cube front larger "Grasp the larger blue cube in front." 0/1
+154 grasp green cube front smaller "Grasp the smaller green cube in front." 1/1
+155 push orange star middle large "Push the large orange star in the middle." 0/1
+156 pull green cup left larger "Pull the larger green cup on the left." 1/1
+157 grasp yellow cube middle large "Grasp the large yellow cube in the middle." 1/1
+158 grasp green cube left small "Grasp the small green cube on the left." 1/1
+159 push green car left large "Push the large green car on the left." 0/1
+160 slide red cup right large "Slide the large red cup on the right." 0/1
+161 push black cup middle larger "Push the larger black cup in the middle." 1/1
+162 pull red car behind larger "Pull the larger red car at the back." 0/1
+163 slide green pyramid left larger "Slide the larger green pyramid on the left." 1/1
+164 slide orange pyramid behind small "Slide the small orange pyramid at the back." 1/1
+165 grasp black pyramid left small "Grasp the small black pyramid on the left." 0/1
+166 push yellow pyramid middle larger "Push the larger yellow pyramid in the middle." 0/1
+167 slide orange star left large "Slide the large orange star on the left." 0/1
+168 slide orange star front large "Slide the large orange star in front." 0/1
+169 grasp orange pyramid middle small "Grasp the small orange pyramid in the middle." 0/1
+170 lift red sphere behind large "Lift the large red sphere at the back." 0/1
+171 grasp red sphere behind large "Grasp the large red sphere at the back." 0/1
+172 slide black cup right smaller "Slide the smaller black cup on the right." 0/1
+173 lift green pyramid right larger "Lift the larger green pyramid on the right." 1/1
+174 slide blue cube middle small "Slide the small blue cube in the middle." 1/1
+175 push green cup front smaller "Push the smaller green cup in front." 0/1
+176 grasp blue cup behind larger "Grasp the larger blue cup at the back." 0/1
+177 slide green cup left smaller "Slide the smaller green cup on the left." 1/1
+178 grasp blue pyramid behind larger "Grasp the larger blue pyramid at the back." 0/1
+179 pull red star behind larger "Pull the larger red star at the back." 0/1
+180 pull red sphere behind larger "Pull the larger red sphere at the back." 0/1
+181 slide orange cup front larger "Slide the larger orange cup in front." 1/1
+182 pull black cup left small "Pull the small black cup on the left." 0/1
+183 lift red cube right large "Lift the large red cube on the right." 1/1
+184 push green star right larger "Push the larger green star on the right." 0/1
+185 pull yellow pyramid behind larger "Pull the larger yellow pyramid at the back." 0/1
+186 lift blue star front small "Lift the small blue star in front." 0/1
+187 rotate green cup front small "Rotate the small green cup in front." 1/1
+188 slide blue star left small "Slide the small blue star on the left." 1/1
+189 lift blue cube behind larger "Lift the larger blue cube at the back." 1/1
+190 rotate red cube right larger "Rotate the larger red cube on the right." 1/1
+191 lift yellow pyramid front small "Lift the small yellow pyramid in front." 1/1
+192 push blue cup middle smaller "Push the smaller blue cup in the middle." 0/1
+193 slide yellow star front larger "Slide the larger yellow star in front." 0/1
+194 rotate red car behind large "Rotate the large red car at the back." 0/1
+195 lift green car right small "Lift the small green car on the right." 1/1
+196 slide black cube left smaller "Slide the smaller black cube on the left." 1/1
+197 lift red sphere behind small "Lift the small red sphere at the back." 0/1
+198 lift orange car middle larger "Lift the larger orange car in the middle." 0/1
+199 push yellow sphere front smaller "Push the smaller yellow sphere in front." 0/1
+200 slide blue star right small "Slide the small blue star on the right." 0/1
+
+overall_success=88/200 (44.0%)

results/gr00t/all_factor/all_factor_Lrandom_f50_n400/SUMMARY.txt

@@ -0,0 +1,3 @@
+# all_factor_Lrandom_f50_n400 — full-factor, TASK-aligned (sample_n=200 seed=42, 200 eps, max_steps=500, no_distractor=0.70, default difficulty); sim_backend=gpu → 任务/采样/prompt/success 与 pi0.5 一致,但数值非 1:1 可比
+seed42: 91/200 (45.5%)
+AVG over 1 seed(s): 45.5%