主干部分fp16,最敏感psnr=25.21,可以考虑对主干部分太敏感的部分回退fp32

This commit is contained in:
qhy
2026-02-11 16:23:21 +08:00
parent 9a08e27a19
commit 352a79035f
64 changed files with 4769 additions and 60 deletions

View File

@@ -9,7 +9,12 @@
"Bash(nvidia-smi:*)", "Bash(nvidia-smi:*)",
"Bash(conda activate unifolm-wma)", "Bash(conda activate unifolm-wma)",
"Bash(conda info:*)", "Bash(conda info:*)",
"Bash(direnv allow:*)" "Bash(direnv allow:*)",
"Bash(ls:*)",
"Bash(for scenario in unitree_g1_pack_camera unitree_z1_dual_arm_cleanup_pencils unitree_z1_dual_arm_stackbox unitree_z1_dual_arm_stackbox_v2 unitree_z1_stackbox)",
"Bash(do for case in case1 case2 case3 case4)",
"Bash(done)",
"Bash(chmod:*)"
] ]
} }
} }

114
run_all_case.sh Normal file
View File

@@ -0,0 +1,114 @@
#!/bin/bash
# 自动执行所有场景的所有case
# 总共5个场景每个场景4个case共20个case
# 设置环境变量(离线模式)
export HF_HUB_OFFLINE=1
export TRANSFORMERS_OFFLINE=1
# 颜色定义
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
BLUE='\033[0;34m'
NC='\033[0m' # No Color
# 定义所有场景
SCENARIOS=(
"unitree_g1_pack_camera"
"unitree_z1_dual_arm_cleanup_pencils"
"unitree_z1_dual_arm_stackbox"
"unitree_z1_dual_arm_stackbox_v2"
"unitree_z1_stackbox"
)
# 定义case数量
CASES=(1 2 3 4)
# 记录开始时间
START_TIME=$(date +%s)
LOG_FILE="run_all_cases_$(date +%Y%m%d_%H%M%S).log"
echo -e "${BLUE}========================================${NC}"
echo -e "${BLUE}开始执行所有场景的case${NC}"
echo -e "${BLUE}总共: ${#SCENARIOS[@]} 个场景 x ${#CASES[@]} 个case = $((${#SCENARIOS[@]} * ${#CASES[@]})) 个任务${NC}"
echo -e "${BLUE}日志文件: ${LOG_FILE}${NC}"
echo -e "${BLUE}========================================${NC}"
echo ""
# 初始化计数器
TOTAL_CASES=$((${#SCENARIOS[@]} * ${#CASES[@]}))
CURRENT_CASE=0
SUCCESS_COUNT=0
FAIL_COUNT=0
# 记录失败的case
declare -a FAILED_CASES
# 遍历所有场景
for scenario in "${SCENARIOS[@]}"; do
echo -e "${YELLOW}>>> 场景: ${scenario}${NC}"
# 遍历所有case
for case_num in "${CASES[@]}"; do
CURRENT_CASE=$((CURRENT_CASE + 1))
case_dir="${scenario}/case${case_num}"
script_path="${case_dir}/run_world_model_interaction.sh"
echo -e "${BLUE}[${CURRENT_CASE}/${TOTAL_CASES}] 执行: ${case_dir}${NC}"
# 检查脚本是否存在
if [ ! -f "${script_path}" ]; then
echo -e "${RED}错误: 脚本不存在 ${script_path}${NC}"
FAIL_COUNT=$((FAIL_COUNT + 1))
FAILED_CASES+=("${case_dir} (脚本不存在)")
continue
fi
# 执行脚本
echo "开始时间: $(date '+%Y-%m-%d %H:%M:%S')"
if bash "${script_path}" >> "${LOG_FILE}" 2>&1; then
echo -e "${GREEN}✓ 成功: ${case_dir}${NC}"
SUCCESS_COUNT=$((SUCCESS_COUNT + 1))
else
echo -e "${RED}✗ 失败: ${case_dir}${NC}"
FAIL_COUNT=$((FAIL_COUNT + 1))
FAILED_CASES+=("${case_dir}")
fi
echo "结束时间: $(date '+%Y-%m-%d %H:%M:%S')"
echo ""
done
echo ""
done
# 计算总耗时
END_TIME=$(date +%s)
DURATION=$((END_TIME - START_TIME))
HOURS=$((DURATION / 3600))
MINUTES=$(((DURATION % 3600) / 60))
SECONDS=$((DURATION % 60))
# 输出总结
echo -e "${BLUE}========================================${NC}"
echo -e "${BLUE}执行完成!${NC}"
echo -e "${BLUE}========================================${NC}"
echo -e "总任务数: ${TOTAL_CASES}"
echo -e "${GREEN}成功: ${SUCCESS_COUNT}${NC}"
echo -e "${RED}失败: ${FAIL_COUNT}${NC}"
echo -e "总耗时: ${HOURS}小时 ${MINUTES}分钟 ${SECONDS}"
echo -e "详细日志: ${LOG_FILE}"
echo ""
# 如果有失败的case列出来
if [ ${FAIL_COUNT} -gt 0 ]; then
echo -e "${RED}失败的case列表:${NC}"
for failed_case in "${FAILED_CASES[@]}"; do
echo -e "${RED} - ${failed_case}${NC}"
done
echo ""
fi
echo -e "${BLUE}========================================${NC}"

File diff suppressed because it is too large Load Diff

61
run_all_psnr.sh Executable file
View File

@@ -0,0 +1,61 @@
#!/bin/bash
set -e
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
cd "$SCRIPT_DIR"
SCENARIOS=(
unitree_g1_pack_camera
unitree_z1_dual_arm_cleanup_pencils
unitree_z1_dual_arm_stackbox
unitree_z1_dual_arm_stackbox_v2
unitree_z1_stackbox
)
CASES=(case1 case2 case3 case4)
total=0
success=0
fail=0
for scenario in "${SCENARIOS[@]}"; do
for case in "${CASES[@]}"; do
case_dir="${scenario}/${case}"
gt_video="${case_dir}/${scenario}_${case}.mp4"
pred_video=$(ls "${case_dir}"/output/inference/*_full_fs*.mp4 2>/dev/null | head -1)
output_file="${case_dir}/psnr_result.json"
total=$((total + 1))
echo "=========================================="
echo "[${total}/20] ${case_dir}"
if [ ! -f "$gt_video" ]; then
echo " SKIP: GT video not found: $gt_video"
fail=$((fail + 1))
continue
fi
if [ -z "$pred_video" ]; then
echo " SKIP: pred video not found in ${case_dir}/output/inference/"
fail=$((fail + 1))
continue
fi
echo " GT: $gt_video"
echo " Pred: $pred_video"
echo " Out: $output_file"
if python3 psnr_score_for_challenge.py \
--gt_video "$gt_video" \
--pred_video "$pred_video" \
--output_file "$output_file"; then
success=$((success + 1))
echo " DONE"
else
fail=$((fail + 1))
echo " FAILED"
fi
done
done
echo "=========================================="
echo "Finished: ${success} success, ${fail} fail, ${total} total"

View File

@@ -571,6 +571,11 @@ def run_inference(args: argparse.Namespace, gpu_num: int, gpu_no: int) -> None:
torch.save(model, prepared_path) torch.save(model, prepared_path)
print(f">>> Prepared model saved ({os.path.getsize(prepared_path) / 1024**3:.1f} GB).") print(f">>> Prepared model saved ({os.path.getsize(prepared_path) / 1024**3:.1f} GB).")
# ---- BF16: only convert the diffusion backbone, keep VAE/CLIP/embedder in FP32 ----
model.model.to(torch.bfloat16)
model.model.diffusion_model.dtype = torch.bfloat16
print(">>> Diffusion backbone (model.model) converted to BF16.")
# Build normalizer (always needed, independent of model loading path) # Build normalizer (always needed, independent of model loading path)
logging.info("***** Configing Data *****") logging.info("***** Configing Data *****")
data = instantiate_from_config(config.data) data = instantiate_from_config(config.data)

View File

@@ -0,0 +1,120 @@
2026-02-11 13:57:30.192579: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 13:57:30.242090: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 13:57:30.242134: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 13:57:30.243443: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 13:57:30.250963: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 13:57:31.177911: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:33<05:33, 33.38s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:07<05:02, 33.64s/it]
27%|██▋ | 3/11 [01:41<04:31, 33.90s/it]
36%|███▋ | 4/11 [02:15<03:58, 34.05s/it]
45%|████▌ | 5/11 [02:49<03:24, 34.07s/it]
55%|█████▍ | 6/11 [03:23<02:50, 34.08s/it]
64%|██████▎ | 7/11 [03:58<02:16, 34.10s/it]
73%|███████▎ | 8/11 [04:32<01:42, 34.12s/it]
82%|████████▏ | 9/11 [05:06<01:08, 34.12s/it]
91%|█████████ | 10/11 [05:40<00:34, 34.11s/it]
100%|██████████| 11/11 [06:14<00:00, 34.10s/it]
100%|██████████| 11/11 [06:14<00:00, 34.04s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_g1_pack_camera/case1/unitree_g1_pack_camera_case1.mp4",
"pred_video": "unitree_g1_pack_camera/case1/output/inference/0_full_fs6.mp4",
"psnr": 34.28787704598647
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_g1_pack_camera"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,120 @@
2026-02-11 14:04:18.542839: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:04:18.593447: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:04:18.593492: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:04:18.594810: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:04:18.602331: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:04:19.529518: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:41, 34.14s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:08, 34.26s/it]
27%|██▋ | 3/11 [01:42<04:33, 34.23s/it]
36%|███▋ | 4/11 [02:16<03:59, 34.23s/it]
45%|████▌ | 5/11 [02:50<03:25, 34.18s/it]
55%|█████▍ | 6/11 [03:25<02:50, 34.18s/it]
64%|██████▎ | 7/11 [03:59<02:16, 34.18s/it]
73%|███████▎ | 8/11 [04:33<01:42, 34.16s/it]
82%|████████▏ | 9/11 [05:07<01:08, 34.15s/it]
91%|█████████ | 10/11 [05:41<00:34, 34.16s/it]
100%|██████████| 11/11 [06:15<00:00, 34.13s/it]
100%|██████████| 11/11 [06:15<00:00, 34.17s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_g1_pack_camera/case2/unitree_g1_pack_camera_case2.mp4",
"pred_video": "unitree_g1_pack_camera/case2/output/inference/50_full_fs6.mp4",
"psnr": 43.756296364111726
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_g1_pack_camera"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,120 @@
2026-02-11 14:11:08.388455: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:11:08.437992: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:11:08.438037: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:11:08.439358: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:11:08.446903: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:11:09.369764: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:40, 34.10s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:08, 34.25s/it]
27%|██▋ | 3/11 [01:42<04:33, 34.25s/it]
36%|███▋ | 4/11 [02:16<03:59, 34.22s/it]
45%|████▌ | 5/11 [02:51<03:25, 34.20s/it]
55%|█████▍ | 6/11 [03:25<02:50, 34.16s/it]
64%|██████▎ | 7/11 [03:59<02:16, 34.14s/it]
73%|███████▎ | 8/11 [04:33<01:42, 34.13s/it]
82%|████████▏ | 9/11 [05:07<01:08, 34.11s/it]
91%|█████████ | 10/11 [05:41<00:34, 34.09s/it]
100%|██████████| 11/11 [06:15<00:00, 34.07s/it]
100%|██████████| 11/11 [06:15<00:00, 34.13s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_g1_pack_camera/case3/unitree_g1_pack_camera_case3.mp4",
"pred_video": "unitree_g1_pack_camera/case3/output/inference/100_full_fs6.mp4",
"psnr": 37.65161306938167
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_g1_pack_camera"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,120 @@
2026-02-11 14:17:57.092085: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:17:57.141607: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:17:57.141661: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:17:57.142984: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:17:57.150517: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:17:58.074812: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:41, 34.20s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:08, 34.29s/it]
27%|██▋ | 3/11 [01:42<04:34, 34.25s/it]
36%|███▋ | 4/11 [02:16<03:59, 34.23s/it]
45%|████▌ | 5/11 [02:51<03:25, 34.19s/it]
55%|█████▍ | 6/11 [03:25<02:50, 34.17s/it]
64%|██████▎ | 7/11 [03:59<02:16, 34.19s/it]
73%|███████▎ | 8/11 [04:33<01:42, 34.18s/it]
82%|████████▏ | 9/11 [05:07<01:08, 34.16s/it]
91%|█████████ | 10/11 [05:41<00:34, 34.13s/it]
100%|██████████| 11/11 [06:15<00:00, 34.11s/it]
100%|██████████| 11/11 [06:15<00:00, 34.17s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_g1_pack_camera/case4/unitree_g1_pack_camera_case4.mp4",
"pred_video": "unitree_g1_pack_camera/case4/output/inference/200_full_fs6.mp4",
"psnr": 33.205596596179475
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_g1_pack_camera"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -1,24 +1,13 @@
2026-02-10 15:38:28.973314: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2026-02-11 14:24:46.595601: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-10 15:38:29.023024: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2026-02-11 14:24:46.645554: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-10 15:38:29.023070: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2026-02-11 14:24:46.645598: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-10 15:38:29.024393: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2026-02-11 14:24:46.646935: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-10 15:38:29.031901: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. 2026-02-11 14:24:46.654595: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-10 15:38:29.955454: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT 2026-02-11 14:24:47.580547: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123 Global seed set to 123
INFO:mainlogger:LatentVisualDiffusion: Running in v-prediction mode >>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
INFO:unifolm_wma.models.diffusion_head.conditional_unet1d:number of parameters: 5.010531e+08 >>> Prepared model loaded.
INFO:unifolm_wma.models.diffusion_head.conditional_unet1d:number of parameters: 5.010531e+08
AE working on z of shape (1, 4, 32, 32) = 4096 dimensions.
INFO:root:Loaded ViT-H-14 model config.
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): hf-mirror.com:443
DEBUG:urllib3.connectionpool:https://hf-mirror.com:443 "HEAD /laion/CLIP-ViT-H-14-laion2B-s32B-b79K/resolve/main/open_clip_pytorch_model.bin HTTP/1.1" 302 0
INFO:root:Loading pretrained ViT-H-14 weights (laion2b_s32b_b79k).
INFO:root:Loaded ViT-H-14 model config.
DEBUG:urllib3.connectionpool:https://hf-mirror.com:443 "HEAD /laion/CLIP-ViT-H-14-laion2B-s32B-b79K/resolve/main/open_clip_pytorch_model.bin HTTP/1.1" 302 0
INFO:root:Loading pretrained ViT-H-14 weights (laion2b_s32b_b79k).
>>> model checkpoint loaded.
>>> Load pre-trained model ...
INFO:root:***** Configing Data ***** INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded. >>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded. >>> unitree_z1_stackbox: data stats loaded.
@@ -36,13 +25,16 @@ INFO:root:***** Configing Data *****
>>> unitree_g1_pack_camera: data stats loaded. >>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated. >>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ... >>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ... >>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5 DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13 DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9 DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096 DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/8 [00:00<?, ?it/s] 0%| | 0/8 [00:00<?, ?it/s]
12%|█▎ | 1/8 [00:34<03:58, 34.12s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ... >>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin DEBUG:PIL.Image:Importing BlpImagePlugin
@@ -92,9 +84,7 @@ DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin DEBUG:PIL.Image:Importing XpmImagePlugin
12%|█▎ | 1/8 [01:14<08:41, 74.51s/it]
DEBUG:PIL.Image:Importing XVThumbImagePlugin DEBUG:PIL.Image:Importing XVThumbImagePlugin
25%|██▌ | 2/8 [01:08<03:24, 34.15s/it] 25%|██▌ | 2/8 [01:08<03:24, 34.15s/it]
@@ -116,6 +106,6 @@ DEBUG:PIL.Image:Importing XVThumbImagePlugin
>>> Step 4: generating actions ... >>> Step 4: generating actions ...
>>> Step 4: interacting with world model ... >>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ... >>> Step 5: generating actions ...
>>> Step 5: interacting with world model ... >>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -1,5 +1,5 @@
{ {
"gt_video": "unitree_z1_dual_arm_cleanup_pencils/case1/unitree_z1_dual_arm_cleanup_pencils_case1.mp4", "gt_video": "unitree_z1_dual_arm_cleanup_pencils/case1/unitree_z1_dual_arm_cleanup_pencils_case1.mp4",
"pred_video": "unitree_z1_dual_arm_cleanup_pencils/case1/output/inference/0_full_fs4.mp4", "pred_video": "unitree_z1_dual_arm_cleanup_pencils/case1/output/inference/0_full_fs4.mp4",
"psnr": 47.911564449209735 "psnr": 48.52515070316814
} }

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_cleanup_pencils"
--n_iter 8 \ --n_iter 8 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,111 @@
2026-02-11 14:29:51.911195: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:29:51.961101: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:29:51.961156: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:29:51.962467: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:29:51.969980: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:29:52.899745: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/8 [00:00<?, ?it/s]
12%|█▎ | 1/8 [00:34<03:59, 34.22s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
25%|██▌ | 2/8 [01:08<03:25, 34.24s/it]
38%|███▊ | 3/8 [01:42<02:51, 34.21s/it]
50%|█████ | 4/8 [02:16<02:16, 34.18s/it]
62%|██████▎ | 5/8 [02:50<01:42, 34.15s/it]
75%|███████▌ | 6/8 [03:24<01:08, 34.11s/it]
88%|████████▊ | 7/8 [03:58<00:34, 34.10s/it]
100%|██████████| 8/8 [04:32<00:00, 34.07s/it]
100%|██████████| 8/8 [04:32<00:00, 34.12s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_cleanup_pencils/case2/unitree_z1_dual_arm_cleanup_pencils_case2.mp4",
"pred_video": "unitree_z1_dual_arm_cleanup_pencils/case2/output/inference/50_full_fs4.mp4",
"psnr": 47.91455867741451
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_cleanup_pencils"
--n_iter 8 \ --n_iter 8 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,111 @@
2026-02-11 14:34:58.016000: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:34:58.066369: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:34:58.066418: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:34:58.067763: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:34:58.075447: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:34:59.008184: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/8 [00:00<?, ?it/s]
12%|█▎ | 1/8 [00:34<03:58, 34.14s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
25%|██▌ | 2/8 [01:08<03:25, 34.21s/it]
38%|███▊ | 3/8 [01:42<02:50, 34.18s/it]
50%|█████ | 4/8 [02:16<02:16, 34.16s/it]
62%|██████▎ | 5/8 [02:50<01:42, 34.11s/it]
75%|███████▌ | 6/8 [03:24<01:08, 34.08s/it]
88%|████████▊ | 7/8 [03:58<00:34, 34.09s/it]
100%|██████████| 8/8 [04:32<00:00, 34.08s/it]
100%|██████████| 8/8 [04:32<00:00, 34.11s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_cleanup_pencils/case3/unitree_z1_dual_arm_cleanup_pencils_case3.mp4",
"pred_video": "unitree_z1_dual_arm_cleanup_pencils/case3/output/inference/100_full_fs4.mp4",
"psnr": 41.260758562627046
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_cleanup_pencils"
--n_iter 8 \ --n_iter 8 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,111 @@
2026-02-11 14:40:06.781951: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:40:06.832600: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:40:06.832649: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:40:06.833982: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:40:06.841504: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:40:07.772162: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/8 [00:00<?, ?it/s]
12%|█▎ | 1/8 [00:34<03:59, 34.18s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
25%|██▌ | 2/8 [01:08<03:25, 34.26s/it]
38%|███▊ | 3/8 [01:42<02:51, 34.23s/it]
50%|█████ | 4/8 [02:16<02:16, 34.20s/it]
62%|██████▎ | 5/8 [02:50<01:42, 34.16s/it]
75%|███████▌ | 6/8 [03:25<01:08, 34.14s/it]
88%|████████▊ | 7/8 [03:59<00:34, 34.12s/it]
100%|██████████| 8/8 [04:33<00:00, 34.12s/it]
100%|██████████| 8/8 [04:33<00:00, 34.15s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_cleanup_pencils/case4/unitree_z1_dual_arm_cleanup_pencils_case4.mp4",
"pred_video": "unitree_z1_dual_arm_cleanup_pencils/case4/output/inference/200_full_fs4.mp4",
"psnr": 47.046499351779815
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_cleanup_pencils"
--n_iter 8 \ --n_iter 8 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,108 @@
2026-02-11 14:45:16.672502: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:45:16.722666: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:45:16.722716: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:45:16.724025: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:45:16.731562: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:45:17.646917: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/7 [00:00<?, ?it/s]
14%|█▍ | 1/7 [00:34<03:24, 34.09s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
29%|██▊ | 2/7 [01:08<02:50, 34.15s/it]
43%|████▎ | 3/7 [01:42<02:16, 34.13s/it]
57%|█████▋ | 4/7 [02:16<01:42, 34.09s/it]
71%|███████▏ | 5/7 [02:50<01:08, 34.09s/it]
86%|████████▌ | 6/7 [03:24<00:34, 34.09s/it]
100%|██████████| 7/7 [03:58<00:00, 34.08s/it]
100%|██████████| 7/7 [03:58<00:00, 34.09s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox/case1/unitree_z1_dual_arm_stackbox_case1.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox/case1/output/inference/5_full_fs4.mp4",
"psnr": 43.97044934749157
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox"
--n_iter 7 \ --n_iter 7 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,108 @@
2026-02-11 14:49:54.901507: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:49:54.951975: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:49:54.952023: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:49:54.953338: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:49:54.960938: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:49:55.888131: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/7 [00:00<?, ?it/s]
14%|█▍ | 1/7 [00:34<03:25, 34.18s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
29%|██▊ | 2/7 [01:08<02:51, 34.26s/it]
43%|████▎ | 3/7 [01:42<02:16, 34.25s/it]
57%|█████▋ | 4/7 [02:16<01:42, 34.22s/it]
71%|███████▏ | 5/7 [02:50<01:08, 34.17s/it]
86%|████████▌ | 6/7 [03:25<00:34, 34.13s/it]
100%|██████████| 7/7 [03:59<00:00, 34.11s/it]
100%|██████████| 7/7 [03:59<00:00, 34.16s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox/case2/unitree_z1_dual_arm_stackbox_case2.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox/case2/output/inference/15_full_fs4.mp4",
"psnr": 43.70773432165555
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox"
--n_iter 7 \ --n_iter 7 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,108 @@
2026-02-11 14:54:33.079012: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:54:33.128851: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:54:33.128900: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:54:33.130229: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:54:33.137786: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:54:34.065218: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/7 [00:00<?, ?it/s]
14%|█▍ | 1/7 [00:34<03:25, 34.18s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
29%|██▊ | 2/7 [01:08<02:51, 34.24s/it]
43%|████▎ | 3/7 [01:42<02:16, 34.25s/it]
57%|█████▋ | 4/7 [02:16<01:42, 34.17s/it]
71%|███████▏ | 5/7 [02:50<01:08, 34.13s/it]
86%|████████▌ | 6/7 [03:24<00:34, 34.11s/it]
100%|██████████| 7/7 [03:58<00:00, 34.09s/it]
100%|██████████| 7/7 [03:58<00:00, 34.13s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox/case3/unitree_z1_dual_arm_stackbox_case3.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox/case3/output/inference/25_full_fs4.mp4",
"psnr": 48.68206289825236
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox"
--n_iter 7 \ --n_iter 7 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,108 @@
2026-02-11 14:59:11.849320: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 14:59:11.899274: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 14:59:11.899322: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 14:59:11.900640: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 14:59:11.908158: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 14:59:12.830387: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/7 [00:00<?, ?it/s]
14%|█▍ | 1/7 [00:34<03:24, 34.11s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
29%|██▊ | 2/7 [01:08<02:50, 34.17s/it]
43%|████▎ | 3/7 [01:42<02:16, 34.13s/it]
57%|█████▋ | 4/7 [02:16<01:42, 34.12s/it]
71%|███████▏ | 5/7 [02:50<01:08, 34.07s/it]
86%|████████▌ | 6/7 [03:24<00:34, 34.07s/it]
100%|██████████| 7/7 [03:58<00:00, 34.06s/it]
100%|██████████| 7/7 [03:58<00:00, 34.09s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox/case4/unitree_z1_dual_arm_stackbox_case4.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox/case4/output/inference/35_full_fs4.mp4",
"psnr": 42.117165235043196
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox"
--n_iter 7 \ --n_iter 7 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -1,10 +1,10 @@
2026-02-11 11:59:27.241485: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2026-02-11 15:03:49.644187: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 11:59:27.291755: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2026-02-11 15:03:49.694117: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 11:59:27.291807: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2026-02-11 15:03:49.694162: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 11:59:27.293169: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2026-02-11 15:03:49.695456: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 11:59:27.300838: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. 2026-02-11 15:03:49.702946: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 11:59:28.228009: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT 2026-02-11 15:03:50.638334: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123 Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ... >>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded. >>> Prepared model loaded.
@@ -31,7 +31,7 @@ DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13 DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9 DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096 DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s] 0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:41, 34.19s/it]>>> Step 0: generating actions ... 9%|▉ | 1/11 [00:34<05:41, 34.19s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ... >>> Step 0: interacting with world model ...
@@ -84,7 +84,7 @@ DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:08, 34.26s/it] 18%|█▊ | 2/11 [01:08<05:08, 34.26s/it]
@@ -115,6 +115,6 @@ DEBUG:PIL.Image:Importing XVThumbImagePlugin
>>> Step 6: generating actions ... >>> Step 6: generating actions ...
>>> Step 6: interacting with world model ... >>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ... >>> Step 7: generating actions ...
>>> Step 7: interacting with world model ... >>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>> >>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -1,5 +1,5 @@
{ {
"gt_video": "/home/qhy/unifolm-world-model-action/unitree_z1_dual_arm_stackbox_v2/case1/unitree_z1_dual_arm_stackbox_v2_case1.mp4", "gt_video": "unitree_z1_dual_arm_stackbox_v2/case1/unitree_z1_dual_arm_stackbox_v2_case1.mp4",
"pred_video": "/home/qhy/unifolm-world-model-action/unitree_z1_dual_arm_stackbox_v2/case1/output/inference/5_full_fs4.mp4", "pred_video": "unitree_z1_dual_arm_stackbox_v2/case1/output/inference/5_full_fs4.mp4",
"psnr": 28.167025381705358 "psnr": 28.167025381705358
} }

View File

@@ -0,0 +1,120 @@
2026-02-11 15:10:45.687888: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:10:45.738006: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:10:45.738054: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:10:45.739410: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:10:45.747229: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:10:46.687896: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:41, 34.16s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:08, 34.26s/it]
27%|██▋ | 3/11 [01:42<04:33, 34.20s/it]
36%|███▋ | 4/11 [02:16<03:59, 34.16s/it]
45%|████▌ | 5/11 [02:50<03:24, 34.15s/it]
55%|█████▍ | 6/11 [03:24<02:50, 34.13s/it]
64%|██████▎ | 7/11 [03:59<02:16, 34.13s/it]
73%|███████▎ | 8/11 [04:33<01:42, 34.13s/it]
82%|████████▏ | 9/11 [05:07<01:08, 34.12s/it]
91%|█████████ | 10/11 [05:41<00:34, 34.11s/it]
100%|██████████| 11/11 [06:15<00:00, 34.09s/it]
100%|██████████| 11/11 [06:15<00:00, 34.13s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox_v2/case2/unitree_z1_dual_arm_stackbox_v2_case2.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox_v2/case2/output/inference/15_full_fs4.mp4",
"psnr": 33.945563782754554
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox_v2"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,120 @@
2026-02-11 15:17:41.661323: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:17:41.711317: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:17:41.711373: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:17:41.712706: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:17:41.720248: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:17:42.650151: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:41, 34.15s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:08, 34.25s/it]
27%|██▋ | 3/11 [01:42<04:33, 34.23s/it]
36%|███▋ | 4/11 [02:16<03:59, 34.18s/it]
45%|████▌ | 5/11 [02:50<03:24, 34.17s/it]
55%|█████▍ | 6/11 [03:25<02:50, 34.15s/it]
64%|██████▎ | 7/11 [03:59<02:16, 34.13s/it]
73%|███████▎ | 8/11 [04:33<01:42, 34.12s/it]
82%|████████▏ | 9/11 [05:07<01:08, 34.09s/it]
91%|█████████ | 10/11 [05:41<00:34, 34.09s/it]
100%|██████████| 11/11 [06:15<00:00, 34.09s/it]
100%|██████████| 11/11 [06:15<00:00, 34.13s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox_v2/case3/unitree_z1_dual_arm_stackbox_v2_case3.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox_v2/case3/output/inference/25_full_fs4.mp4",
"psnr": 31.86126241517472
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox_v2"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,120 @@
2026-02-11 15:24:38.487806: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:24:38.538144: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:24:38.538200: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:24:38.539554: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:24:38.547185: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:24:39.470885: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/11 [00:00<?, ?it/s]
9%|▉ | 1/11 [00:34<05:41, 34.18s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
18%|█▊ | 2/11 [01:08<05:09, 34.35s/it]
27%|██▋ | 3/11 [01:42<04:34, 34.33s/it]
36%|███▋ | 4/11 [02:17<03:59, 34.26s/it]
45%|████▌ | 5/11 [02:51<03:25, 34.22s/it]
55%|█████▍ | 6/11 [03:25<02:50, 34.18s/it]
64%|██████▎ | 7/11 [03:59<02:16, 34.14s/it]
73%|███████▎ | 8/11 [04:33<01:42, 34.13s/it]
82%|████████▏ | 9/11 [05:07<01:08, 34.09s/it]
91%|█████████ | 10/11 [05:41<00:34, 34.09s/it]
100%|██████████| 11/11 [06:15<00:00, 34.09s/it]
100%|██████████| 11/11 [06:15<00:00, 34.15s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_dual_arm_stackbox_v2/case4/unitree_z1_dual_arm_stackbox_v2_case4.mp4",
"pred_video": "unitree_z1_dual_arm_stackbox_v2/case4/output/inference/35_full_fs4.mp4",
"psnr": 39.90908062249536
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_dual_arm_stackbox_v2"
--n_iter 11 \ --n_iter 11 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,123 @@
2026-02-11 15:31:35.657972: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:31:35.707733: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:31:35.707792: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:31:35.709109: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:31:35.716616: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:31:36.648540: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/12 [00:00<?, ?it/s]
8%|▊ | 1/12 [00:34<06:15, 34.17s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
17%|█▋ | 2/12 [01:08<05:42, 34.24s/it]
25%|██▌ | 3/12 [01:42<05:07, 34.20s/it]
33%|███▎ | 4/12 [02:16<04:33, 34.19s/it]
42%|████▏ | 5/12 [02:50<03:58, 34.14s/it]
50%|█████ | 6/12 [03:24<03:24, 34.09s/it]
58%|█████▊ | 7/12 [03:58<02:50, 34.07s/it]
67%|██████▋ | 8/12 [04:32<02:16, 34.07s/it]
75%|███████▌ | 9/12 [05:06<01:42, 34.05s/it]
83%|████████▎ | 10/12 [05:41<01:08, 34.06s/it]
92%|█████████▏| 11/12 [06:15<00:34, 34.08s/it]
100%|██████████| 12/12 [06:49<00:00, 34.07s/it]
100%|██████████| 12/12 [06:49<00:00, 34.10s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 8: generating actions ...
>>> Step 8: interacting with world model ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_stackbox/case1/unitree_z1_stackbox_case1.mp4",
"pred_video": "unitree_z1_stackbox/case1/output/inference/5_full_fs4.mp4",
"psnr": 49.42336701518203
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_stackbox"
--n_iter 12 \ --n_iter 12 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,123 @@
2026-02-11 15:39:01.409308: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:39:01.459136: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:39:01.459190: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:39:01.460507: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:39:01.468019: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:39:02.395912: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/12 [00:00<?, ?it/s]
8%|▊ | 1/12 [00:34<06:15, 34.10s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
17%|█▋ | 2/12 [01:08<05:41, 34.18s/it]
25%|██▌ | 3/12 [01:42<05:07, 34.15s/it]
33%|███▎ | 4/12 [02:16<04:33, 34.13s/it]
42%|████▏ | 5/12 [02:50<03:58, 34.09s/it]
50%|█████ | 6/12 [03:24<03:24, 34.08s/it]
58%|█████▊ | 7/12 [03:58<02:50, 34.06s/it]
67%|██████▋ | 8/12 [04:32<02:16, 34.03s/it]
75%|███████▌ | 9/12 [05:06<01:42, 34.03s/it]
83%|████████▎ | 10/12 [05:40<01:08, 34.03s/it]
92%|█████████▏| 11/12 [06:14<00:34, 34.02s/it]
100%|██████████| 12/12 [06:48<00:00, 34.00s/it]
100%|██████████| 12/12 [06:48<00:00, 34.05s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 8: generating actions ...
>>> Step 8: interacting with world model ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_stackbox/case2/unitree_z1_stackbox_case2.mp4",
"pred_video": "unitree_z1_stackbox/case2/output/inference/15_full_fs4.mp4",
"psnr": 48.88265200549669
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_stackbox"
--n_iter 12 \ --n_iter 12 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,123 @@
2026-02-11 15:46:27.304090: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:46:27.354074: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:46:27.354120: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:46:27.355468: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:46:27.363130: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:46:28.290783: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/12 [00:00<?, ?it/s]
8%|▊ | 1/12 [00:34<06:15, 34.15s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
17%|█▋ | 2/12 [01:08<05:42, 34.24s/it]
25%|██▌ | 3/12 [01:42<05:07, 34.22s/it]
33%|███▎ | 4/12 [02:16<04:33, 34.16s/it]
42%|████▏ | 5/12 [02:50<03:58, 34.11s/it]
50%|█████ | 6/12 [03:24<03:24, 34.08s/it]
58%|█████▊ | 7/12 [03:58<02:50, 34.04s/it]
67%|██████▋ | 8/12 [04:32<02:16, 34.02s/it]
75%|███████▌ | 9/12 [05:06<01:41, 33.99s/it]
83%|████████▎ | 10/12 [05:40<01:07, 33.99s/it]
92%|█████████▏| 11/12 [06:14<00:33, 33.97s/it]
100%|██████████| 12/12 [06:48<00:00, 33.96s/it]
100%|██████████| 12/12 [06:48<00:00, 34.04s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 8: generating actions ...
>>> Step 8: interacting with world model ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_stackbox/case3/unitree_z1_stackbox_case3.mp4",
"pred_video": "unitree_z1_stackbox/case3/output/inference/25_full_fs4.mp4",
"psnr": 50.884297816906816
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_stackbox"
--n_iter 12 \ --n_iter 12 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"

View File

@@ -0,0 +1,123 @@
2026-02-11 15:53:52.504337: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2026-02-11 15:53:52.554351: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-02-11 15:53:52.554397: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-02-11 15:53:52.555718: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-02-11 15:53:52.563252: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-02-11 15:53:53.493343: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Global seed set to 123
>>> Loading prepared model from ckpts/unifolm_wma_dual.ckpt.prepared.pt ...
>>> Prepared model loaded.
INFO:root:***** Configing Data *****
>>> unitree_z1_stackbox: 1 data samples loaded.
>>> unitree_z1_stackbox: data stats loaded.
>>> unitree_z1_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox: data stats loaded.
>>> unitree_z1_dual_arm_stackbox: normalizer initiated.
>>> unitree_z1_dual_arm_stackbox_v2: 1 data samples loaded.
>>> unitree_z1_dual_arm_stackbox_v2: data stats loaded.
>>> unitree_z1_dual_arm_stackbox_v2: normalizer initiated.
>>> unitree_z1_dual_arm_cleanup_pencils: 1 data samples loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: data stats loaded.
>>> unitree_z1_dual_arm_cleanup_pencils: normalizer initiated.
>>> unitree_g1_pack_camera: 1 data samples loaded.
>>> unitree_g1_pack_camera: data stats loaded.
>>> unitree_g1_pack_camera: normalizer initiated.
>>> Dataset is successfully loaded ...
✓ KV fused: 66 attention layers
>>> Generate 16 frames under each generation ...
DEBUG:h5py._conv:Creating converter from 3 to 5
DEBUG:PIL.PngImagePlugin:STREAM b'IHDR' 16 13
DEBUG:PIL.PngImagePlugin:STREAM b'pHYs' 41 9
DEBUG:PIL.PngImagePlugin:STREAM b'IDAT' 62 4096
0%| | 0/12 [00:00<?, ?it/s]
8%|▊ | 1/12 [00:34<06:15, 34.14s/it]>>> Step 0: generating actions ...
>>> Step 0: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 1: generating actions ...
DEBUG:PIL.Image:Importing BlpImagePlugin
DEBUG:PIL.Image:Importing BmpImagePlugin
DEBUG:PIL.Image:Importing BufrStubImagePlugin
DEBUG:PIL.Image:Importing CurImagePlugin
DEBUG:PIL.Image:Importing DcxImagePlugin
DEBUG:PIL.Image:Importing DdsImagePlugin
DEBUG:PIL.Image:Importing EpsImagePlugin
DEBUG:PIL.Image:Importing FitsImagePlugin
DEBUG:PIL.Image:Importing FitsStubImagePlugin
DEBUG:PIL.Image:Importing FliImagePlugin
DEBUG:PIL.Image:Importing FpxImagePlugin
DEBUG:PIL.Image:Image: failed to import FpxImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing FtexImagePlugin
DEBUG:PIL.Image:Importing GbrImagePlugin
DEBUG:PIL.Image:Importing GifImagePlugin
DEBUG:PIL.Image:Importing GribStubImagePlugin
DEBUG:PIL.Image:Importing Hdf5StubImagePlugin
DEBUG:PIL.Image:Importing IcnsImagePlugin
DEBUG:PIL.Image:Importing IcoImagePlugin
DEBUG:PIL.Image:Importing ImImagePlugin
DEBUG:PIL.Image:Importing ImtImagePlugin
DEBUG:PIL.Image:Importing IptcImagePlugin
DEBUG:PIL.Image:Importing JpegImagePlugin
DEBUG:PIL.Image:Importing Jpeg2KImagePlugin
DEBUG:PIL.Image:Importing McIdasImagePlugin
DEBUG:PIL.Image:Importing MicImagePlugin
DEBUG:PIL.Image:Image: failed to import MicImagePlugin: No module named 'olefile'
DEBUG:PIL.Image:Importing MpegImagePlugin
DEBUG:PIL.Image:Importing MpoImagePlugin
DEBUG:PIL.Image:Importing MspImagePlugin
DEBUG:PIL.Image:Importing PalmImagePlugin
DEBUG:PIL.Image:Importing PcdImagePlugin
DEBUG:PIL.Image:Importing PcxImagePlugin
DEBUG:PIL.Image:Importing PdfImagePlugin
DEBUG:PIL.Image:Importing PixarImagePlugin
DEBUG:PIL.Image:Importing PngImagePlugin
DEBUG:PIL.Image:Importing PpmImagePlugin
DEBUG:PIL.Image:Importing PsdImagePlugin
DEBUG:PIL.Image:Importing QoiImagePlugin
DEBUG:PIL.Image:Importing SgiImagePlugin
DEBUG:PIL.Image:Importing SpiderImagePlugin
DEBUG:PIL.Image:Importing SunImagePlugin
DEBUG:PIL.Image:Importing TgaImagePlugin
DEBUG:PIL.Image:Importing TiffImagePlugin
DEBUG:PIL.Image:Importing WebPImagePlugin
DEBUG:PIL.Image:Importing WmfImagePlugin
DEBUG:PIL.Image:Importing XbmImagePlugin
DEBUG:PIL.Image:Importing XpmImagePlugin
DEBUG:PIL.Image:Importing XVThumbImagePlugin
17%|█▋ | 2/12 [01:08<05:41, 34.19s/it]
25%|██▌ | 3/12 [01:42<05:07, 34.15s/it]
33%|███▎ | 4/12 [02:16<04:32, 34.12s/it]
42%|████▏ | 5/12 [02:50<03:58, 34.09s/it]
50%|█████ | 6/12 [03:24<03:24, 34.05s/it]
58%|█████▊ | 7/12 [03:58<02:50, 34.03s/it]
67%|██████▋ | 8/12 [04:32<02:16, 34.04s/it]
75%|███████▌ | 9/12 [05:06<01:42, 34.03s/it]
83%|████████▎ | 10/12 [05:40<01:08, 34.01s/it]
92%|█████████▏| 11/12 [06:14<00:34, 34.03s/it]
100%|██████████| 12/12 [06:48<00:00, 34.02s/it]
100%|██████████| 12/12 [06:48<00:00, 34.05s/it]
>>> Step 1: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 2: generating actions ...
>>> Step 2: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 3: generating actions ...
>>> Step 3: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 4: generating actions ...
>>> Step 4: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 5: generating actions ...
>>> Step 5: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 6: generating actions ...
>>> Step 6: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 7: generating actions ...
>>> Step 7: interacting with world model ...
>>>>>>>>>>>>>>>>>>>>>>>>
>>> Step 8: generating actions ...
>>> Step 8: interacting with world model ...

View File

@@ -0,0 +1,5 @@
{
"gt_video": "unitree_z1_stackbox/case4/unitree_z1_stackbox_case4.mp4",
"pred_video": "unitree_z1_stackbox/case4/output/inference/35_full_fs4.mp4",
"psnr": 47.85197517791449
}

View File

@@ -20,5 +20,6 @@ dataset="unitree_z1_stackbox"
--n_iter 12 \ --n_iter 12 \
--timestep_spacing 'uniform_trailing' \ --timestep_spacing 'uniform_trailing' \
--guidance_rescale 0.7 \ --guidance_rescale 0.7 \
--perframe_ae --perframe_ae \
--fast_policy_no_decode
} 2>&1 | tee "${res_dir}/output.log" } 2>&1 | tee "${res_dir}/output.log"