$ /lustre/fsw/portfolios/nvr/users/shyang/miniconda3/envs/diffusion-pipe/bin/python /lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/SCOPE/inference.py --model_dir /lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/models/SCOPE --input_image /lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/SCOPE/examples/example_0/image.png --action_path /lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/SCOPE/examples/example_0/action.parquet --prompt 'In a whimsical, toy-inspired garden, the first-person view reveals a tactical weapon aimed forward along a sunlit path. The vibrant environment includes oversized flowers, colorful building blocks, and towering structures. In the foreground, a carved wooden sign reads ROSE GARDEN, accompanied by giant puzzle pieces scattered across the ground. The surroundings are filled with fairy lights hanging from trees, creating a warm, golden glow. The garden looks like a miniature diorama, set in a charming and playful atmosphere.' --output_dir /lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/outputs/scope_fastvideo/full/runs/example_0/scope_3step --height 480 --width 832 --max_frames 81 --num_inference_steps 3 --seed 0 --fps 20 --dmd_denoising_steps 1000,757,522 --comparison_mode

/lustre/fsw/portfolios/nvr/users/shyang/miniconda3/envs/diffusion-pipe/lib/python3.12/site-packages/torch/cuda/__init__.py:61: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you.
  import pynvml  # type: ignore[import]
Loading models from: "/lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/models/SCOPE/models_t5_umt5-xxl-enc-bf16.pth"
Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file.
Loaded model: {
    "model_name": "wan_video_text_encoder",
    "model_class": "diffsynth.models.wan_video_text_encoder.WanTextEncoder",
    "extra_kwargs": null
}
Loading models from: "/lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/models/SCOPE/Wan2.2_VAE.pth"
Detected non-safetensors files, which may cause slower loading. It's recommended to convert it to a safetensors file.
Loaded model: {
    "model_name": "wan_video_vae",
    "model_class": "diffsynth.models.wan_video_vae.WanVideoVAE38",
    "extra_kwargs": null
}
Using wan_video_text_encoder from "/lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/models/SCOPE/models_t5_umt5-xxl-enc-bf16.pth".
No wan_video_dit models available. This is not an error.
Using wan_video_vae from "/lustre/fs1/portfolios/nvr/projects/nvr_elm_llm/users/shyang/FastVideo_LoRA/models/SCOPE/Wan2.2_VAE.pth".
No wan_video_image_encoder models available. This is not an error.
No wan_video_motion_controller models available. This is not an error.
No wan_video_vace models available. This is not an error.
No wan_video_vap models available. This is not an error.
No wans2v_audio_encoder models available. This is not an error.
No wan_video_animate_adapter models available. This is not an error.

VAE encoding:   0%|          | 0/1 [00:00<?, ?it/s]
VAE encoding: 100%|██████████| 1/1 [00:00<00:00,  6.99it/s]
VAE encoding: 100%|██████████| 1/1 [00:00<00:00,  6.98it/s]

  0%|          | 0/3 [00:00<?, ?it/s]
 33%|███▎      | 1/3 [00:00<00:01,  1.12it/s]
 67%|██████▋   | 2/3 [00:01<00:00,  1.14it/s]
100%|██████████| 3/3 [00:02<00:00,  1.14it/s]
100%|██████████| 3/3 [00:02<00:00,  1.14it/s]

VAE decoding:   0%|          | 0/1 [00:00<?, ?it/s]
VAE decoding: 100%|██████████| 1/1 [00:01<00:00,  1.90s/it]
VAE decoding: 100%|██████████| 1/1 [00:01<00:00,  1.90s/it]

Saving video:   0%|          | 0/81 [00:00<?, ?it/s]
Saving video:   1%|          | 1/81 [00:00<00:08,  9.01it/s]
Saving video:  85%|████████▌ | 69/81 [00:00<00:00, 384.47it/s]
Saving video: 100%|██████████| 81/81 [00:00<00:00, 285.40it/s]