demo.py download model despite --ckpt passed the path

#4
by CHNtentes - opened

This is my command:

v2x@v2x-OMEN-Desktop:~/ltg/Lumina-T2X/lumina_next_t2i$ python -u demo.py --ckpt ../lumina_next_sft/ --ema

In ../lumina_next_sft directory, I already downloaded model in this repo using huggingface-cli.

However, it prints:

/home/v2x/miniconda3/envs/executorch/lib/python3.12/site-packages/diffusers/models/transformers/transformer_2d.py:34: FutureWarning: `Transformer2DModelOutput` is deprecated and will be removed in version 1.0.0. Importing `Transformer2DModelOutput` from `diffusers.models.transformer_2d` is deprecated and this will be removed in a future version. Please use `from diffusers.models.modeling_outputs import Transformer2DModelOutput`, instead.
  deprecate("Transformer2DModelOutput", "1.0.0", deprecation_message)
> initializing model parallel with size 1
> initializing ddp with size 1
> initializing pipeline with size 1
Loaded model arguments: {
  "model": "NextDiT_2B_GQA_patch2",
  "image_size": 1024,
  "vae": "sdxl",
  "precision": "bf16",
  "grad_precision": "fp32",
  "grad_clip": 2.0,
  "wd": 0.0,
  "qk_norm": true,
  "model_parallel_size": 1
}
Creating lm: Gemma-2B
/home/v2x/miniconda3/envs/executorch/lib/python3.12/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
  warnings.warn(
Downloading shards:   0%|                                                                                                       | 0/2 [00:00<?, ?it/s]
model-00001-of-00002.safetensors:   0%|                                                                                   | 0.00/4.95G [00:00<?, ?B/s]

What model is being downloaded here?

Alpha-VLLM org

Hi @CHNtentes ,
you need to download the Gemma-2b model from huggingface-cli in advance.

PommesPeter changed discussion status to closed

Sign up or log in to comment