Text-to-Video
Diffusers
Safetensors
English
t2v
text to video
ghibli-video
ghibli video
t2v-lora
t2v lora
t2v ghibli
t2v-ghibli
Instructions to use danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Download config.yaml from danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA: direct link, hf CLI and curl.
- Browser
- Download file 1.76 kB
-
https://huggingface.co/danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA/resolve/main/config.yaml
- Command line
-
hf download hf://danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA/config.yaml
-
curl -L -o config.yaml https://huggingface.co/danhtran2mind/zeroscope_v2_576w-Ghibli-LoRA/resolve/main/config.yaml
1.76 kB
| pretrained_model_path: ./ckpts/zeroscope_v2_576w | |
| output_dir: ./zeroscope_v2_576w-Ghibli-LoRA | |
| train_data: | |
| path: ./data/ghibli/videos | |
| width: 384 | |
| height: 384 | |
| use_bucketing: true | |
| gradient_accumulation_steps: 2 | |
| batch_size: 1 | |
| sample_start_idx: 1 | |
| fps: 16 | |
| frame_step: 1 | |
| n_sample_frames: 24 | |
| validation_data: | |
| prompt: | |
| - Anime style desert with brown-beige hills, clear sky, sparse clouds, no life, | |
| emphasizing barren desolation. | |
| - Anime style cherry blossoms, white-pink petals, yellow stamens, dark branches, | |
| blurred background, highlighting still beauty. | |
| sample_preview: false | |
| num_frames: 24 | |
| width: 384 | |
| height: 384 | |
| num_inference_steps: 15 | |
| guidance_scale: 12 | |
| spatial_scale: 0 | |
| noise_prior: 0 | |
| single_spatial_lora: true | |
| train_temporal_lora: true | |
| random_hflip_img: -1 | |
| extra_train_data: [] | |
| max_spatial_loras: 10 | |
| dataset_types: | |
| - folder | |
| validation_steps: 5000 | |
| trainable_modules: null | |
| extra_unet_params: null | |
| train_batch_size: 1 | |
| max_train_steps: 5000 | |
| learning_rate: 0.0005 | |
| lr_scheduler: constant | |
| lr_warmup_steps: 0 | |
| adam_beta1: 0.9 | |
| adam_beta2: 0.999 | |
| adam_weight_decay: 0.0001 | |
| adam_epsilon: 1.0e-08 | |
| gradient_accumulation_steps: 1 | |
| gradient_checkpointing: true | |
| text_encoder_gradient_checkpointing: true | |
| checkpointing_steps: 5000 | |
| resume_from_checkpoint: null | |
| resume_step: null | |
| mixed_precision: fp16 | |
| use_8bit_adam: true | |
| enable_xformers_memory_efficient_attention: true | |
| enable_torch_2_attn: true | |
| seed: null | |
| use_offset_noise: false | |
| rescale_schedule: false | |
| offset_noise_strength: 0.0 | |
| extend_dataset: false | |
| cache_latents: true | |
| cached_latent_dir: null | |
| use_unet_lora: true | |
| unet_lora_modules: [] | |
| text_encoder_lora_modules: [] | |
| save_pretrained_model: true | |
| lora_rank: 16 | |
| lora_path: '' | |
| lora_unet_dropout: 0.1 | |
| logger_type: tensorboard | |
| kwargs: {} | |