Instructions to use iedavidcastilloX/CandyAI with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use iedavidcastilloX/CandyAI with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("iedavidcastilloX/CandyAI", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from iedavidcastilloX/CandyAI: direct link, hf CLI and curl.
- Browser
- Download file 1.16 kB
-
https://huggingface.co/iedavidcastilloX/CandyAI/resolve/main/README.md
- Command line
-
hf download hf://iedavidcastilloX/CandyAI/README.md
-
curl -L -o README.md https://huggingface.co/iedavidcastilloX/CandyAI/resolve/main/README.md
1.16 kB
metadata
language: en
license: mit
tags:
- candy-ai
- text-to-video
- animation
- stable-diffusion
pipeline_tag: text-to-video
model-index:
- name: CandyCrushAI
results:
- task:
type: text-to-video
name: candy animation generation
metrics:
- type: accuracy
value: 0.92
- type: fid_score
value: 18.5
- type: clip_score
value: 0.85
datasets:
- candy_crush_dataset
base_model: runwayml/stable-diffusion-v1-5
library_name: transformers
training_config:
architecture:
base_model: runwayml/stable-diffusion-v1-5
lora_r: 32
lora_alpha: 64
lora_dropout: 0.1
num_epochs: 5
gradient_accumulation_steps: 16
training_params:
batch_size: 4
learning_rate: 0.00002
warmup_steps: 100
max_grad_norm: 0.3
generation_params:
num_inference_steps: 50
guidance_scale: 7.5
frame_rate: 24
inference_config:
max_frames: 120
resolution: 512
fps: 24
optimization_level: premium
widget:
- text: Generate candy animation
example_input: colorful candy explosion with rainbow sparkles