Instructions to use studyOverflow/egrpo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use studyOverflow/egrpo with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("studyOverflow/egrpo", 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
| GPU_NUM=8 # 2,4,8 | |
| MODEL_PATH="./ckpt/flux" | |
| OUTPUT_DIR="data/laion_rl_embeddings" # Updated for LAION dataset | |
| PROMPT_DIR="/data2/dataset/laion-220k/short_captions.txt" # Path to LAION captions | |
| echo "[INFO] Processing LAION-220k dataset captions..." | |
| echo "[INFO] Model path: $MODEL_PATH" | |
| echo "[INFO] Output directory: $OUTPUT_DIR" | |
| echo "[INFO] Prompt file: $PROMPT_DIR" | |
| torchrun --nproc_per_node=$GPU_NUM --master_port 19002 \ | |
| fastvideo/data_preprocess/preprocess_flux_embedding_rlpt.py \ | |
| --model_path $MODEL_PATH \ | |
| --output_dir $OUTPUT_DIR \ | |
| --prompt_dir $PROMPT_DIR |