Skip to main content

Troubleshooting

Common issues and solutions when using OpenTryOn.

Import Errors​

Issue: ModuleNotFoundError​

Error:

ModuleNotFoundError: No module named 'tryon'

Solution: Ensure OpenTryOn is installed:

pip install -e .

Or install dependencies:

pip install -r requirements.txt

CUDA Issues​

Issue: CUDA Out of Memory​

Error:

RuntimeError: CUDA out of memory

Solutions:

  1. Reduce batch size
  2. Reduce image resolution
  3. Use CPU instead of GPU:
import os
os.environ["CUDA_VISIBLE_DEVICES"] = ""

Issue: CUDA Not Available​

Error:

CUDA not available, using CPU

Solutions:

  1. Check CUDA installation:
import torch
print(torch.cuda.is_available())
  1. Install CUDA-compatible PyTorch:
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118

Model Checkpoint Issues​

Issue: Checkpoint Not Found​

Error:

FileNotFoundError: Checkpoint not found

Solution:

  1. Verify checkpoint path in .env file
  2. Ensure checkpoint file exists at specified path
  3. Download missing checkpoints:
# U2Net cloth segmentation
wget https://huggingface.co/levihsu/OOTDiffusion/resolve/main/cloth_segm.pth

Environment Variable Issues​

Issue: Environment Variables Not Loaded​

Error:

KeyError: 'U2NET_CLOTH_SEG_CHECKPOINT_PATH'

Solution: Always load environment variables first:

from dotenv import load_dotenv
load_dotenv()

# Then import OpenTryOn modules
from tryon.preprocessing import segment_garment

Performance Issues​

Issue: Slow Processing​

Solutions:

  1. Use GPU if available
  2. Reduce image resolution
  3. Process in batches
  4. Pre-load models for reuse
# Pre-load model
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
net = load_cloth_segm_model(device, checkpoint_path)

# Reuse model for multiple images
for image in images:
    result = extract_garment(image, net=net, device=device)

Dependency Conflicts​

Issue: Version Conflicts​

Solution: Use conda environment:

conda env create -f environment.yml
conda activate opentryon

Still Having Issues?​

  1. Check GitHub Issues
  2. Join Discord
  3. Open a new issue with:
    • Error message
    • Python version
    • OpenTryOn version
    • Steps to reproduce