Spaces:
Running
Running
A newer version of the Streamlit SDK is available:
1.51.0
β‘ CleanSpeak Quick Start Guide
π Get Started in 3 Steps
1. Setup (One-time)
# Option A: Use setup script
./setup.sh
# Option B: Manual setup
python3 -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
2. Run the App
streamlit run app.py
3. Open in Browser
Your browser should open automatically to http://localhost:8501
If not, manually navigate to that URL.
π― How to Use
- Type a comment in the text box
- Click "π Detect Toxicity"
- See the result:
- β No = Safe comment
- π¨ Yes = Toxic comment detected
π Key Files
app.py- Main applicationrequirements.txt- DependenciesREADME.md- Full documentationDEMO.md- Demo examplessetup.sh- Automated setup script
π Troubleshooting
Issue: "No module named 'streamlit'"
Solution: Activate your virtual environment
source venv/bin/activate
pip install -r requirements.txt
Issue: Model download is slow
Solution: Model downloads automatically on first run (~500MB)
- First run will take a few minutes
- Subsequent runs are instant
Issue: "Port 8501 already in use"
Solution: Kill the existing process or use a different port
# Kill existing process
pkill -f streamlit
# Or use different port
streamlit run app.py --server.port 8502
π‘ Quick Tips
- First run: Model downloads from Hugging Face (~2-3 minutes)
- Subsequent runs: Instant startup
- No training needed: Uses pre-trained model
- Yes/No output: Simple binary classification
- Detailed breakdown: See individual toxicity types
π Need Help?
- Check
README.mdfor detailed documentation - See
DEMO.mdfor example inputs/outputs - Open an issue on GitHub
Happy detecting! π¬β¨