Dataset Weight Reduction
Calculate absorption energies and reduce dataset weight by removing low-importance columns using ML predictions and SHAP analysis.
OC2020 Absorption Analysis
Started: 1/15/2024, 2:30:00 PM
Stage: SHAP • Status: running
Molecular Descriptors Reduction
Started: 1/15/2024, 11:00:00 AM
Stage: REDUCTION • Status: completed
Completed: 1/15/2024, 3:30:00 PM
✅ Weight Reduction Complete
Reduced Dataset Summary
Ready for Quantum ML
The dataset has been optimized by removing low-importance columns and filtering rows based on SHAP feature criteria. This compact dataset is now ready for quantum machine learning processing.
✓ Absorption energies calculated for 15,847 compounds
✓ Feature importance analyzed with SHAP (1,247 → 156 features)
✓ Low-importance columns removed (87.5% reduction)
✓ Dataset rows filtered by SHAP criteria (15,847 → 2,341 compounds)
✓ Compact dataset optimized for quantum processing
ML Absorption Energy Prediction
Train machine learning models to predict molecular absorption energies from structural features. These predictions will be used for downstream analysis and feature importance calculation.