Joint triplet loss with semi-hard constraint for data augmentation and disease prediction using gene expression data
The joint triplet loss model with a semi-hard constraint (JTSC) is designed to represent data with a small number of samples. JTSC uses a strict selection process for semi-hard samples to construct triplets, combined with an angular loss function to achieve better prediction results in the tasks.
- Python: 3.8.5
- Numpy: 1.19.2
- Sklearn: 0.24.2
- PyTorch: 1.7.1
- Pandas: 1.1.3
For any assistance, contact [email protected]