Seonglae Cho실제 응용 분야에서는 대부분의 경우 테스트 분포가 학습 분포와 일치하지 않는다
iteration means The number of passes to complete one epoch.
batch size means Total number of training examples present in a single batch
Model Training Notion
- Data preprocessing
- model architecture
- Reasonable Loss Function
- 처음에는 적은 data overfitting하고 적은 iteration
- Find Learning Rate , Model Regularization 계수 최적 찾고
- if loss > 3 * original loss, quit early (learning rate is too high)
- increase data size & epoch iteration
Model Training Usages
