models_training¶
- Configuration Instruction
- Image Recognition
- Image Classification
- Tricks for Training
- Catalogue
- 1. Choice of Optimizers
- 2. Choice of Learning Rate and Learning Rate Declining Strategy
- 3. Choice of Batch_size
- 4. Choice of Weight_decay
- 5. Choice of Label_smoothing
- 6. Change the Crop Area and Stretch Transformation Degree of the Images for Small Models
- 7. Use Data Augmentation to Improve Accuracy
- 8. Determine the Tuning Strategy by Train_acc and Test_acc
- 9. Improve the Accuracy of Your Own Data Set with Existing Pre-trained Models
- Reference