Table 1 Data statistics for training of images and diffraction patterns with YOLOv5.
From: Machine learning-based real-time object locator/evaluator for cryo-EM data collection
Mode | hole | xtal | diff | lowmagxtal |
|---|---|---|---|---|
Numbers of original images | 300 | 331 | 751 | 33 |
Resized dimension of images | 800 × 800 | 800 × 800 | 1024 × 1024 | 800 × 800 |
Model | YOLOv5s | YOLOv5x | YOLOv5x | YOLOv5x |
Training time (h) | ~1.9 | ~4 | ~20 | ~1 |
