3.10 Cache写策略

Transferred 102/344 items from weights/yolov7_training.pt Scaled weight_decay = 0.0005 Optimizer groups: 58 .bias, 58 conv.weight, 61 other train: Scanning &#39;datasets\PlantLeaf\train\labels&#39; images and labels... 2002 found, 0 missing, 9 empty, 0 corrupted: 100%|██████████| 2002/2002 [00:03<00:00, 505.46it/s] train: New cache created: datasets\PlantLeaf\train\labels.cache QtFusion 0.5.4 Python-3.10.16 (Windows 10) val: Scanning &#39;datasets\PlantLeaf\valid\labels&#39; images and labels... 311 found, 0 missing, 1 empty, 0 corrupted: 100%|██████████| 311/311 [00:00<00:00, 447.13it/s] val: New cache created: datasets\PlantLeaf\valid\labels.cache QtFusion 0.5.4 Python-3.10.16 (Windows 10) autoanchor: Analyzing anchors... anchors/target = 4.55, Best Possible Recall (BPR) = 0.9999 Image sizes 640 train, 640 test Using 1 dataloader workers Logging results to runs\detect\train_v7_PlantLeaf19 Starting training for 9 epochs...