车牌检测模型训练结果（标注人员未标注前的数据）

Testing: [100/100][2279/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.477
Testing: [100/100][2280/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.514
Testing: [100/100][2281/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.559
Testing: [100/100][2282/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.594
Testing: [100/100][2283/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.638
Testing: [100/100][2284/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.684
Testing: [100/100][2285/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.729
Testing: [100/100][2286/2286] | Best_MAP: 0.7294 | Time: 0.052 | Data: 0.034 | Det: 0.015 | Post: 0.003 | LR: 0.00000000 | Total: 119.782
Collecting plate results (1/2)
Writing results json to ckpts/ssd/lpd/ssdlite/ssdlite_MV1-1.0-BN-PROB02-FL_512x512_1x/results/detections_LPD_val_results.json
Loading and preparing results...
DONE (t=0.02s)
creating index...
index created!
Running per image evaluation...
useSegm (deprecated) is not None. Running bbox evaluation
Evaluate annotation type *bbox*
DONE (t=1.10s).
Accumulating evaluation results...
DONE (t=0.16s).
~~~~ Mean and per-category AP @ IoU=[0.50,0.95] ~~~~
73.3
plate = 73.3
~~~~ Summary metrics ~~~~
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.733
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.922
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.868
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.495
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.660
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.772
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.815
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.833
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.833
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.571
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.753
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.862
Wrote COCO eval results to: ckpts/ssd/lpd/ssdlite/ssdlite_MV1-1.0-BN-PROB02-FL_512x512_1x/results/detection_results.pkl
INFO:root:Success saving log curve
INFO:root:Training Done!
INFO:root:Best mAP: 73.33%
guochunhe@gpunode3:~/PET/PytorchEveryThing$ python3 /home/guochunhe/PET/PytorchEveryThing/tools/ssd/train_net_multi_gpu.py --cfg cfgs/ssd/lpd/ssdlite/ssdlite_MV1-1.0-BN-PROB02-FL_512x512_1x.yaml