search

facebookresearch / unbiased-teacher-v2

PyTorch code for CVPR 2022 paper Unbiased Teacher v2 Semi-supervised Object Detection for Anchor-free and Anchor-based Detectors
MIT License
89 stars 19 forks source link

About verification, the accuracy is 0 #9

Open wmjlincy opened 1 year ago

wmjlincy commented 1 year ago

My inference code is: python train_net.py --eval-only --num-gpus 1 --config configs/Faster-RCNN/coco-standard/faster_rcnn_R_50_FPN_ut2_sup10_run0.yaml SOLVER.IMG_PER_BATCH_LABEL 16 SOLVER.IMG_PER_BATCH_UNLABEL 16 MODEL.WEIGHTS ./output/model_final.pth

but my output result is below, but during training, inference is normal, Can you tell me how to fix it?

[02/23 15:23:05 d2.evaluation.evaluator]: Inference done 26990/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:33 [02/23 15:23:10 d2.evaluation.evaluator]: Inference done 27059/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:28 [02/23 15:23:15 d2.evaluation.evaluator]: Inference done 27126/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:23 [02/23 15:23:20 d2.evaluation.evaluator]: Inference done 27195/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:17 [02/23 15:23:25 d2.evaluation.evaluator]: Inference done 27265/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:12 [02/23 15:23:30 d2.evaluation.evaluator]: Inference done 27333/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:07 [02/23 15:23:35 d2.evaluation.evaluator]: Inference done 27400/27432. Dataloading: 0.0018 s/iter. Inference: 0.0732 s/iter. Eval: 0.0005 s/iter. Total: 0.0755 s/iter. ETA=0:00:02 [02/23 15:23:38 d2.evaluation.evaluator]: Total inference time: 0:34:31.245365 (0.075518 s / iter per device, on 1 devices) [02/23 15:23:38 d2.evaluation.evaluator]: Total inference pure compute time: 0:33:27 (0.073197 s / iter per device, on 1 devices) [02/23 15:23:49 d2.evaluation.coco_evaluation]: Preparing results for COCO format ... [02/23 15:23:49 d2.evaluation.coco_evaluation]: Saving results to ./output/inference/coco_instances_results.json [02/23 15:24:03 d2.evaluation.coco_evaluation]: Evaluating predictions with unofficial COCO API... Loading and preparing results... DONE (t=8.83s) creating index... index created! [02/23 15:24:12 d2.evaluation.fast_eval_api]: Evaluate annotation type bbox [02/23 15:24:33 d2.evaluation.fast_eval_api]: COCOeval_opt.evaluate() finished in 20.61 seconds. [02/23 15:24:34 d2.evaluation.fast_eval_api]: Accumulating evaluation results... [02/23 15:24:41 d2.evaluation.fast_eval_api]: COCOeval_opt.accumulate() finished in 7.13 seconds. Average Precision (AP) @[ IoU=0.50:0.95 area= all maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50 area= all maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.75 area= all maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 area= small maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 area=medium maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 area= large maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 area= all maxDets= 1 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 area= all maxDets= 10 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 area= all maxDets=100 ] = 0.001 Average Recall (AR) @[ IoU=0.50:0.95 area= small maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 area=medium maxDets=100 ] = 0.001 Average Recall (AR) @[ IoU=0.50:0.95 area= large maxDets=100 ] = 0.001 [02/23 15:24:41 d2.evaluation.coco_evaluation]: Evaluation results for bbox: AP AP50 AP75 APs APm APl
0.000 0.000 0.000 0.000 0.000 0.001
[02/23 15:24:41 d2.evaluation.coco_evaluation]: Per-category bbox AP: category AP category AP category AP
Passenger_Car 0.000 Truck_Light 0.000 Truck_Heavy 0.000
Bus_Small 0.000 Bus_Big 0.000 Van 0.000
Tricycle 0.000 Cyclist 0.000 Other 0.000

[02/23 15:24:42 d2.engine.defaults]: Evaluation results for coco_2017_val in csv format: [02/23 15:24:42 d2.evaluation.testing]: copypaste: Task: bbox [02/23 15:24:42 d2.evaluation.testing]: copypaste: AP,AP50,AP75,APs,APm,APl [02/23 15:24:42 d2.evaluation.testing]: copypaste: 0.0001,0.0003,0.0000,0.0000,0.0000,0.0007

Abdullah955 commented 1 year ago

have able to fix it ? i'm facing the same issue