training efficientDet-d0 over VOC0712 dataset, cpu cost is very high, always over 85%. can anybody tell me what happens ? thanks

[y78h11b09]

[midasklr]

I train efficientdet d0 on voc0712, i didnt get attention with cpu cost, have u test on voc2007? whats the mAP u got? mine is as: AP for aeroplane = 0.8431 AP for bicycle = 0.8494 AP for bird = 0.7730 AP for boat = 0.6617 AP for bottle = 0.5781 AP for bus = 0.8470 AP for car = 0.8763 AP for cat = 0.8507 AP for chair = 0.6052 AP for cow = 0.8176 AP for diningtable = 0.6236 AP for dog = 0.8137 AP for horse = 0.8629 AP for motorbike = 0.8339 AP for person = 0.7785 AP for pottedplant = 0.5108 AP for sheep = 0.8022 AP for sofa = 0.6860 AP for train = 0.8568 AP for tvmonitor = 0.7898 Mean AP = 0.7630

[quangtn266]

I train efficientdet d0 on voc0712, i didnt get attention with cpu cost, have u test on voc2007? whats the mAP u got? mine is as: AP for aeroplane = 0.8431 AP for bicycle = 0.8494 AP for bird = 0.7730 AP for boat = 0.6617 AP for bottle = 0.5781 AP for bus = 0.8470 AP for car = 0.8763 AP for cat = 0.8507 AP for chair = 0.6052 AP for cow = 0.8176 AP for diningtable = 0.6236 AP for dog = 0.8137 AP for horse = 0.8629 AP for motorbike = 0.8339 AP for person = 0.7785 AP for pottedplant = 0.5108 AP for sheep = 0.8022 AP for sofa = 0.6860 AP for train = 0.8568 AP for tvmonitor = 0.7898 Mean AP = 0.7630

How do you get your results? I tried to train efficientdet- d1, but mAp is so low. Can you share your configurations?

[majian-stu]

I run the eval.py with the pre-trained model EfficientDet-d0 on VOC2007 test dataset, but the map is only 56. Do you know why? Can you share the model you trained?