shengshijieshao

> 在train.py的418行模型加载的权重始终是cfg.pretrained,而不是最新训练的结果，此处代码是否有问题呢? > `eval_model = Yolov4(cfg.pretrained, n_classes=cfg.classes, inference=True)` > 例如mAP始终为0的问题。 我在训练的过程中map一直是0，不知道啥问题啊

please update it!!!!!!thanks!!!!!!!!!!!!!!

python train.py -l 0.001 -g 0,1 -pretrained ./yolov4.pth -classes 12 -dir ../VisDrone2019-DET -train_label_path ./data/train.txt the 0,1 is the id of GPU,it means you can train this model by the first...

emmm还有啊，我明明设置了testnum为500，testbatchsize为300，但他却提示 The number of test sets cannot be less than the test batch size. INFO:tensorflow:Total 4750 Trains DataSets INFO:tensorflow:Total 250 Test DataSets 我都没分测试集和训练集。。。。

大佬负样本指的是什么啊？

also want a train or test demo

I had the same issue and when I removed all channels except the 'defaults' channel CondaHTTPError: HTTP 000 CONNECTION FAILED for url So now I can't connect the default repo,and...