mei123hao

It seems that it is impossible to download the whole dataset in China!!

7

I am trying to download the whole dataset for traning. However,I have been downloading for two weeks through the .py scripts. But the conection of the networks seems very slow...

我想知道验证的时候如何获取对应的图片名称？我一直获取不到

2

for i, (image, target) in enumerate(self.val_loader): image = image.to(self.device) target = target.to(self.device) with torch.no_grad(): outputs = model(image) self.metric.update(outputs[0], target) pixAcc, mIoU = self.metric.get() logger.info("Sample: {:d}, validation pixAcc: {:.3f}, mIoU: {:.3f}".format(...

hello! when i run the command to eval my model , i got valueerror like this: Found 500 images in the folder ../datasets/citys/leftImg8bit/val 2019-11-16 13:16:18,593 semantic_segmentation INFO: Start validation, Total...

关于mobilenetv3预训练权重

3

大神你好，感谢你提供的代码。我也尝试用mobilenetv3主干网络去修改yolo，用了这个仓库 `https://github.com/d-li14/mobilenetv3.pytorch.git` 所提供的权重。但是训练测试结果貌似没有达到你的指标。只有48%的map0.5。而且训练过程发现loss没办法收敛到1一下。但是我用darknet53的主干网是可以的，请问是不是它所提供的预训练权重的问题？您训练出来的voc权重也是基于这个仓库所提供的主干网络权重去训练的吗？