zhangxiaopang88

many false positive boxes have scores above 0.9，Is the regression not good? can you give me any suggestions?

你好，我之前遇到楼上同样的错误，我也按照您说的方法修改了，但是上个问题解决了，又出现一下错误，您看是哪边出问题了 Traceback (most recent call last): File "test_myimg.py", line 176, in main() File "test_myimg.py", line 46, in main test(file_list[0], model_path) File "test_myimg.py", line 52, in test net.load_state_dict(torch.load(model_path)) File "/home/jan/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line...

@gjy3035

@gjy3035 我把训练的gpu改为[0,1]就解决这个问题了。我测试自己的图片需要把自己的图片生成密度图吗，现在出现以下问题，请问您遇到过吗 test_myimg.py:99: UserWarning: volatile was removed and now has no effect. Use `with torch.no_grad():` instead. img = Variable(img[None,:,:,:],volatile=True).cuda() /home/jan/anaconda3/lib/python3.6/site-packages/torch/tensor.py:287: UserWarning: non-inplace resize is deprecated warnings.warn("non-inplace resize is deprecated") Traceback...

您好，很感谢你能提供这么方便的工具，代码中我只改了路径，转换后测试结果全为0.9999，请问您知道这是什么原因吗

Hello, I also encountered this problem, which has not been solved yet. Could you send me a copy of your document, please? thank you.

> How to solved this problem？ Hello, I also encountered this problem, which has not been solved yet. Could you send me a copy of your document, please? thank you.

谢谢你，现在才知道 --dataset 这个参数是不能随意输入的，会导致tfrecord 的路径不对，就会报这个错误，还是应该把作者的代码看一下才可以

我也遇到同样的问题，类别识别为0 ，没有框，请问您解决了吗，能说说怎么解决这一类问题吗 @yangxue0827

谢谢你，我已经训练自己的数据了，可是发现会有一些漏检的问题，网上说是样本不够，现在在增加样本，我是想问题下batch_size为1还可以用其他的参数去改变batch_size吗 @yangxue0827