kyle1993 / simplest_fasterrcnn_pytorch Goto Github PK

你好，我非常感兴趣您的这个项目，觉得很简洁；但是当我在运行train.py 的时候，出现了以下的错误，您能帮忙解决一下吗？谢谢！
Traceback (most recent call last):
File "train.py", line 25, in
losses = model.train_step(img,bbox,label,scale,epoch)
File "/home/a/Bcw_data/simplest_FasterRcnn_pytorch/model/faster_rcnn_vgg16.py", line 178, in train_step
features, roi, rpn_loc_loss, rpn_cls_loss = self.train_extracor_and_rpn(img, bbox, scale, retain_graph=True)
File "/home/a/Bcw_data/simplest_FasterRcnn_pytorch/model/faster_rcnn_vgg16.py", line 209, in train_extracor_and_rpn
rpn_loc, rpn_score, roi, anchor = self.rpn(features, img_size, scale, training=True)
File "/home/a/anaconda3/envs/bcw_env/lib/python3.6/site-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/home/a/Bcw_data/simplest_FasterRcnn_pytorch/model/regin_proposal_network.py", line 195, in forward
roi = self.proposcal_creator(rpn_locs[0].cpu().detach().numpy(),rpn_scores[0].cpu().detach().numpy(),anchor, img_size,scale=scale,training=training)
File "/home/a/Bcw_data/simplest_FasterRcnn_pytorch/model/creator.py", line 450, in call
keep = np.where((hs >= min_size) & (ws >= min_size))[0]
TypeError: le() received an invalid combination of arguments - got (numpy.ndarray), but expected one of:

• (Tensor other)
  didn't match because some of the arguments have invalid types: (numpy.ndarray)
• (Number other)
  didn't match because some of the arguments have invalid types: (numpy.ndarray)

dear,when i run the test.py, has the problem as follow:

Connected to pydev debugger (build 182.4505.26)
Traceback (most recent call last):
File "/home/yjs/Downloads/pycharm/helpers/pydev/pydevd.py", line 1664, in
main()
File "/home/yjs/Downloads/pycharm/helpers/pydev/pydevd.py", line 1658, in main
globals = debugger.run(setup['file'], None, None, is_module)
File "/home/yjs/Downloads/pycharm/helpers/pydev/pydevd.py", line 1068, in run
pydev_imports.execfile(file, globals, locals) # execute the script
File "/home/yjs/Downloads/pycharm/helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "/home/yjs/Downloads/simplest_FasterRcnn_pytorch-master/test.py", line 10, in
model = FasterRCNNVGG16(opt)
File "/home/yjs/Downloads/simplest_FasterRcnn_pytorch-master/model/faster_rcnn_vgg16.py", line 102, in init
vgg16 = vgg16_bn(pretrained=True)
File "/home/yjs/.conda/envs/py36-tf13/lib/python3.6/site-packages/torchvision/models/vgg.py", line 130, in vgg16_bn
return VGG(make_layers(cfg['D'], batch_norm=True), **kwargs)
TypeError: init() got an unexpected keyword argument 'pretrained'

running on ubuntu18.04,cuda10+cudnn7.5.0

RT，我看到你的代码中时不时的要把数据从numpy上切换到cuda.tensor上或者从cuda.tensor切换到numpy上面，这一点我在chenyuntc/simple-faster-rcnn-pytorch的代码中也有发现。请问你知道有关这方面的技巧吗？让代码运行更快的技巧。就是应该在什么时候从numpy->cuda，而什么时候从cuda->numpy。或者如果你知道相关论文博客链接的话也可以。

请问如何才能改变batchsize的大小呢？