frcnn训练代码运行

pascal_voc.py中classes改为kitti中的两类

运行trainval代码的问题

Traceback (most recent call last):
  File "trainval_my.py", line 191, in <module>
    imdb.num_classes, training=True)
  File "frcnn-jwy/lib/roi_data_layer/roibatchLoader.py", line 54, in __init__
    self.ratio_list_batch[left_idx:(right_idx+1)] = target_ratio
TypeError: can't assign a numpy.int64 to a torch.FloatTensor

把target_ratio转换成tensor

self.ratio_list_batch[left_idx:(right_idx+1)] = torch.tensor(target_ratio)

2. 开始训练后，iter1有loss，之后的iterloss都变成nan
   lr太大了，改成1e-4就可以正常训练啦