hizhangp / yolo_tensorflow

Tensorflow implementation of YOLO, including training and test phase.
MIT License
795 stars 442 forks source link

改变config文件中CELL_SIZE改变后程序出现错误 #92

Open pengzi2017 opened 5 years ago

pengzi2017 commented 5 years ago

我将config文件中的CELL_SIZE由7变成14后,运行程序出现了错误,显示是pascal_voc.py文件第44行两个矩阵赋值时现状不一致。 labels[count, :, :, :] = self.gt_labels[self.cursor]['label'] ValueError: could not broadcast input array from shape (7,7,7) into shape (14,14,7) 我自己检查了检查了很久,梳理了其中的逻辑关系,但就是找不出bug在哪儿。

pengzi2017 commented 5 years ago

程序中将CELL_SIZE由7变为14后,根本就没有任何关于7的定义或别的内容,定义的现状也是(14,14,5+len(classes)),但是程序运行后就会出现(7,7,5+len(classes))的现状,感觉很诡异。

XiangqianMa commented 5 years ago

I think the output cell size is decided by the size of input image and the stride of network,not only the cell size in config. py.