CHINESE-OCR icon indicating copy to clipboard operation
CHINESE-OCR copied to clipboard

请问如何用crnn训练自己的数据集,具体怎么修改参数啊?

Open Cony07f5 opened this issue 6 years ago • 5 comments


Traceback (most recent call last): File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/common_shapes.py", line 686, in _call_cpp_shape_fn_impl input_tensors_as_shapes, status) File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/errors_impl.py", line 473, in exit c_api.TF_GetCode(self.status.status)) tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimension 1 in both shapes must be equal, but are 11 and 5531. Shapes are [512,11] and [512,5531]. for 'Assign_36' (op: 'Assign') with input shapes: [512,11], [512,5531].

作者,您好!我训练自己的数据,10个数字加1一个背景类,然后就报这个错误,不知道怎么修改?求解作者.

Cony07f5 avatar Jul 31 '18 08:07 Cony07f5

你的类别弄混了,5531是原来的,11是你自己的,把所有5532换成11,应该差不多了

18813055625 avatar Aug 02 '18 09:08 18813055625

@18813055625 首先感谢您! 是的,我已经修改了,但是训练的时候loss值非常低,后面还要看看要改哪里才行.

Cony07f5 avatar Aug 03 '18 01:08 Cony07f5

增加训练步数,增加训练样本,其他我也没有什么好办法

18813055625 avatar Aug 07 '18 02:08 18813055625

训练样本和验证样本要来自同一分布

Sierkinhane avatar Sep 28 '18 02:09 Sierkinhane

@Cony07f5 你好!我也遇到同样的问题,请问在哪里修改参数(5531改成自己的数)? 非常感谢

chinesesuperliu avatar Dec 18 '18 12:12 chinesesuperliu