search

shartoo / luna16_multi_size_3dcnn

An implement of paper "Multi-level Contextual 3D CNNs for False Positive Reduction in Pulmonary Nodule Detection"
114 stars 44 forks source link

running data_prepare.py occured ValueError #3

Open HbueSky opened 6 years ago

HbueSky commented 6 years ago

process images 1.3.6.1.4.1.14519.5.2.1.6279.6001.108197895896446896160048741492.mhd error... (<type 'exceptions.Exception'>, ':', ValueError('could not broadcast input array from shape (40,40,25) into shape (40,40,26)',)) Traceback (most recent call last): File "/home/.../luna16_multi_size_3dcnn-master/data_prepare.py", line 145, in extract_fake_cubic_from_mhd int(v_center[2] - 13):int(v_center[2] + 13)] ValueError: could not broadcast input array from shape (40,40,25) into shape (40,40,26) Traceback (most recent call last): File "/home/.../luna16_multi_size_3dcnn-master/data_prepare.py", line 145, in extract_fake_cubic_from_mhd int(v_center[2] - 13):int(v_center[2] + 13)] ValueError: could not broadcast input array from shape (40,40,0) into shape (40,40,26) process images 1.3.6.1.4.1.14519.5.2.1.6279.6001.108197895896446896160048741492.mhd error... (<type 'exceptions.Exception'>, ':', ValueError('could not broadcast input array from shape (40,40,0) into shape (40,40,26)',)) Traceback (most recent call last): File "/home/.../luna16_multi_size_3dcnn-master/data_prepare.py", line 145, in extract_fake_cubic_from_mhd int(v_center[2] - 13):int(v_center[2] + 13)] ValueError: could not broadcast input array from shape (40,40,22) into shape (40,40,26)

a56256324 commented 6 years ago

Have you solved the problem?I have a similar problem, will it affect the following procedure? Thank you~

shartoo commented 6 years ago

Sorry,it's my problem. The index is out of range,some modify should be done in the method to check range before slice.

cherry-ball commented 6 years ago

I have the same problem, and I wanna know how to solve it. Really looking forward to your answer. Thanks a lot.

qiuliwang commented 6 years ago

你好，请问出现这个错误的话会影响后面的操作吗？我也遇到了这个问题。

shartoo commented 6 years ago

@HbueSky @wangqiuli0102 @cherry-ball @a56256324 This is not a fatal error,it's fine to skip this error. It does not effect any later operation.

guyucowboy commented 6 years ago

shartoo，你好，如果修改的话，是不是先判断一下，无法组成slice的数据（导致 越界），都舍弃掉？多谢！

qishuaizheng commented 5 years ago

您好，为什么第一个尺寸的模型可以跑通，其他两个尺寸的模型跑不通呀？错误如下：“InvalidArgumentError (see above for traceback): logits and labels must be broadcastable: logits_size=[128,2] labels_size=[32,2] [[node softmax_cross_entropy_with_logits_sg (defined at H:/自己的软件/代码文献/1/luna16_multi_size_3dcnn-master/new_model.py:267)。” 万分感谢，谢谢。 #

qiuliwang commented 5 years ago

logits_size=[128,2] labels_size=[32,2]这个地方，label和你模型预测的维度不一致