search

mks0601 / A-Convolutional-Neural-Network-Cascade-for-Face-Detection

TensorFlow implementation of "A Convolutional Neural Network Cascade for Face Detection", CVPR 2015
http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Li_A_Convolutional_Neural_2015_CVPR_paper.pdf
109 stars 62 forks source link

Error in training 12-net #21

Closed mariamallaa closed 4 years ago

mariamallaa commented 4 years ago

Hi, when trying to run train_12net.py I'm getting the error: Loading positive training db... Loading negative training db... Traceback (most recent call last): File "train_12net.py", line 16, in net_12 = model.detect_12Net(input_12_node,target_12_node) File "D:\GP\A-Convolutional-Neural-Network-Cascade-for-Face-Detection-master\A-Convolutional-Neural-Network-Cascade-for-Face-Detection-master\model.py", line 45, in init self.conv2_shape = tf.concat(0,[[-1],[tf.reduce_prod(tf.slice(tf.shape(self.h_conv2),[1],[3]),0)]]) File "C:\Users\Mariam Alaa.conda\envs\opencv\lib\site-packages\tensorflow\python\util\dispatch.py", line 180, in wrapper return target(*args, **kwargs) File "C:\Users\Mariam Alaa.conda\envs\opencv\lib\site-packages\tensorflow\python\ops\array_ops.py", line 1604, in concat dtype=dtypes.int32).get_shape().assert_has_rank(0) File "C:\Users\Mariam Alaa.conda\envs\opencv\lib\site-packages\tensorflow\python\framework\tensor_shape.py", line 997, in assert_has_rank raise ValueError("Shape %s must have rank %d" % (self, rank)) ValueError: Shape (2, 1) must have rank 0

Please can you help me?

mariamallaa commented 4 years ago

Okay I solved the issue by replacing self.conv2_shape = tf.concat(0,[[-1],[tf.reduce_prod(tf.slice(tf.shape(self.h_conv2),[1],[3]),0)]]) by self.conv2_shape = tf.concat([[-1],[tf.reduce_prod(tf.slice(tf.shape(self.h_conv2),[1],[3]),0)]],0)