tensorflow model import problem

edit

OpenCV Error: Assertion failed (size == (int)dstBlob.total()) in parseTensor, file /home/zero/Program/opencv-3.4.0/modules/dnn/src/tensorflow/tf_importer.cpp, line 179 terminate called after throwing an instance of 'cv::Exception' what(): /home/zero/Program/opencv-3.4.0/modules/dnn/src/tensorflow/tf_importer.cpp:179: error: (-215) size == (int)dstBlob.total() in function parseTensor

my model definition:

with tf.variable_scope(scope, 'opencv_v1', [images, num_classes]):

image_gray = images[:,:,:,0:1]
images_input = tf.placeholder_with_default(image_gray, shape=[None, 39, 39, 1], name='InputPlaceholder')

net = slim.conv2d(images_input, 16, [4, 4], padding='VALID', scope='conv1')
end_points['conv1'] = net

net = slim.max_pool2d(net, [2, 2], 2, scope='pool1')
end_points['pool1'] = net

net = slim.conv2d(net, 32, [3, 3], padding='VALID', scope='conv2')
end_points['conv2'] = net

net = slim.max_pool2d(net, [2, 2], 2, scope='pool2')
end_points['pool2'] = net

net = slim.conv2d(net, 10, [8, 8], padding='VALID', scope='conv3')
end_points['conv2'] = net

flat1 = tf.reshape(net,[-1, 10])
net = tf.concat([flat1], 1)

output = tf.add(net, tf.zeros([num_classes]), name="Output")
end_points['Output'] = output