Can this model train with gray scale code?

[Github-Vicente]

The code is concise and worked great. But when I change the batch size to 30 in main.py. It will show File "/home/BicycleGAN/model.py", line 261, in train _, summary_str_d, D_loss_curr = self.sess.run([self.D_solver, self.d_loss_sum, self.loss_D], feed_dict={self.image_A: batch_imagesA, self.image_B: batch_imagesB, self.z: batch_z}) File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 887, in run run_metadata_ptr) File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1086, in _run str(subfeed_t.get_shape()))) ValueError: Cannot feed value of shape (1, 256, 256, 3) for Tensor u'input_images:0', which has shape '(30, 256, 256, 3)' Is there anything else that I need to revise?

Thank you very much.

[prakashpandey9]

@GamuzaLu ...You can't train with batch_size = 30 using existing code. The code uses batch_size = 1 and you are passing 30 images in a batch. See function load_batch_image in load_data.py file which returns a single pair of an image.