cannot reshape array of size 0 into shape (64,128,128,1)

[anonymous530]

When I tried to run bg_nodm.py, it has something wrong:
      ValueError: cannot reshape array of size 0 into shape (64,128,128,1)
Maybe  three is something wrong : mask.astype(dtype=np.float32).reshape(
                    self.resolution, self.resolution, 1), im) 
            and
              full_output = np.concatenate( (cl2, mask.reshape(self.batch_size, 128, 128, 1)), axis=3)

[tatarchm]

Could you please provide some extra information about the issue? Did you run the script in train or test mode? Did you download the additional materials (data/snapshots/backgrounds) before running the script?

[z362194037]

@AndrewSoul do you solve the issue? I met the same question

[anonymous530]

Emm...No, I don't.. On 08/03/2017 17:37, z362194037 wrote: @AndrewSoul do you solve the issue? I get the same question — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[z362194037]

@AndrewSoul I think change the path="../data" into your own path such as "home/xxx/xxx/data" in def load_test_set(normal, bg, rad_factor=1000.0)which in tf_utils.py work.

[tatarchm]

@z362194037 Thank you for pointing this out. The data path was relative to the working directory, not the script location. I've updated the corresponding section of the code.