i have the follow problem.

help please

Hi,

The problem occurs due to the input format that you are using. You have to feed input in the following format - `BxCxHxW`

. You have used the channel as the last dimension(`BxHxWxC`

- this is wrong).

Thanks

I’ll see if it works

Hi,

sorry, I still have the same problem this is the part of the image rescaling:

def process_img(self, image):

i2 = np.array(image.raw_data)

i2 = i2.reshape((self.im_height, self.im_width, 4))

i2 = i2.mean(2)

i2 = i2.astype(np.float32)

i2 *= 1.0/255.0

i2 = cv2.resize(i2, (84,84))

i2 = np.reshape(i2, [1,84,84])

self.front_camera = i2

self.front_camera is the one to introduce to the convolutional neural network please help me