RuntimeError: output with shape doesn’t match the broadcast shape

1

how should i go on this type of error
NB AM USING A CPU

transform = transforms.Compose([
 transforms.Resize(256),
 transforms.CenterCrop(224),
 transforms.ToTensor()
])

transform_normalize = transforms.Normalize(
     mean=[0.485, 0.456, 0.406],
     std=[0.229, 0.224, 0.225]
 )
img = Image.open('image.jpeg')

transformed_img = transform(img)

input = transform_normalize(transformed_img)
input = input.unsqueeze(0)

image
image731×425 19.3 KB

2

I think if you use only one value in mean, one value in std, then it will start working, i.e.,

transform_normalize = transforms.Normalize(
     mean=[0.485],
     std=[0.229]
 )
3

after i use this an error is raised

transform_normalize = transforms.Normalize(
     mean=[0.485],
     std=[0.229]
 )

image
image708×393 19 KB

4

maybe change number of in_channels to 1 in nn.Conv2d, i.e.,

nn.Conv2d(1, out_channels, kernel_size)
5

how do i do that.please show

6

this has solved my problem i just grayed the image

import cv2
import numpy as np
img = cv2.imread('jim.jpeg')
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
img2 = np.zeros_like(img)
img2[:,:,0] = gray
img2[:,:,1] = gray
img2[:,:,2] = gray
cv2.imwrite('jim.jpeg', img2)