Let sample.py can generate caption for gray image.

The sample.py cannot be used to generate caption for gray image.In the data_loader.py, all of the images have been converted to 'RGB' format. But when generating captions using sample.py for single image, this conversion is missed. So for gray image, it will have such RuntimeError: Given groups=1, weight of size [64, 3, 7, 7], expected input[1, 1, 224, 224] to have 3 channels, but got 1 channels instead.

tutorials/03-advanced/image_captioning/sample.py

@@ -14,7 +14,7 @@ from PIL import Image
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 
 def load_image(image_path, transform=None):
-    image = Image.open(image_path)
+    image = Image.open(image_path).convert('RGB')
     image = image.resize([224, 224], Image.LANCZOS)
     
     if transform is not None:
@@ -78,4 +78,4 @@ if __name__ == '__main__':
     parser.add_argument('--hidden_size', type=int , default=512, help='dimension of lstm hidden states')
     parser.add_argument('--num_layers', type=int , default=1, help='number of layers in lstm')
     args = parser.parse_args()
-    main(args)
+    main(args)