PyTorch Video Dataset Loading Implementation

RaivoKoot · RaivoKoot · commit 48004723c5c9 · 2020-11-13T05:26:07.000Z
diff --git a/README.md b/README.md
@@ -89,17 +89,17 @@ demo_dataset
 │       └───0002
 │             ├───img_00001.jpg
 │             .
-│             └───img_00017.jpg
+│             └───img_00018.jpg
 │
 └───running # arbitrary folder naming
         ├───0001  # arbitrary video folder naming
         │     ├───img_00001.jpg
         │     .
-        │     └───img_00017.jpg
+        │     └───img_00015.jpg
         └───0002
               ├───img_00001.jpg
               .
-              └───img_00017.jpg
+              └───img_00015.jpg
 
  
 ```
diff --git a/demo.py b/demo.py
@@ -59,7 +59,7 @@
         test_mode=False
     )
 
-    sample = dataset[1]
+    sample = dataset[2]
     # tensor of shape (NUM_SEGMENTS*FRAMES_PER_SEGMENT) x CHANNELS x HEIGHT x WIDTH
     frame_tensor = sample[0]
     print('Video Tensor Size:', frame_tensor.size())

Original file line number	Diff line number	Diff line change
`@@ -59,7 +59,7 @@`
`59`	`59`	`test_mode=False`
`60`	`60`	`)`
`61`	`61`
`62`		`- sample = dataset[1]`
	`62`	`+ sample = dataset[2]`
`63`	`63`	`# tensor of shape (NUM_SEGMENTS*FRAMES_PER_SEGMENT) x CHANNELS x HEIGHT x WIDTH`
`64`	`64`	`frame_tensor = sample[0]`
`65`	`65`	`print('Video Tensor Size:', frame_tensor.size())`