- In this practice problem, we are provided with a dataset of ~25k images from a wide range of natural scenes from all around the world such as a forest or a mountain. Our task is to identify which natural scene can the image be categorized into.
- There are 17034 images in training dataset and 7301 images for test data. The categories of natural scenes and their corresponding labels in the dataset are as follows - {'buildings' -> 0, 'forest' -> 1, 'glacier' -> 2, 'mountain' -> 3, 'sea' -> 4, 'street' -> 5 }
-
The Evaluation metric for this competition is accuracy.
-
Got 91% accuracy on test dataset.
-
Checkout the competition here