Classifying ImageNet classes with ResNet50 Using pytorch by setting up the pre-trained network. Start by obtaining 10 images that are similar to Imagenet classes and classifying them. Then choose 10 images from 5 different classes (2 images/class). Report the confusion matrix, the accuracy, the f-score, precision and recall of the classifier. Special Thanks to the following people for providing the images for this project:
-
Photo 0_0 by Ankith Choudhary on Unsplash
-
Photo 0_1 by Michael Milverton on Unsplash
-
lynx: Photo 1_0 by Zdeněk Macháček on Unsplash
-
lynx: Photo 1_1 by Zdeněk Macháček on Unsplash
-
mangoose: Photo 2_0 by Chandan Siddaramaia on Unsplash
-
mangoose: Photo 2_1 by Dušan veverkolog on Unsplash
-
obelisk: Photo 3_0 by Sofia Vila Flor on Unsplash
-
obelisk: Photo 3_1 by Louis Hansel on Unsplash
-
cock: Photo 4_1 by Muhammad Rosyid on Unsplash