Lab-Challenge

This is my implementation for Udacity Pytorch Challenge Scholarship Project. It is an image classifier that classifies flower images. The used dataset is Oxford's 102 flower dataset. I used pretrained model (resnet152) and changed the classifier to a one fully connected layer with dropout. I freezed conv layers and trained the model. accuracy : 92%