Used transfer learning with InceptionV3 model. I've frozen the weights of the lower layers of InceptionV3 model, concatenated them with two dense layers. Then I've trained the dense layers only.
! pip install -q kaggle
! mkdir ~/.kaggle
! cp kaggle.json ~/.kaggle/
! chmod 600 ~/.kaggle/kaggle.json
! kaggle datasets download jessicali9530/caltech256
I've used two categories (that are frog and Giraffe) from the following dataset: https://www.kaggle.com/jessicali9530/caltech256
The architecture of the neural network can be found here: https://arxiv.org/abs/1512.00567v3