Predicting the type of art by using CNN and FastAI (Accuracy = 94%)

The Jupyter notebook contained in this repo predicts the type of art from the available classes (paintings, drawings, iconographies, sculptures and engravings) by being trained upon a set of 9000 images from a Kaggle dataset and using CNNs and fastai.

Links:

• The official blog post about this project is here.
• The Kaggle Notebook for this project is here.
• The dataset which has been used for this is here.