Artist classification by painting

National Action Council for Minorities in Engineering(NACME) Google Applied Machine Learning Intensive (AMLI) at the University of Arkansas

Developed by:

• member1 - Anh Tran - Uark
• member2 - Rodrigo Mouron - Uark
• member3 - Kaveon Ware - UAPB

Description

We created a model that will predict the name of the Artist who's painting we analyze. We used the knowledge we gained from the colabs we have worked on to make our model more accurate. We used our own model which we developed.

Usage instructions

To run our project there are a couple of things needed to get started.

1. The user must download the dataset from Kaggle at: https://www.kaggle.com/competitions/painter-by-numbers/data?select=train_1.zip. This will download the training dataset. The entire library of photos is 90gb so we decided to download a portion of it. This file should be around 5gb in size.
2. Download all our code and place the file of photos in a folder with our code
3. Download all the libraries. We used the pycharm IDE to code, the modules the user must install are: pillow, tensorflow, keras, sklearn, os, and tqdm.

Once the user has completed these 3 steps, the program should run.