This project was to determine how effective a neural network would be in classifying whether or not an image was a Monet or not. Final results was the creation of a model that could correctly pick if an image was a Monet or not roughly 85% of the time.
A full overview of interpetation of results can be found at my blog.
Images used were a selection of images from Kaggle's Painter by Numbers and C. Monet Gallery for images by other artists and by Monet, respectively. The data used in training my model can be downloaded here if one wants to replicate the process.
A requirements.txt file has been included to quickly build the needed environment. After creating an environment running pip install -r requirements.txt will install all the required packages, with the exception of opencv2. OpenCV2 was not an integral part of the project, and served only to show the color composition of the paintings under analysis.
Am currently working to supplement this analysis with a natural language processing project involving the interpretation of his works.
Feel free to reach out to me if you have any questions.