The purpose of this project is to train a painting classifier that correctly labels paintings' artist, genre and style. The datatset used is public and can be found at wiki-art Github. The classifier can have multiple real-life use cases like such as forgery detection, objects retrieval in fine-art paintings, and archiving and retrieving artworks.
- Machine Learning
- Data Visualization
- Predictive Modeling
- Deep Learning
- Image recognition
- Data Mining
- Python
- Pandas
- Jupyter
- Google Cloud Storage
- Fastai Library
- Larger dataset of paintings
- data exploration/descriptive statistics
- data processing/cleaning
- web application
the complete EDA (exploratory data Analysis), model training and testing can be found by clicking on the links below, which will redirect to the jupyter notebooks created for the project.
- Feel free to contact me for information o nthe project and on how to collaborate. ji:grinning-face