- Machine Learning with K-Nearest Neighbors, Decision Tree, and Random Forest Classification algorithms
- Cross-validation using pipelines with column transformers
- Dimensionality reduction with Principal Component Analysis (PCA)
- Exploratory data analysis and data visualization
The purpose of this project is to promote understanding -- my own and others' -- of fundamental data science and machine learning concepts and tools.
-
data.csv: A dataset consisting of 400 fruits with attributes related to the fruit type, subtype, weight, volume, and a set of 28,224 variables that represent RGB color values of an image of the fruit. The dataset is relatively simple and clean, but rich enough to explore classification, regression, dimensionality reduction, and other tasks. -
classification.ipynb: Jupyter notebook providing an analysis of how machine learning models can accurately classify fruits according to their type based on the image, weight and volume data, with explanations of the methods being used and links to external resources along the way. -
toy.py: a script used in classification.ipynb to show examples of certain concepts using toy data.
- Jupyer Notebook
- Python (3.10.8)
- Scikit-Learn (1.1.3)
- Pandas (1.5.2)
- NumPy (1.23.5)
- SciPy (1.9.3)
- Matplotlib (3.6.2)
- Seaborn (0.12.1)
- Bokeh (3.0.3)
- Pillow (9.3.0)
- regex (2.2.1)
This project currently consists of one notebook (classification.ipynb), but it is possible that additional notebooks will be added in the future.