Test task

Counting islands. Classical Algorithms

• Island Counter: Additional utility function to count islands in a binary matrix.

python3 count_islands.py

Regression on the tabular data. General Machine Learning

• Exploratory Data Analysis (EDA): The eda.ipynb notebook provides an overview of the data, including data cleaning, feature analysis, and visualization.
• Model Training: train.py includes preprocessing steps, model selection, training, and evaluation.
• Reproducible Code: Well-documented Python scripts that can be run from the terminal.
• Test predictions: hidden_test_preds.csv

Digit Classifier

A Python script for digit classification using CNN, Random Forest, and Logistic Regression models. Includes image preprocessing and inference functionalities.

Features

• CNN Classifier: Uses PyTorch-based CNN for digit recognition.
• Random Forest Classifier: Implements scikit-learn's RandomForestClassifier.
• Logistic Regression Classifier: Implements scikit-learn's LogisticRegression.
• Image Preprocessing: Handles resizing, normalization, and cropping. Install dependencies: To execute the MNIST script, use the following command:

python3 run_mnist.py

Install dependencies:

pip install -e requirements.txt