RudraxDave/ForestFires_Prediction

This dataset contains weather data from 2 regions in Algeria over the period of 3 months and the goal is to predict if a fire occurred at any day within that period. To create a real-world scenario, we want to predict if there will be a fire in a future date as provided by the dataset. The fire prediction is based on weather data collected from the regions. Problem type: Classification. # EE559_Project - 2-class problem - Dataset (# data pts.) - training: 184 - Class 0: 69 (37.5%) - Class 1: 115 (62.5%) - test: 60 - Sequential Backward Selection - Most contributing features: - ISI > Rain > DMC > FFMC > DC > RH > BUI > Ws > Temperature - Required reference systems - Trivial system \ `python3 trivial.py` - Test F1-score: 0.5 - Test Accuracy: 0.5 - Baseline system \ `python3 baseline.py` - Drop "Date" - Test F1-score: 0.6286 - Test Accuracy: 0.7833 - Technique 1: Perceptron Learning (Drop "Date")\ `python3 perceptron.py --M 4 --epoch 200 --plot_title perceptron` (M-fold cross-validation) - Val F1-score: 0.9113 - Val Accuracy: 0.9076 - Test F1-score: 0.8846 - Test Accuracy: 0.9 `python3 perceptron.py --M 4 --epoch 200 --normalization --plot_title p_norm` - Apply min-max normalization to all features - Val F1-score: 0.9405 - Val Accuracy: 0.9457 - Test F1-score: 0.8679 - Test Accuracy: 0.8833 `python3 perceptron.py --M 4 --epoch 200 --standardization --plot_title p_std` - Apply standardization to all features - Val F1-score: 0.9368 - Val Accuracy: 0.9457 - Test F1-score: 0.92 - Test Accuracy: 0.93 `python3 perceptron.py --M 4 --epoch 200 --standardization --feat_reduction --plot_title p_feat_reduct` - Four least contributing features: Temperature -> Ws -> BUI -> RH - Drop (1,2,3,4) features - Val F1-score: (0.9725, 0.9805, 0.9763, 0.9875) - Val Accuracy: (0.9674, 0.9728, 0.9728, 0.9837) - Test F1-score: (0.9583, 0.9787, 0.9388, 0.9583) - Test Accuracy: (0.9667, 0.9833, 0.95, 0.9667) `python3 perceptron.py --standardization --feat_reduction --extra_feat --plot_title p_add_1_feat` - Val F1-score: 0.9882 - Val Accuracy: 0.9783 - Test F1-score: 0.9787 - Test Accuracy: 0.9833 - Technique 2: KNN Classifier (Drop "Date", with Standardization)\ `python3 kNN.py --M 4 --k 7 --plot_title kNN` - The following results are for k = (2, 3, 4, 5, 6, 7, 8) - Val F1-score: (0.8287, 0.8532, 0.8605, 0.8745, 0.886, 0.8678, 0.8739) - Val Accuracy: (0.8478, 0.8641, 0 …