COVID-19 Case Forecasting Model

from pathlib import Path
import numpy as np
import pandas as pd
from joblib import load
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_absolute_error, mean_squared_error

BASE_DIR = Path(__file__).resolve().parent

TARGET_COLUMN = "Case_Count"
SCALED_COLUMNS = ["Case_Count", "Monitoring_Week", "Monitoring_Month", "Monitoring_Year"]
FUTURE_WEEKS = 12

# Read the cleaned/scaled data and scaler file.
data = pd.read_excel(BASE_DIR / "Covid19_Cleaned_After_Scaling.xlsx")
scaler = load(BASE_DIR / "Covid19_scaler.joblib")

# X = inputs, y = what we want to predict.
feature_columns = [column for column in data.columns if column != TARGET_COLUMN]
X = data[feature_columns]
y = data[TARGET_COLUMN]

# Split data into training and testing sets.
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Train model.
model = RandomForestRegressor(random_state=42)
model.fit(X_train, y_train)

# Evaluate model performance on test data.
r2 = model.score(X_test, y_test)
y_pred = model.predict(X_test)
mae = mean_absolute_error(y_test, y_pred)
mse = mean_squared_error(y_test, y_pred)
rmse = np.sqrt(mse)

# Convert metrics from scaled values to case counts.
case_count_index = SCALED_COLUMNS.index(TARGET_COLUMN)
mae_cases = mae * scaler.scale_[case_count_index]
mse_cases = mse * (scaler.scale_[case_count_index] ** 2)
rmse_cases = rmse * scaler.scale_[case_count_index]

print(f"R^2: {r2:.4f}")
print(f"MAE (Case Count): {mae_cases:.2f}")
print(f"MSE (Case Count): {mse_cases:.2f}")
print(f"RMSE (Case Count): {rmse_cases:.2f}")