xgboost_tune.py

import xgboost as xgb
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error, r2_score, median_absolute_error, mean_absolute_error
from sklearn.metrics.pairwise import cosine_similarity
import numpy as np
from sklearn.model_selection import RandomizedSearchCV
from scipy.stats import uniform, randint

data = pd.read_csv('dataset_paper_final.csv')

X = data[['hydrophobicity', 'aromaticity', 'isoelectric_point', 'molecular_weight', 'sequence_length', 'instability_index', 'cystein_count']]
y = data['retention_time']
print('Data in input and output loaded')

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
print('training and test data split')


param_dist = {
    'n_estimators': randint(400, 600),
    'max_depth': randint(8, 15),
    'learning_rate': uniform(0.1, 0.2),
    'colsample_bytree': uniform(0.6, 0.3),
    'gamma': uniform(0.1, 0.3),
    'reg_alpha': uniform(0.5, 1),
    'reg_lambda': uniform(5, 8)
}

xgb_model = xgb.XGBRegressor(objective='reg:squarederror')

random_search = RandomizedSearchCV(
    estimator=xgb_model,
    param_distributions=param_dist,
    n_iter=50,
    scoring='neg_mean_squared_error',
    cv=5,
    verbose=1,
    random_state=42,
    n_jobs=-1
)

random_search.fit(X_train, y_train)

print("Best parameters:", random_search.best_params_)
best_model = random_search.best_estimator_
y_train_pred =best_model.predict(X_train)
y_test_pred = best_model.predict(X_test)
y_pred = best_model.predict(X_test)

y_test_2D = np.array(y_test).reshape(1,-1)
y_pred_2D = np.array(y_pred).reshape(1,-1)


train_mae = mean_absolute_error(y_train, y_train_pred)
test_mae = mean_absolute_error(y_test, y_test_pred)
train_r2 = r2_score(y_train, y_train_pred)
test_r2 = r2_score(y_test, y_test_pred)
cosine_similarity = cosine_similarity(y_test_2D, y_pred_2D)
print('errors calculated')

train_median_absolute_error = median_absolute_error(y_train, y_train_pred)
test_median_absolute_error = median_absolute_error(y_test, y_test_pred)

print(f'Training Mean Absolute Error: {train_mae}')
print(f'Testing Mean Absolute Error: {test_mae}')
print(f'Training R^2 Score: {train_r2}')
print(f'Testing R^2 Score: {test_r2}')
print(f'Training Median Absolute Error: {train_median_absolute_error}')
print(f'Testing Median Absolute Error: {test_median_absolute_error}')
print(f'Training Cosine Similarity: {cosine_similarity}')