# Always start by ignoring future warnings
import warnings; warnings.filterwarnings("ignore")
import numpy as np
np.random.seed(42)
 

# Data setup
import pandas as pd
url = 'https://raw.githubusercontent.com/jbrownlee/Datasets/master/pima-indians-diabetes.data.csv'
cols = ['Pregnancies','Glucose','BloodPressure','SkinThickness','Insulin','BMI','DiabetesPedigree','Age','Outcome']
df = pd.read_csv(url, names=cols)
print(df.shape)
print(df.head(3))

# What do the columns mean?
for col in df.columns:
    print(col)

# Check for missing values
print(df.isnull().sum())

# Quick data summary
print(df.describe())

# Check for zeros as missing in medical columns
cols_check = ['Glucose', 'BloodPressure', 'SkinThickness', 'Insulin', 'BMI']
for c in cols_check:
    zero_count = (df[c] == 0).sum()
    print(f"{c}: {zero_count} zeros")

# Replace zeros with column median for selected features
for c in cols_check:
    df[c] = df[c].replace(0, df[c].median())
print(df[cols_check].head(3))

# Plot histograms for features
import matplotlib.pyplot as plt
df.hist(bins=15, figsize=(12,7))
plt.tight_layout()
plt.show()

# Visualize glucose by outcome using a boxplot
import seaborn as sns
sns.boxplot(data=df, x='Outcome', y='Glucose')
plt.title('Glucose by Diabetes Outcome')
plt.show()

# See how many have diabetes vs not
sns.countplot(x='Outcome', data=df)
plt.title('Class Balance: Diabetes')
plt.show()

# Set up features and label arrays
X = df.drop('Outcome', axis=1)
y = df['Outcome']
print(X.shape, y.shape)

# Split to train and test sets
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
print(X_train.shape, X_test.shape)

# Train logistic regression model
from sklearn.linear_model import LogisticRegression
model = LogisticRegression(max_iter=500, solver='lbfgs')
model.fit(X_train, y_train)
print('Training complete.')

# Predict and check accuracy
from sklearn.metrics import accuracy_score
y_pred = model.predict(X_test)
acc = accuracy_score(y_test, y_pred)
print(f"Test Accuracy: {acc:.2f}")

# Confusion matrix to see detailed results
from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay
cm = confusion_matrix(y_test, y_pred)
disp = ConfusionMatrixDisplay(cm, display_labels=['No Diabetes','Diabetes'])
disp.plot()
plt.show()

# Which features matter most?
import numpy as np
coef = model.coef_[0]
for name, value in zip(X.columns, coef):
    print(f'{name}: {value:.2f}')

# Try the model on a new patient (input example)
input_vals = []
for c in X.columns:
    val = float(input(f'Enter {c}: '))
    input_vals.append(val)
prediction = model.predict([input_vals])[0]
if prediction == 1:
    print('Prediction: Diabetes present')
else:
    print('Prediction: No diabetes')

# Stretch: Try a random forest for extra accuracy
from sklearn.ensemble import RandomForestClassifier
rf = RandomForestClassifier(n_estimators=100, random_state=42)
rf.fit(X_train, y_train)
rf_acc = rf.score(X_test, y_test)
print(f"Random Forest Test Accuracy: {rf_acc:.2f}")

# Mini project: Try predicting a new disease outcome!
# 1. Find a different health dataset online.
# 2. Load, clean, and explore it with these same steps.
# 3. Train a classifier, make predictions, and visualize results.
# Share your project with a friend and teach them what you learned.
"If you enjoyed this, subscribe to our channel for more fun projects!"