Lab_ML(27/01/25)

# %%
#insurance

# %%
import pandas as pd
df = pd.read_csv('insurance.csv')
df

# %%
df['sex'] = df['sex'].astype('category')
df['sex'] = df['sex'].cat.codes
df

# %%
df['smoker'] = df['smoker'].astype('category')
df['smoker'] = df['smoker'].cat.codes

df['region'] = df['region'].astype('category')
df['region'] = df['region'].cat.codes
df

# %%
df.isnull().sum()

# %%
x = df.drop(columns='charges')
x

# %%
y = df['charges']

from sklearn.model_selection import train_test_split
X_train,X_test,y_train,y_test = train_test_split(x,y,test_size=0.3, random_state=0)

from sklearn.linear_model import LinearRegression
lr = LinearRegression()
lr.fit(X_train,y_train)

# %%
c = lr.intercept_
c

# %%
m=lr.coef_
m

# %%
y_pred_train = lr.predict(X_train)
y_pred_train

# %%
import matplotlib.pyplot as plt

# Create the scatter plot
fig, ax = plt.subplots()
scatter = ax.scatter(y_train, y_pred_train, c='#3f3f3f', picker=True)  # Enable picking
plt.xlabel('Actual Charges')
plt.ylabel('Predicted Charges')

# Function to handle pick events (clicking on a point)
def on_pick(event):
    ind = event.ind[0]  # Get the index of the clicked point
    actual = y_train.iloc[ind]
    predicted = y_pred_train[ind]

    # Create or update the annotation
    if hasattr(on_pick, 'annotation'):
        on_pick.annotation.remove()  # Remove the previous annotation
    on_pick.annotation = ax.annotate(
        f"Actual: {actual:.2f}\nPredicted: {predicted:.2f}",
        xy=(y_train.iloc[ind], y_pred_train[ind]),
        xytext=(10, 10), textcoords='offset points',
        bbox=dict(boxstyle='round,pad=0.5', fc='yellow', alpha=0.5),
        arrowprops=dict(arrowstyle='->')
    )
    fig.canvas.draw()  # Redraw the figure to show the annotation

# Connect the pick event to the function
fig.canvas.mpl_connect('pick_event', on_pick)

# Show the plot
plt.show()

# %%
from sklearn.metrics import mean_absolute_error, r2_score

mae = mean_absolute_error(y_train, y_pred_train)
r2 = r2_score(y_train, y_pred_train)
print(f"MAE: {mae}")
print(f"R-squared: {r2}")

# %%
y_pred_test = lr.predict(X_test)
y_pred_test

# %%
import matplotlib.pyplot as plt

# Create the scatter plot
fig, ax = plt.subplots()
scatter = ax.scatter(y_train, y_pred_train, c='#3f3f3f', picker=True)  # Enable picking
plt.xlabel('Actual Charges')
plt.ylabel('Predicted Charges')

# Function to handle pick events (clicking on a point)
def on_pick(event):
    ind = event.ind[0]  # Get the index of the clicked point
    actual = y_test.iloc[ind]
    predicted = y_pred_test[ind]

    # Create or update the annotation
    if hasattr(on_pick, 'annotation'):
        on_pick.annotation.remove()  # Remove the previous annotation
    on_pick.annotation = ax.annotate(
        f"Actual: {actual:.2f}\nPredicted: {predicted:.2f}",
        xy=(y_test.iloc[ind], y_pred_test[ind]),
        xytext=(10, 10), textcoords='offset points',
        bbox=dict(boxstyle='round,pad=0.5', fc='yellow', alpha=0.5),
        arrowprops=dict(arrowstyle='->')
    )
    fig.canvas.draw()  # Redraw the figure to show the annotation

# Connect the pick event to the function
fig.canvas.mpl_connect('pick_event', on_pick)

# Show the plot
plt.show()

# %%
from sklearn.metrics import mean_absolute_error, r2_score

mae = mean_absolute_error(y_train, y_pred_train)
r2 = r2_score(y_test, y_pred_test)
print(f"MAE: {mae}")
print(f"R-squared: {r2}")