# Suppress warnings for a clean output
import warnings; warnings.filterwarnings('ignore')
import numpy as np
np.random.seed(42)
 
 

# Data setup (Titanic Dataset)
import pandas as pd
url = 'https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv'
df = pd.read_csv(url)
print(df.shape)
print(df.head(3))

# Checking missing values
print(df.isnull().sum())
 
 

# Fill missing Age values with the median
median_age = df['Age'].median()
df['Age'] = df['Age'].fillna(median_age)
print(df['Age'].isnull().sum())
 
 

# Selecting numerical columns for scaling
num_cols = ['Age', 'Fare']
print(df[num_cols].head())
 
 

# Normalize Age and Fare between 0 and 1
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
df_scaled = df.copy()
df_scaled[num_cols] = scaler.fit_transform(df[num_cols])
print(df_scaled[num_cols].head())
 
 

# Standardize Age and Fare
from sklearn.preprocessing import StandardScaler
std_scaler = StandardScaler()
df_standard = df.copy()
df_standard[num_cols] = std_scaler.fit_transform(df[num_cols])
print(df_standard[num_cols].head())
 
 

# Label encoding the Sex column
from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
df_encoded = df.copy()
df_encoded['Sex'] = le.fit_transform(df['Sex'])
print(df_encoded[['Sex']].head(8))
 
 

# One-hot encoding the Embarked column
df_ohe = pd.get_dummies(df, columns=['Embarked'], prefix='Emb')
print(df_ohe.filter(like='Emb').head(8))
 
 

# Combine scaling and encoding in one step
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
num_features = ['Age','Fare']
cat_features = ['Sex','Embarked']
numeric_transformer = MinMaxScaler()
categorical_transformer = LabelEncoder()
# We need a helper for label encoding in pipelines
import numpy as np
from sklearn.base import BaseEstimator, TransformerMixin
class CustomLabelEncoder(BaseEstimator, TransformerMixin):
    def fit(self, X, y=None):
        self.le = LabelEncoder()
        self.le.fit(X)
        return self
    def transform(self, X):
        return self.le.transform(X).reshape(-1,1)
from sklearn.preprocessing import OneHotEncoder
preprocessor = ColumnTransformer(transformers=[
    ('num', MinMaxScaler(), num_features),
    ('cat', OneHotEncoder(), cat_features)
], remainder='drop')
X = df[num_features+cat_features].copy()
X_transformed = preprocessor.fit_transform(X)
print(X_transformed[:5])
 
 

# Mini-Project: Scale, Encode, and Split Data
from sklearn.model_selection import train_test_split
features = ['Pclass', 'Sex', 'Age', 'Fare', 'Embarked']
df_temp = df[features].copy()
# Fill missing Embarked
common_embarked = df_temp['Embarked'].mode()[0]
df_temp['Embarked'] = df_temp['Embarked'].fillna(common_embarked)
# One-hot encode Embarked, label encode Sex
df_temp['Sex'] = LabelEncoder().fit_transform(df_temp['Sex'])
df_temp = pd.get_dummies(df_temp, columns=['Embarked'], prefix='Emb')
# Normalize Age and Fare
scaler2 = MinMaxScaler()
df_temp[['Age','Fare']] = scaler2.fit_transform(df_temp[['Age','Fare']])
X = df_temp
y = df['Survived']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)
print('Train shape:', X_train.shape)
print('Test shape:', X_test.shape)
 
 

# Practice: Manually normalize a small list
data = [10, 20, 30, 40, 50]
min_val = min(data)
max_val = max(data)
data_norm = [(x - min_val) / (max_val - min_val) for x in data]
print('Original:', data)
print('Normalized:', data_norm)
 
 

# User input practice: Enter numbers to normalize
nums = input('Enter numbers separated by spaces: ')
nums = [float(x) for x in nums.split()]
min_n = min(nums)
max_n = max(nums)
normalized = [(x - min_n) / (max_n - min_n) for x in nums]
print('Normalized:', normalized)
 
 

# Challenge: Find all columns in Titanic data that are not numeric
non_numeric = df.select_dtypes(include=['object']).columns
print('Non-numeric columns:', list(non_numeric))