import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import openml
import warnings
warnings.filterwarnings('ignore')

# Load customer satisfaction survey data from OpenML
dataset = openml.datasets.get_dataset(42178)
df_satis, _, _, _ = dataset.get_data(dataset_format='dataframe')
print(df_satis.shape)
print(df_satis.head(3))

# Load marketing campaign performance data
dataset = openml.datasets.get_dataset(1461)
df_marketing, _, _, _ = dataset.get_data(dataset_format='dataframe')
df_marketing.columns = ['age','job','marital','education','default','balance','housing','loan','contact','day','month','duration','campaign','pdays','previous','poutcome','response']
print(df_marketing.shape)
print(df_marketing.head(3))

# Load synthetic Net Promoter Score (NPS) survey data
np.random.seed(42)
df_nps = pd.DataFrame({'CustomerID': range(1,501),
                      'Age': np.random.randint(18,70,500),
                      'Region': np.random.choice(['North','South','East','West'],500),
                      'NPS_Score': np.random.randint(0,11,500)})
print(df_nps.shape)
print(df_nps.head(3))

# Beginner Example 1: Count survey responses by gender
ax = df_satis['gender'].value_counts().plot(kind='bar', color='skyblue')
plt.title('Number of Survey Respondents by Gender')
plt.xlabel('Gender')
plt.ylabel('Count')
plt.tight_layout()
plt.show()

# Beginner Example 2: Plot NPS Score distribution
plt.figure(figsize=(8,4))
sns.histplot(df_nps['NPS_Score'], bins=11, kde=False, color='green')
plt.title('Distribution of NPS Scores')
plt.xlabel('NPS Score')
plt.ylabel('Customer Count')
plt.tight_layout()
plt.show()

# Beginner Example 3: Pie chart of marketing campaign responses
response_counts = df_marketing['response'].value_counts()
plt.figure(figsize=(6,6))
plt.pie(response_counts, labels=response_counts.index, autopct='%1.1f%%', colors=['#89cff0','#ffcccb'])
plt.title('Customer Response Rate in Campaign')
plt.show()

# Intermediate Example 1: NPS by region
region_nps = df_nps.groupby('Region')['NPS_Score'].mean()
region_nps.plot(kind='bar', color=['#0072B2', '#D55E00', '#CC79A7', '#009E73'])
plt.title('Average NPS Score by Region')
plt.xlabel('Region')
plt.ylabel('Average NPS Score')
plt.ylim(0,10)
plt.tight_layout()
plt.show()

# Intermediate Example 2: Monthly charges vs churn rate
churn_rate = df_satis.groupby('MonthlyCharges')['Churn'].apply(lambda x: (x == 'Yes').mean())
plt.figure(figsize=(10,4))
plt.plot(churn_rate.index, churn_rate.values, marker='o', linestyle='-')
plt.title('Churn Rate by Monthly Charges')
plt.xlabel('Monthly Charges')
plt.ylabel('Churn Rate')
plt.grid(True)
plt.tight_layout()
plt.show()

# Intermediate Example 3: Cross-tabulation heatmap of Internet service vs Churn
crosstab = pd.crosstab(df_satis['InternetService'], df_satis['Churn'])
sns.heatmap(crosstab, annot=True, fmt='d', cmap='YlGnBu')
plt.title('Churn by Internet Service Type')
plt.xlabel('Churn')
plt.ylabel('Internet Service')
plt.tight_layout()
plt.show()

# Advanced Example 1: NPS segmentation by age group and region
df_nps['AgeGroup'] = pd.cut(df_nps['Age'], bins=[17,29,39,49,59,70], labels=['18-29','30-39','40-49','50-59','60-69'])
pivot = df_nps.pivot_table(index='AgeGroup', columns='Region', values='NPS_Score', aggfunc='mean')
plt.figure(figsize=(8,5))
sns.heatmap(pivot, annot=True, cmap='coolwarm', center=5, linewidths=0.5)
plt.title('Average NPS Score by Age Group and Region')
plt.ylabel('Age Group')
plt.xlabel('Region')
plt.tight_layout()
plt.show()

# Advanced Example 2: Visualizing customer retention trends using cohort analysis
np.random.seed(0)
dates = pd.date_range('2021-01-01', periods=24, freq='ME')
df_cohort = pd.DataFrame({'CustomerID': np.random.randint(1000,2000,len(dates)),
                         'Signup_Month': dates,
                         'Active_Users': np.random.randint(50,300,len(dates))})
plt.figure(figsize=(10,4))
plt.plot(df_cohort['Signup_Month'], df_cohort['Active_Users'], marker='o')
plt.title('Customer Retention Trend Over Time')
plt.xlabel('Signup Month')
plt.ylabel('Active Users')
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

# Advanced Example 3: Word cloud of open-ended customer feedback
from wordcloud import WordCloud, STOPWORDS
df_feedback = pd.DataFrame({'CustomerID':[1,2,3,4,5],
    'Feedback':['Great service and friendly staff','Delivery was slow and packaging was poor',
    'Excellent quality, will buy again','Customer support needs improvement','Good value for money']})
text = ' '.join(df_feedback['Feedback'])
wordcloud = WordCloud(stopwords=STOPWORDS, background_color='white', width=600, height=400).generate(text)
plt.figure(figsize=(8,4))
plt.imshow(wordcloud, interpolation='bilinear')
plt.axis('off')
plt.title('Customer Feedback Highlights')
plt.show()

# Error Handling Example 1: Detect missing values in survey data
missing_counts = df_satis.isnull().sum()
print('Missing values per column:')
print(missing_counts[missing_counts > 0])

# Error Handling Example 2: Incorrect groupby on categorical fields
try:
    result = df_marketing.groupby('balance')['age'].mean()
    print(result.head())
except Exception as e:
    print('Error:', e)
    print('Check if grouping field is too granular and should be binned.')

# Error Handling Example 3: Misinterpreting NPS score distribution
promoters = (df_nps['NPS_Score'] >= 9).sum()
detractors = (df_nps['NPS_Score'] <= 6).sum()
passives = ((df_nps['NPS_Score'] >= 7) & (df_nps['NPS_Score'] <= 8)).sum()
total = len(df_nps)
nps = ((promoters - detractors) / total) * 100
print(f'NPS = {nps:.1f} (should range from -100 to 100)')

# Pattern: Customer segmentation by satisfaction level
df_satis['Satisfaction_Level'] = pd.cut(df_satis['MonthlyCharges'], bins=[0,30,60,90,120], labels=['Basic','Standard','Premium','Elite'])
segment_counts = df_satis['Satisfaction_Level'].value_counts().sort_index()
segment_counts.plot(kind='bar', color='orange')
plt.title('Customer Segmentation by Monthly Charges')
plt.xlabel('Segment')
plt.ylabel('Customer Count')
plt.tight_layout()
plt.show()

# Pattern: Cross-tab of churn by payment method
crosstab2 = pd.crosstab(df_satis['PaymentMethod'], df_satis['Churn'], normalize='index')
plt.figure(figsize=(8,5))
sns.heatmap(crosstab2, annot=True, cmap='viridis', fmt='.2f')
plt.title('Churn Rate by Payment Method (Proportion)')
plt.xlabel('Churn')
plt.ylabel('Payment Method')
plt.tight_layout()
plt.show()

# Pattern: Trend analysis using moving average on active users
df_cohort['MA_3'] = df_cohort['Active_Users'].rolling(window=3, min_periods=1).mean()
plt.figure(figsize=(10,4))
plt.plot(df_cohort['Signup_Month'], df_cohort['Active_Users'], label='Active Users')
plt.plot(df_cohort['Signup_Month'], df_cohort['MA_3'], label='3-Month Moving Average', linestyle='--')
plt.title('Customer Retention Trend - 3 Month Smoothing')
plt.xlabel('Signup Month')
plt.ylabel('Active Users')
plt.legend()
plt.tight_layout()
plt.show()

# End-to-End Example: From survey data to business recommendation
# Step 1: Visualize satisfaction by contract type
contract_churn = pd.crosstab(df_satis['Contract'], df_satis['Churn'], normalize='index')
contract_churn.plot(kind='bar', stacked=True, color=['#e74c3c','#2ecc71'])
plt.title('Churn Rate by Contract Type')
plt.xlabel('Contract Type')
plt.ylabel('Proportion')
plt.legend(title='Churn')
plt.tight_layout()
plt.show()

# Step 2: Recommend a churn reduction action
highest_churn = contract_churn['Yes'].idxmax()
print(f'Customers on {highest_churn} contracts have the highest churn. Consider retention offers targeting this group.')