import warnings; warnings.filterwarnings("ignore")
# Data setup
import pandas as pd
import kagglehub, os
import numpy as np
np.random.seed(42)
path = kagglehub.dataset_download('heeraldedhia/groceries-dataset')
files = os.listdir(path)
csv_file = [f for f in files if f.endswith('.csv')][0]
df = pd.read_csv(os.path.join(path, csv_file))
print(df.shape)
print(df.head(3))

# Let us see the basic structure of our dataset
df.info()

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

# View top purchased items
top_items = df['itemDescription'].value_counts().head(10)
print(top_items)

# Visualize most popular items with a histogram
import matplotlib.pyplot as plt
plt.figure(figsize=(10,4))
top_items.plot(kind='bar', color='skyblue')
plt.ylabel('Frequency')
plt.title('Top 10 Purchased Grocery Items')
plt.show()

# How many items does each customer buy per shopping trip?
items_per_trip = df.groupby(['Member_number', 'Date']).size()
plt.hist(items_per_trip, bins=20, color='purple', edgecolor='white')
plt.xlabel('Number of Items per Trip')
plt.ylabel('Number of Transactions')
plt.title('Distribution of Items Bought Each Trip')
plt.show()

# Scatterplot: Items per trip over time for one customer
sample_member = df['Member_number'].sample(1, random_state=42).iloc[0]
member_transactions = df[df['Member_number'] == sample_member]
trips = member_transactions.groupby('Date').size().reset_index(name='ItemsBought')
plt.scatter(trips['Date'], trips['ItemsBought'], alpha=0.7, color='orange')
plt.xlabel('Date')
plt.ylabel('Items Bought in Trip')
plt.title(f'Shopping Pattern Over Time: Member {sample_member}')
plt.xticks(rotation=45)
plt.show()

# Preprocess data for a heatmap: Days and items
item_by_day = df.groupby(['Date', 'itemDescription']).size().unstack(fill_value=0)
item_totals = item_by_day.sum(axis=0).sort_values(ascending=False)[:8]
small = item_by_day[item_totals.index]
small = small.T

# Draw the heatmap
import seaborn as sns
plt.figure(figsize=(12,5))
sns.heatmap(small, cmap='YlGnBu', linewidths=0.5, cbar_kws={'label': 'Count'})
plt.ylabel('Item')
plt.xlabel('Date')
plt.title('Heatmap of Item Purchases Across Days')
plt.show()

# Practice: Make your own histogram
your_col = input("Type column name: 'Member_number' or 'itemDescription': ")
vc = df[your_col].value_counts().head(10)
vc.plot(kind='bar', color='salmon')
plt.title(f'Your Top 10: {your_col}')
plt.show()

# Challenge: Try a heatmap for days and members
pivot = df.pivot_table(index='Date', columns='Member_number', values='itemDescription', aggfunc='count', fill_value=0)
sample_cols = pivot.columns[:15]
plt.figure(figsize=(12,8))
sns.heatmap(pivot[sample_cols], cmap='mako', cbar_kws={'label': 'Items Bought'})
plt.title('Shopping Heatmap: Dates vs. 15 Members')
plt.ylabel('Date')
plt.xlabel('Member Number')
plt.show()