import warnings; warnings.filterwarnings("ignore")

# Let us import Pandas.
import pandas as pd
# Also import numpy for numerical calculations.
import numpy as np

# Create a simple Series of fruit counts
fruits = pd.Series([3, 5, 2], index=["apples", "bananas", "cherries"])

print(fruits)

# Accessing Series values by label
print("Bananas count:", fruits["bananas"])

# Make a small DataFrame for a class
data = {
    "name": ["Ana", "Ben", "Cleo"],
    "score": [90, 85, 88],
    "passed": [True, True, True]
}
students = pd.DataFrame(data)
print(students)

# Show the shape and column names
print("Shape:", students.shape)
print("Columns:", students.columns.tolist())

# Load data from a CSV file
url = "https://raw.githubusercontent.com/jbrownlee/Datasets/master/shampoo.csv"
sales = pd.read_csv(url)
print("Shape:", sales.shape)
print(sales.head())

# Plot the shampoo sales data
import matplotlib.pyplot as plt

plt.plot(sales["Month"], sales["Sales"])
plt.title("Monthly Shampoo Sales")
plt.xlabel("Month")
plt.ylabel("Sales")
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

# Accessing a single column
monthly_sales = sales["Sales"]
print(monthly_sales.head())

# Describing the numbers
print(monthly_sales.describe())

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

# Fill missing values, if any, with 0
sales_filled = sales.fillna(0)
print(sales_filled.head())

# Making a new column from other columns
sales_filled["LogSales"] = sales_filled["Sales"].apply(lambda x: 0 if x == 0 else np.log(x))
print(sales_filled.head())

# Filtering rows: sales over 500
high_sales = sales_filled[sales_filled["Sales"] > 500]
print(high_sales)

# Sorting the DataFrame by sales
sorted_sales = sales_filled.sort_values(by="Sales", ascending=False)
print(sorted_sales.head())

# Grouping sales by the first year
sales_filled["Year"] = sales_filled["Month"].apply(lambda x: str(x).split("-")[0])
yearly_sales = sales_filled.groupby("Year")["Sales"].sum()
print(yearly_sales)

# Iterating over a DataFrame, row by row
for i, row in sales_filled.iterrows():
    print("Month:", row["Month"], "Sales:", row["Sales"])
    if i > 2:
        break
    

# Mini-project: Find the best month
best_month = sales_filled.loc[sales_filled["Sales"].idxmax(), "Month"]
print("Best month for shampoo sales was:", best_month)

# Mini-project: Simple growth calculation
first = sales_filled["Sales"].iloc[0]
last = sales_filled["Sales"].iloc[-1]
growth = (last - first) / first * 100
print("Total sales growth over the period:", round(growth, 2), "%")

# Catching errors when accessing columns
try:
    print(sales["profit"])
except KeyError:
    print("There is no profit column in our data.")
    

# User input with a DataFrame
column = input("Type a column name to see (Month or Sales): ")
if column in sales.columns:
    print(sales[column].head())
else:
    print("That column is not in our data.")