import warnings; warnings.filterwarnings("ignore")
# We turn off warnings to keep our notebook clean.

import numpy as np
# NumPy lets us work with powerful math tools in Python.

# Let us make our first simple NumPy array
arr = np.array([10, 20, 30, 40, 50])
print(arr)

# NumPy arrays work a bit like lists, but they are more powerful.
print("First value:", arr[0])
print("Middle value:", arr[2])
print("Last value:", arr[-1])

# Math on arrays: add 5 to every value
new_arr = arr + 5
print(new_arr)

# What is inside an array? Let us check types and shapes.
print("Array type:", type(arr))
print("Data type:", arr.dtype)
print("Shape:", arr.shape)

# Making arrays of zeros or random numbers
zeroes = np.zeros(5)
print("All zeros:", zeroes)
rand_nums = np.random.rand(5)
print("Random values:", rand_nums)

# Generating random integers between 1 and 100
rand_ints = np.random.randint(1, 101, size=10)
print("Random integers:", rand_ints)

# Reshaping arrays: turning a 1D array into 2D (like a table)
table = np.arange(1, 7).reshape((2,3))
print(table)

# Accessing groups of values (slicing)
print(arr[1:4])  # from 2nd to 4th item (not including the last index)

# What happens if you use an index too large?
try:
    print(arr[10])
except IndexError:
    print("That position does not exist! Try a smaller number.")
    

import pandas as pd
url = "https://raw.githubusercontent.com/jbrownlee/Datasets/master/shampoo.csv"
shampoo = pd.read_csv(url)
print("Shape:", shampoo.shape)
print(shampoo.head())

# Visualizing the time series using matplotlib
import matplotlib.pyplot as plt
plt.figure(figsize=(10,4))
plt.plot(shampoo["Sales"])
plt.title("Shampoo Sales Over Time")
plt.xlabel("Month")
plt.ylabel("Sales (Thousands)")
plt.show()

# Converting pandas data to a NumPy array for number crunching
sales_arr = shampoo["Sales"].values
print(type(sales_arr))
print(sales_arr[:10])

# Simple stats: understanding your data
print("Mean:", np.mean(sales_arr))
print("Minimum:", np.min(sales_arr))
print("Maximum:", np.max(sales_arr))

# Creating a new array: normalized sales as percentages of max
norm_sales = sales_arr / np.max(sales_arr) * 100
print(norm_sales[:5])

# Find months where sales were above 350 (thousands)
high_months = np.where(sales_arr > 350)[0]
print("Months with highest sales:", high_months)

# Mini-project part 1: Generate synthetic sales data for 24 months
np.random.seed(42)
fake_sales = np.random.normal(loc=300, scale=50, size=24)
print("First 5 fake sales:", fake_sales[:5])

# Mini-project part 2: Plot real vs. synthetic sales
plt.figure(figsize=(10,4))
plt.plot(sales_arr[:24], label="Real Sales")
plt.plot(fake_sales, label="Simulated Sales")
plt.legend()
plt.title("Compare Real and Simulated Sales")
plt.xlabel("Month")
plt.ylabel("Sales (Thousands)")
plt.show()

# Some best practices for working with NumPy arrays
# 1. Always set random seeds for reproducible results.
# 2. Keep your arrays the right shape for operations.
# 3. Label your charts and data for clarity.

# Troubleshooting: When arrays do not match up
try:
    arr_a = np.array([1,2,3])
    arr_b = np.array([4,5])
    arr_sum = arr_a + arr_b
except ValueError as e:
    print("Error! Arrays are not the same size:", e)
    

# Extra tip: You can use input() to get numbers from a user
user_num = input("Type a sales number: ")
user_num = float(user_num)
print("You entered:", user_num)

# Challenge: Find average only for months with sales above 300
high_sales = sales_arr[sales_arr > 300]
avg_high = np.mean(high_sales)
print("Average of high sales:", avg_high)