การใช้ดัชนีบูลีนกับ DataFrames

การทำดัชนีแบบบูลีนคืออะไร

import pandas as pd

df = pd.DataFrame({
    'product': ['A', 'B', 'C', 'D'],
    'revenue': [500, 1500, 800, 2000]
})

# Boolean condition produces a Series of True/False
mask = df['revenue'] > 1000
print(mask)
# 0    False
# 1     True
# 2    False
# 3     True

การใช้มาสก์บูลีน

import pandas as pd

df = pd.DataFrame({
    'product': ['A', 'B', 'C', 'D'],
    'revenue': [500, 1500, 800, 2000]
})

mask = df['revenue'] > 1000
filtered = df[mask]
print(filtered)
#   product  revenue
# 1       B     1500
# 3       D     2000

เงื่อนไขแบบอินไลน์โดยไม่ใช้ตัวแปรมาสก์

import pandas as pd

df = pd.DataFrame({
    'city': ['NYC', 'LA', 'Chicago', 'Houston'],
    'population': [8_336_817, 3_979_576, 2_693_976, 2_320_268]
})

# Inline boolean filter
large_cities = df[df['population'] > 3_000_000]
print(large_cities)
#   city  population
# 0  NYC   8336817
# 1   LA   3979576

การรวมเงื่อนไขด้วย & (AND)

import pandas as pd

df = pd.DataFrame({
    'name': ['Alice', 'Bob', 'Carol', 'Dave'],
    'age': [25, 35, 28, 45],
    'salary': [50000, 90000, 70000, 120000]
})

# Both conditions must be true
result = df[(df['age'] > 27) & (df['salary'] > 60000)]
print(result)
#     name  age  salary
# 1    Bob   35   90000
# 2  Carol   28   70000

การรวมเงื่อนไขด้วย | (OR)

import pandas as pd

df = pd.DataFrame({
    'product': ['Laptop', 'Mouse', 'Monitor', 'Keyboard'],
    'price': [1200, 25, 300, 80],
    'category': ['Electronics', 'Accessories', 'Electronics', 'Accessories']
})

# Rows where price > 200 OR category is Accessories
result = df[(df['price'] > 200) | (df['category'] == 'Accessories')]
print(result)
#     product  price     category
# 0    Laptop   1200  Electronics
# 1     Mouse     25  Accessories
# 2   Monitor    300  Electronics
# 3  Keyboard     80  Accessories

การปฏิเสธเงื่อนไขด้วย ~

import pandas as pd

df = pd.DataFrame({
    'order_id': [1, 2, 3, 4],
    'status': ['shipped', 'cancelled', 'delivered', 'refunded']
})

# Keep only rows that are NOT cancelled
active = df[~(df['status'] == 'cancelled')]
print(active)
#    order_id     status
# 0         1    shipped
# 2         3  delivered
# 3         4   refunded

การกรองค่าข้อความ

import pandas as pd

df = pd.DataFrame({
    'country': ['USA', 'Germany', 'UK', 'USA', 'France'],
    'sales': [400, 200, 150, 600, 300]
})

# Filter rows where country equals USA
us_sales = df[df['country'] == 'USA']
print(us_sales)
#   country  sales
# 0     USA    400
# 3     USA    600

# Filter rows where country starts with 'G'
g_countries = df[df['country'].str.startswith('G')]
print(g_countries)
#    country  sales
# 1  Germany    200

การกรองหลายคอลัมน์

import pandas as pd

df = pd.DataFrame({
    'region': ['North', 'South', 'North', 'East'],
    'year': [2022, 2023, 2023, 2022],
    'profit': [5000, -200, 8000, 3000]
})

is_north = df['region'] == 'North'
is_2023 = df['year'] == 2023
is_profitable = df['profit'] > 0

result = df[is_north & is_2023 & is_profitable]
print(result)
#   region  year  profit
# 2  North  2023    8000

การใช้ np.where สำหรับคอลัมน์ตามเงื่อนไข

import pandas as pd
import numpy as np

df = pd.DataFrame({
    'score': [45, 72, 88, 61, 95]
})

# Create a new column based on a condition
df['grade'] = np.where(df['score'] >= 70, 'Pass', 'Fail')
print(df)
#    score grade
# 0     45  Fail
# 1     72  Pass
# 2     88  Pass
# 3     61  Fail
# 4     95  Pass

การกำหนดค่าให้ชุดย่อยที่กรองแล้ว

import pandas as pd

df = pd.DataFrame({
    'item': ['A', 'B', 'C', 'D'],
    'stock': [0, 5, 0, 12]
})

# Mark out-of-stock items
df.loc[df['stock'] == 0, 'status'] = 'Out of Stock'
df.loc[df['stock'] > 0, 'status'] = 'Available'
print(df)
#   item  stock        status
# 0    A      0  Out of Stock
# 1    B      5     Available
# 2    C      0  Out of Stock
# 3    D     12     Available

การนับและหาผลรวมของมาสก์

import pandas as pd

df = pd.DataFrame({'value': [10, 55, 30, 80, 20, 70]})

mask = df['value'] > 40
print('Count of rows > 40:', mask.sum())      # 3
print('Fraction > 40:', mask.mean().round(2)) # 0.5

# Now apply
print(df[mask])
#    value
# 1     55
# 3     80
# 5     70

ตรวจสอบความเข้าใจอย่างรวดเร็ว

สรุปบทเรียน