Pandas: Getting Started with Pandas

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import numpy as np

# Customer data: name(str), age(int), salary(float)
names = np.array(['Alice', 'Bob', 'Charlie'])
ages = np.array([28, 34, 25])
salaries = np.array([75000.0, 92000.0, 68000.0])

# Problem 1: mixing types forces upcast to object
mixed = np.array(['Alice', 28, 75000.0])
print(mixed.dtype)  # object — loses numeric operations!

# Problem 2: cannot compute mean on object array
try:
    print(np.mean(mixed))  # Error or meaningless result
except TypeError:
    print("Cannot compute mean on object array")

# Problem 3: missing values need special handling
ages_with_nan = np.array([28, 34, np.nan, 25])
print(np.mean(ages_with_nan))  # nan — need np.nanmean()

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import pandas as pd

# One DataFrame handles mixed types naturally
df = pd.DataFrame({
    'name': ['Alice', 'Bob', 'Charlie'],
    'age': [28, 34, 25],
    'salary': [75000.0, 92000.0, 68000.0],
    'hire_date': pd.to_datetime(['2022-03-15', '2019-08-01', '2023-01-10'])
})

print(df.dtypes)
# name                 object
# age                   int64
# salary              float64
# hire_date    datetime64[ns]

print(df['salary'].mean())  # 78333.33 — works naturally
print(df[df['age'] > 27])   # filter by condition

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import numpy as np
import pandas as pd

# NumPy: must remember column order
arr = np.array([['Alice', 28, 75000.0],
                ['Bob', 34, 92000.0]])
# What is column 2? Have to check documentation
print(arr[0, 2])  # 75000.0 — but what does this mean?

# Pandas: column names are self-documenting
df = pd.DataFrame(arr, columns=['name', 'age', 'salary'])
df['age'] = df['age'].astype(int)
df['salary'] = df['salary'].astype(float)
print(df['salary'])  # column name says it all
# 0    75000.0
# 1    92000.0
print(df.loc[0, 'name'])  # Alice — label-based access

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import numpy as np
import pandas as pd

# NumPy: nan in integer array forces float conversion
arr = np.array([1, 2, np.nan, 4])
print(arr.dtype)  # float64 — integers lost!
print(np.mean(arr))  # nan — need np.nanmean()

# Pandas: handles missing values per column
s = pd.Series([1, 2, pd.NA, 4], dtype='Int64')  # nullable integer
print(s)  # keeps integer type with <NA>
print(s.mean())  # 2.333 — auto-skips missing

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import pandas as pd
from io import StringIO

# Read from CSV string
csv_data = """name,age,salary
Alice,28,75000
Bob,34,92000
Charlie,25,68000"""
df = pd.read_csv(StringIO(csv_data))
print(df)

# Write to different formats
# df.to_csv('output.csv', index=False)
# df.to_excel('output.xlsx', sheet_name='Employees')
# df.to_json('output.json', orient='records')

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import numpy as np
import pandas as pd

# ============================================
# Comprehensive comparison: NumPy vs Pandas
# for student grade analysis workflow
# ============================================

# --- Raw data ---
students = ['Alice', 'Bob', 'Charlie', 'Carol', 'David']
math = [85, 92, 78, 95, 88]
english = [90, 85, 82, 88, 91]
science = [88, 90, 75, 93, 86]

# --- NumPy approach ---
scores_np = np.array([math, english, science])  # shape: (3, 5)
# Which row is which subject? Must remember: row 0=math, 1=english, 2=science
# Which column is which student? Must remember order
avg_per_student = scores_np.mean(axis=0)
print("NumPy - Average per student (by position):")
for i, avg in enumerate(avg_per_student):
    print(f"  Student {i}: {avg:.1f}")  # Who is Student 0?

# --- Pandas approach ---
df = pd.DataFrame({
    'student': students,
    'math': math,
    'english': english,
    'science': science
})
df['average'] = df[['math', 'english', 'science']].mean(axis=1)
df['grade'] = df['average'].map(
    lambda x: 'A' if x >= 90 else 'B' if x >= 85 else 'C'
)
print("\nPandas - Student grades:")
print(df[['student', 'average', 'grade']])

# Top performers
top = df[df['grade'] == 'A']
print(f"\nA-grade students: {top['student'].tolist()}")

# Subject statistics
print("\nSubject statistics:")
print(df[['math', 'english', 'science']].describe())

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

NumPy - Average per student (by position):
  Student 0: 87.7
  Student 1: 89.0
  Student 2: 78.3

Pandas - Student grades:
   student  average grade
0    Alice     87.7     B
1      Bob     89.0     B
2  Charlie     78.3     C
3    Carol     92.0     A
4    David     88.3     B

A-grade students: ['Carol']

Subject statistics:
        math  english  science
count   5.00     5.00     5.00
mean   87.60    87.20    86.40
std     6.19     3.70     7.60
min    78.00    82.00    75.00
max    95.00    91.00    93.00

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.

import pandas as pd
import numpy as np

print(f"Pandas version: {pd.__version__}")
print(f"NumPy version: {np.__version__}")

# Quick functionality check
df = pd.DataFrame({'x': [1, 2, 3], 'y': [4, 5, 6]})
print(df)
print(f"\nMean of x: {df['x'].mean()}")

> Output: Run this in a local Python environment (pandas 2.x). The Piston server does not have pandas pre-installed — install it locally (`pip install pandas`) to follow along. Actual values may vary slightly depending on your pandas version.