AI Development Lesson 3: NumPy & Pandas

NumPy is the math engine for AI — fast array operations. Pandas is the data manipulation library — every AI project starts here.

NumPy

import numpy as np

# Arrays (much faster than Python lists!)
a = np.array([1, 2, 3, 4, 5])
b = np.array([10, 20, 30, 40, 50])

# Vectorized operations (no loops!)
a + b          # [11, 22, 33, 44, 55]
a * 2          # [2, 4, 6, 8, 10]
a ** 2         # [1, 4, 9, 16, 25]
np.sqrt(a)    # element-wise sqrt
np.dot(a, b)  # 550 (dot product)

# 2D arrays (matrices)
matrix = np.array([[1,2,3],[4,5,6],[7,8,9]])
matrix.shape   # (3, 3)
matrix[0]      # first row: [1,2,3]
matrix[:,1]    # second column: [2,5,8]
matrix.T       # transpose
np.linalg.inv(matrix)  # inverse

Pandas

import pandas as pd

# Load data
df = pd.read_csv("data.csv")        # from file
df = pd.read_json("data.json")

# Explore
df.head(5)          # first 5 rows
df.info()           # column types, nulls
df.describe()       # stats (mean, std, min, max)
df.shape            # (rows, columns)

# Filter
df[df["age"] > 30]
df[(df["age"] > 30) & (df["role"] == "admin")]

# Transform
df["full_name"] = df["first"] + " " + df["last"]
df["age_bucket"] = pd.cut(df["age"], bins=[0,18,65,100], labels=["youth","adult","senior"])

# Handle missing data
df.dropna()                      # remove rows with NaN
df.fillna(0)                     # fill NaN with 0
df["salary"].fillna(df["salary"].mean())

# Group and aggregate
df.groupby("department")["salary"].mean()
df.groupby("role").agg({"salary": "mean", "age": ["min","max"]})