Master NumPy - Scientific Computing

NumPy Arrays Fundamentals

NumPy arrays are the heart of scientific computing - fast, memory-efficient, and mathematically powerful multidimensional arrays.

Function	Creates	Use Case
`np.array()`	From lists	Basic arrays
`np.zeros()`	Zero arrays	Matrix initialization
`np.ones()`	Ones arrays	Weight initialization
`np.arange()`	Sequences	Index arrays
`np.linspace()`	Even spacing	Plotting data

Hello NumPy World

🚀 python numpy_intro.py

[1 2 3]

📈 Array Created!

Shape: (3,)
Dtype: int64
Memory: 24 bytes
Speed: 100x faster than lists

import numpy as np

# Create basic array
arr = np.array([1, 2, 3])
print("Array:", arr)
print("Shape:", arr.shape)
print("Type:", arr.dtype)

# Common array creation
zeros = np.zeros(5)
ones = np.ones((3, 3))
range_arr = np.arange(0, 10, 2)
linspace = np.linspace(0, 1, 5)

print("Zeros:", zeros)
print("3x3 Ones:\n", ones)
print("Range:", range_arr)
print("Linspace:", linspace)
                

Vectorized Operations

Universal Functions (ufunc)

a = np.array([1, 2, 3])
b = np.array([4, 5, 6])

print(a + b)  # [5 7 9]
print(a * 2)  # [2 4 6]
print(np.sqrt(a))  # [1. 1.41 1.73]
                        

Advanced Indexing

arr = np.array([10, 20, 30, 40, 50])
print(arr[1:4])     # [20 30 40]
print(arr[::2])     # [10 30 50]
print(arr[[0, 2, 4]])  # [10 30 50]
                        

Reshaping Arrays

flat = np.arange(12)
matrix = flat.reshape(3, 4)
print("Matrix:\n", matrix)
print("Transposed:\n", matrix.T)
                        

Statistical Functions

data = np.random.randn(1000)
print("Mean:", np.mean(data))
print("Std:", np.std(data))
print("Min/Max:", np.min(data), np.max(data))
                        

Multidimensional Arrays

[[1 2 3] [4 5 6] [7 8 9]]

🔢 3x3 Matrix

1

2

3

4

5

6

7

8

9

import numpy as np

# 2D Matrix
matrix = np.array([[1, 2, 3],
                   [4, 5, 6],
                   [7, 8, 9]])

print("Matrix shape:", matrix.shape)
print("Row 1:", matrix[0])
print("Column 2:\n", matrix[:, 1])
print("Element [1,2]:", matrix[1, 2])

# Matrix operations
print("Sum all:", np.sum(matrix))
print("Row sums:", np.sum(matrix, axis=1))
print("Column sums:", np.sum(matrix, axis=0))
                

Broadcasting & Linear Algebra

# Broadcasting: arrays of different shapes
a = np.array([1, 2, 3])
b = 2
print("Broadcasting:", a * b)  # [2 4 6]

# Dot product
A = np.array([[1, 2], [3, 4]])
B = np.array([[5, 6], [7, 8]])
print("Matrix product:\n", np.dot(A, B))

# Eigenvalues
eigenvals = np.linalg.eigvals(A)
print("Eigenvalues:", eigenvals)

# Solve linear system Ax = b
x = np.linalg.solve(A, np.array([10, 20]))
print("Solution x:", x)
                

🔍 Broadcasting Example

a = np.array([[1,2,3], [4,5,6]])
b = np.array([10, 20, 30])
a + b → [[11 22 33], [14 25 36]]

Random Numbers & Statistics

# Random arrays
np.random.seed(42)  # Reproducible
rand_normal = np.random.randn(3, 3)  # Normal distribution
rand_uniform = np.random.rand(3, 3)  # Uniform [0,1]
rand_int = np.random.randint(0, 100, (3, 3))

print("Normal:\n", rand_normal)
print("Uniform:\n", rand_uniform)

# Advanced statistics
data = np.random.randn(10000)
print("Percentiles:", np.percentile(data, [25, 50, 75]))
print("Correlation:", np.corrcoef(np.random.randn(50), np.random.randn(50)))
                

Performance Optimization

Memory Views & Slicing

# Views (no copy)
arr = np.arange(10)
view = arr[::2]
view[0] = 999
print("Original modified:", arr)  # [999  1  2 999  4 ...]

# Vectorized vs loops (100x faster)
data = np.arange(1000000)
%timeit np.sum(data)        # Fast
%timeit sum(data.tolist())  # Slow
                

Universal Functions

# Compile for speed
from numba import jit
@jit(nopython=True)
def fast_sum(arr):
    total = 0
    for i in arr:
        total += i
    return total
                

Production Applications

Machine Learning
TensorFlow, PyTorch

Data Science
Pandas, Matplotlib

Scientific Computing
Physics, Engineering

Production Projects

1. Image Processing Pipeline

Load, filter, transform, and analyze images with NumPy arrays

2. Financial Time Series

Moving averages, volatility, correlations, risk metrics

3. Monte Carlo Simulations

1M+ path simulations for pricing derivatives and risk analysis

4. Neural Network from Scratch

Forward/backward pass, optimization, matrix operations

Scientific Computing Expert!

⚡ Production NumPy Applications:

🧠 ML Model Training
Matrix ops, gradients

📈 Data Analysis
Statistics, aggregation

🖼️ Computer Vision
Image processing

💰 Algorithmic Trading
Real-time analytics

🚀 Production Speed
100x faster than pure Python

Power machine learning, data science, and scientific research! 🚀

NumPy Mastery

NumPy - Scientific Computing