NumPy

A concise NumPy cheatsheet

NumPy Basics

TaskCodeDescription
Import NumPyimport numpy as npImport NumPy library
Check versionnp.versionDisplay NumPy version
Create arraynp.array([1,2,3])Create 1D array
Create 2D arraynp.array([[1,2,3],[4,5,6]])Create 2D array
Create zerosnp.zeros((3,3))3×3 matrix of zeros
Create onesnp.ones((2,4))2×4 matrix of ones
Create constant arraynp.full((2,3), 7)Matrix filled with 7
Create identity matrixnp.eye(3)3×3 identity matrix
Create random arraynp.random.rand(2,3)Uniform random numbers
Create random intnp.random.randint(1,10,(3,3))Random integers

Array Inspection

TaskCodeDescription
Shape of arrayarr.shape(rows, columns)
Dimensionsarr.ndimNo. of dimensions
Total elementsarr.sizeTotal elements
Data typearr.dtypeElement data type
Change typearr.astype(float)Convert type
Item sizearr.itemsizeBytes per element
Info summaryarr.flagsMemory info

Array Creation Helpers

TaskCodeDescription
Range arraynp.arange(0,10,2)Evenly spaced values
Linspacenp.linspace(0,1,5)5 values from 0 to 1
Random normalnp.random.randn(3,3)Standard normal distribution
Empty arraynp.empty((2,2))Uninitialized values
Identity matrixnp.identity(4)4×4 identity matrix
Diagonal matrixnp.diag([1,2,3])Diagonal matrix

Mathematical Operations

TaskCodeDescription
Add arraysa + bElement-wise addition
Subtract arraysa - bElement-wise subtraction
Multiply arraysa * bElement-wise multiplication
Divide arraysa / bElement-wise division
Dot productnp.dot(a,b)Matrix multiplication
Matrix producta @ bMatrix multiplication operator
Exponentiationnp.exp(a)e^a
Square rootnp.sqrt(a)√a
Logarithmnp.log(a)Natural log
Powernp.power(a,2)a²
Roundingnp.round(a,2)Round to 2 decimals

Statistical Operations

TaskCodeDescription
Sumnp.sum(a)Total sum
Meannp.mean(a)Average
Mediannp.median(a)Median
Standard deviationnp.std(a)Std. deviation
Variancenp.var(a)Variance
Min valuenp.min(a)Minimum
Max valuenp.max(a)Maximum
Argmin indexnp.argmin(a)Index of min value
Argmax indexnp.argmax(a)Index of max value
Cumulative sumnp.cumsum(a)Running total
Cumulative productnp.cumprod(a)Running product

Array Indexing & Slicing

TaskCodeDescription
Get elementa[0], a[1,2]Access element
Slice arraya[1:4]Elements 1–3
2D slicinga[0:2, 1:3]Subarray slice
Step slicinga[::2]Every 2nd element
Reversea[::-1]Reverse array
Conditionala[a > 5]Filter values
Fancy indexinga[[0,2,4]]Select specific indices
Replace valuesa[a < 0] = 0Conditional assignment

Reshaping & Combining

TaskCodeDescription
Reshapea.reshape(3,2)Change shape
Flattena.flatten()1D version
Ravela.ravel()Flattened view
Transposea.TSwap rows/columns
Concatenatenp.concatenate((a,b), axis=0)Stack vertically
Stack horizontallynp.hstack((a,b))Horizontal merge
Stack verticallynp.vstack((a,b))Vertical merge
Split arraynp.split(a, 3)Split into parts

Linear Algebra

TaskCodeDescription
Matrix productnp.dot(A,B)Multiply matrices
TransposeA.TTranspose matrix
Determinantnp.linalg.det(A)Determinant
Inversenp.linalg.inv(A)Matrix inverse
Eigenvaluesnp.linalg.eig(A)Eigenvalues/vectors
Ranknp.linalg.matrix_rank(A)Rank of matrix
Solve linear eq.np.linalg.solve(A, b)Solve Ax = b

Broadcasting

TaskCodeDescription
Add scalara + 10Add scalar to array
Multiply scalara * 5Multiply array
Add vectorsa + bAutomatically expands shapes
Different shapesa + np.array([1,2,3])Matches smaller dimension

Random Module

TaskCodeDescription
Random float [0,1)np.random.rand(3,3)Uniform random
Random integersnp.random.randint(0,10,(3,3))Random ints
Normal distributionnp.random.randn(3,3)Standard normal
Random choicenp.random.choice([1,2,3,4])Random pick
Shuffle arraynp.random.shuffle(a)In-place shuffle
Seed for repeatnp.random.seed(42)Reproducibility

Useful Utilities

TaskCodeDescription
Unique elementsnp.unique(a)Get unique values
Sort arraynp.sort(a)Sort elements
Concatenate arraysnp.concatenate((a,b))Join arrays
Compare arraysnp.array_equal(a,b)Check equality
Copy arrayb = a.copy()Deep copy
Repeat elementsnp.repeat(a, 2)Repeat each element
Tile arraynp.tile(a, 3)Repeat whole array

Logical Operations

TaskCodeDescription
Element-wise ANDnp.logical_and(a>0, a<5)Logical AND
Element-wise ORnp.logical_or(a>5, a==0)Logical OR
Element-wise NOTnp.logical_not(a)Logical negation
Any True?np.any(a > 0)Any element True
All True?np.all(a > 0)All elements True
Where conditionnp.where(a>0, 'Yes', 'No')Conditional output