Tensor math operations in TensorFlow - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a tensor in TensorFlow?

A tensor is a multi-dimensional array used to represent data in TensorFlow. It can have any number of dimensions, like scalars (0D), vectors (1D), matrices (2D), or higher.

beginner

How do you add two tensors in TensorFlow?

You can add two tensors using tf.add(tensor1, tensor2) or simply using the + operator if tensors have the same shape.

intermediate

What does broadcasting mean in tensor operations?

Broadcasting lets TensorFlow automatically expand smaller tensors to match the shape of larger tensors during math operations, so you don't have to manually reshape them.

beginner

How do you multiply two tensors element-wise in TensorFlow?

Use tf.multiply(tensor1, tensor2) or the * operator to multiply tensors element by element, assuming they have compatible shapes.

intermediate

What function computes the dot product of two tensors in TensorFlow?

Use tf.tensordot(tensor1, tensor2, axes) to compute the dot product along specified axes. For matrix multiplication, tf.matmul is commonly used.

Which TensorFlow function adds two tensors element-wise?

A tf.add()

B tf.matmul()

C tf.multiply()

D tf.reshape()

tf.add() adds two tensors element-wise.

What happens if you add tensors of different shapes without broadcasting?

A The smaller tensor is ignored

B TensorFlow automatically reshapes them

C An error occurs

D The operation returns zeros

TensorFlow throws an error if shapes are incompatible and broadcasting is not possible.

Which operation performs element-wise multiplication of tensors?

A tf.tensordot()

B tf.matmul()

C tf.add()

D tf.multiply()

tf.multiply() multiplies tensors element-wise.

What does tf.matmul() do?

A Matrix multiplication

B Reshape tensor

C Element-wise addition

D Calculate mean

tf.matmul() performs matrix multiplication between two tensors.

Broadcasting helps by:

A Reducing tensor size

B Automatically expanding smaller tensors to match larger ones

C Changing tensor data types

D Sorting tensor elements

Broadcasting automatically expands smaller tensors to compatible shapes for operations.

Explain how broadcasting works in TensorFlow tensor math operations.

Think about how smaller arrays can be used with bigger ones without errors.

Describe the difference between element-wise multiplication and matrix multiplication in TensorFlow.

One multiplies matching elements, the other combines rows and columns.