Shouldn't X be (m,n) matrix when vectorizing neural network

Course Q&A Deep Learning Specialization Neural Networks and Deep Learning
1

When vectorizing calculation for neural networks for entire set of m samples, X is mentioned as an (n, m) matrix. The equation for calculation of Z is Z = transpose(W) * X + B where W is (m, 1) matrix, transpose(W) will be (1, m). So, shouldn’t X be (m, n) matrix. Otherwise, matrix multiplication itself will not work.

2

Hi @Gaurav_Malhotra, here’s a thread to clarify your query. Thanks.

1 Like
3

Also, it’s a good habit to check your query once over the DeepLearning.AI community platform. Most of the times, a similar kind of question and solution has already being asked and discussed, which saves a lot of your own time to further your learning. Thank you!