Doubt in numpy vstack

Question

I wanted to ask that the following code doesn't work

c=vstack(tsne_data,label_1000).T

where shape of tsne_data is (1000,2) and the shape of label_1000 is (1000,)

but the following code works for the same

c=vstack(tsne_data.T,label_1000).T

for the same shape of tsne_data and label_1000

so I wanted to know that why do we need to use transpose with tsne_data in vstack

Answer 1

The vstack function in numpy will stack the sequence vertically (row wise).

If you see the below example shape of the row in a and b are not similar so you won't be able to add b to a.

In order to do that you will have to find the transpose of it so that shape of rows in a and b are similar