Why np.tile() is used? | C2_W1_Lab02_CoffeeRoasting_TF

Course Q&A Machine Learning Specialization Advanced Learning Algorithms
1

I am wondering why the data is tiled this much, what exactly is the purpose of tiling (or, repeating) the data for 1000 times !?

Xt = np.tile(Xn,(1000,1))
Yt= np.tile(Y,(1000,1))   
print(Xt.shape, Yt.shape)

Link : Coursera | Online Courses & Credentials From Top Educators. Join for Free | Coursera

2

As we know, training a good machine learning model requires a sufficiently large dataset. If you have very little data and want to expand it with similar values, you can use tile() to artificially increase the dataset size.

1 Like
3

Thank you.