22.02.2024 Views

Daniel Voigt Godoy - Deep Learning with PyTorch Step-by-Step A Beginner’s Guide-leanpub

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Output

torch.Size([5, 5, 3, 3])

Its shape indicates it produced five filters for each one of the five input channels

(25 filters in total), and each filter is 3x3 pixels.

fig = sbs_cnn2.visualize_filters('conv2')

Figure 6.13 - Visualizing filters for conv2 layer

Learning Rates

It is time to have "the talk." It cannot be postponed any longer—we need to talk

about choosing a learning rate! It is no secret that the learning rate is the most

important hyper-parameter of all—it drives the update of the parameters; that is,

it drives how fast a model learns (hence, learning rate).

444 | Chapter 6: Rock, Paper, Scissors

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!