← Deep learning
Lesson 3 of 5
Convolutional networks
Watch, read the card, then practise. Convolutions · Pooling
The video for this lesson is being recorded.
Meanwhile, the notes below and the 60-second cards cover the same ideas.
Practise this topicConvolutions
A convolution slides a small filter over an image, computing a weighted sum at each spot to detect a pattern.
Remember: Filter count = how many patterns a layer looks for.
Pooling
Pooling shrinks a feature map, keeping the strongest signals.
Remember: Pooling has no weights to learn.
Try the example in Code Lab
size, f, pad, stride = 28, 3, 0, 1 print((size - f + 2 * pad) // stride + 1)