A Recipe for Training Neural Networks


Wow. Andrej Karpathy doesn't post often, but when he does it's worth reading. This most recent post is golden. He starts by sharing why all of the one-line tutorials on model training are just not how training networks happens in real life. He then goes deep on how that process should actually go. In the process, he drops wonderful little tidbits like:

Now that we understand our data can we reach for our super fancy Multi-scale ASPP FPN ResNet and begin training awesome models? For sure no. That is the road to suffering.

Incredible resource from one of the leading practitioners in AI today.


Want to receive more content like this in your inbox?