iT邦幫忙

第 11 屆 iT 邦幫忙鐵人賽

DAY 26
0
Google Developers Machine Learning

ML Study Jam Journey系列 第 26

Day 26 Art and Science of Machine Learning (cont.)

Multi-class problems

  • one vs all
  • one vs one

Use one softmax loss for all possible classes

Dense representations - Inefficient in space and compute

Embedding

  • Feature columns (like layers)
  • latent features
  • Custom Estimator

Keras Models

  • High-level deep neural networks library (supports multiple backends)
  • Fast prototyping

上一篇
Day 25 Art and Science of Machine Learning (cont.)
下一篇
Day 27 Art and Science of Machine Learning (cont.)
系列文
ML Study Jam Journey30

尚未有邦友留言

立即登入留言