ML in Google Products

在這個篇章 強調的部分在於 一個產品或是大系統 不可能靠單一ML就完成。

影片中的例子
[情境]

You might look at a
business problem, say, how to forecast whether an item will go out of stock and think of it as a single machine learning model that you have to build.In practice, you will have to build many ML models to solve this problem.

first model : predict demand for the product at the store location
second model : predict the inventory of this item at your supplier’s
warehouses and nearby stores
third model : predict how long it is going to take them to stock your product and use this to predict which supplier you will ask to refill the shelf, and when

對於 milk 和 對於 乾燥食品 的model 也會有所不同，我推測是因為 賞味期的長短不同。

[觀念]
Avoid the trap of thinking of building monolithic, one-model-solves-the-whole-problem
solutions.

Ref :
Google Team
Google ML Study Jam