- Neural Networks with Keras Cookbook
- V Kishore Ayyadevara
- 119字
- 2021-07-02 12:46:35
Defining weights for rows
In the Predicting house prices recipe, we learned about defining a custom loss function. However, we are not in a position yet to assign a higher weightage for certain rows over others. (We did a similar exercise for a credit default prediction case study where we assigned higher weightage to one class over the other; however, that was a classification problem, and the current problem that we are solving is a continuous variable-prediction problem.)
In this section, we will define weights for each row and then pass them to the custom_loss function that we will define.
We will continue working on the same dataset that we analyzed in the Stock price prediction recipe.
推薦閱讀
- Learning Cython Programming
- TypeScript Blueprints
- Oracle Database In-Memory(架構與實踐)
- 精通網絡視頻核心開發技術
- Nexus規?;疭crum框架
- UML 基礎與 Rose 建模案例(第3版)
- Windows Phone 7.5:Building Location-aware Applications
- 從零開始學C語言
- CRYENGINE Game Development Blueprints
- Getting Started with Python
- Java Web開發實例大全(基礎卷) (軟件工程師開發大系)
- Appcelerator Titanium:Patterns and Best Practices
- Mastering SciPy
- PostgreSQL 12 High Availability Cookbook
- Android 5從入門到精通