Comments (1)
您好,这个问题很关键,我是这么理解的
(1)目前多变量预测是首先使用一层embeding网络将变量维度C映射为深度特征的通道维度d_{model},这个过程会使得不同变量之间的信息相互融合,因此在TimesNet中使用fft计算的已经是「融合后特征」的周期性的,这个周期性已经隐式包含了不同变量的信息。同时TimesNet中还使用了多周期性的设计,用于覆盖不同长度的周期。因此,我觉得目前的设计是可以比较好的覆盖多元时间序列不同频率的。不过,目前这种平均的设计也是trade-off performance and efficiency之后的结果,如果做更加细粒度的建模是会得到更好的结果的。
(2)最近也有很多文章探索了variate-independent的设计,例如Dlinear、PatchTST,不过他们也都是将模型在不同变量间share。总体来看,目前多元时间序列预测上,不同变量的差异性目前探索还比较初步。
from time-series-library.
Related Issues (20)
- Something wrong on short_term_forecast HOT 4
- Insufficient GPU memory for the long term forecast task HOT 1
- 想问一下为什么我自己的数据 只有两列,会出现Nan这种情况 HOT 2
- 无法复现结果,用了torch1.7.1也一样不行
- add SCINet into baseline
- Non-overlapping sliding window for anomaly detection experiments HOT 1
- 分类实验结果无法复现 HOT 2
- multivariable test results HOT 1
- Error in Timesnet Short term Forecasting Model Training (modelConfig)
- CPU-heavy operation slowing training HOT 3
- Would you mind add "One Fits All: Power General Time Series Analysis by Pretrained LM“ HOT 1
- May I ask if the anomaly detection methods implemented in tslib need to be classified? Are they based on forecasting or reconstruction? HOT 1
- How to feed my sensor dataset into the TimesNet model for classification task HOT 1
- 为什么认为将所有的seq_len设置为96是一种公平的表现 HOT 1
- Add the new models to leaderboard HOT 1
- About the parameters features and target in the code
- Question about TimesNet HOT 1
- Clarifications on exp_anomaly_detections.py HOT 1
- Add TimesFM to the library
- An inquiry about TimesNet HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from time-series-library.