LTSF-Linear
Crossformer
Our great sponsors
LTSF-Linear | Crossformer | |
---|---|---|
1 | 1 | |
1,783 | 355 | |
7.3% | 13.0% | |
4.5 | 3.7 | |
3 months ago | 5 months ago | |
Python | Python | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
LTSF-Linear
-
[D] Transformer effectiveness for time series forecasting (doubts)
Found relevant code at https://github.com/cure-lab/DLinear + all code implementations here
Crossformer
-
The Transformer Blueprint
Great write up. There were a few grammatical errors - I’d suggest piping it through a LLM prompted to find grammar errors :-)
Another area of interesting research is the use of transformer like models for highly dimensional time series prediction. While language and vision are interesting and have their uses, my opinion is the application of these techniques for multidimensional non linear effects in time series may ultimately have more broad and significant impact. Ex:
https://github.com/Thinklab-SJTU/Crossformer
What are some alternatives?
tidle-brain - TiDLE time-series forecasting in Tensorflow (Google. paper 2023)
sktime-dl - DEPRECATED, now in sktime - companion package for deep learning based on TensorFlow
ETSformer - PyTorch code for ETSformer: Exponential Smoothing Transformers for Time-series Forecasting
PyPOTS - A Python toolbox/library for reality-centric machine/deep learning and data mining on partially-observed time series with PyTorch, including SOTA neural network models for science tasks of imputation, classification, clustering, and forecasting on incomplete (irregularly-sampled) multivariate time series with NaN missing values/data.
flow-forecast - Deep learning PyTorch library for time series forecasting, classification, and anomaly detection (originally for flood forecasting).
gluonts - Probabilistic time series modeling in Python