Search
Google Research Releases Upgraded TimesFM-3 Architecture for Multivariate Time Series Forecasting
15 hour ago /
Read about 0 minute
Author：小编
PingWest, September 2 news, according to explainx, Google Research released TimesFM-3 on August 31. This model enables accurate multivariate time series forecasting in a single forward pass, significantly outperforming other forecasting models across multiple mainstream benchmarks. Compared to version 2.5, the 3.0 version can jointly predict multiple related signals at once and autonomously infer relative changes between variables. The model weights have been uploaded to Hugging Face, and Omar Sanseviero, Head of Developer Relations at Google DeepMind, promoted the release on X on September 1.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
