TSC-FM Benchmark
In-Context Learning
Standard Univariate results available

TimEE

An end-to-end time series classifier pretrained on synthetic tasks that predicts from labeled support examples without target-specific training.

PaperCodeCheckpoint

Using TimEE for time series classification

These scores come from an independent reproduction of the released checkpoint with its default four-member preprocessing ensemble and full training splits. Standard Univariate covers all 128 UCR datasets. Multivariate and low-shot results are unavailable in this benchmark; the reproduced scores are not substituted with the authors' paper values.

Evaluated configurations

Default four-member preprocessing ensemble

Published evaluation coverage

  • Standard: Univariate
  • 5-shot: No published results in this setting.
  • 10-shot: No published results in this setting.
Organization
University of Freiburg · Zuse School ELIZA Darmstadt · Prior Labs · ELLIS Institute Tübingen
Parameters
4.5M
Checkpoint used
Official univariate checkpoint
Input Type
Univariate time series
Inference
Training-free in-context classification

TimEE classification benchmark results

Each row is a separate configuration; coverage is reported per row. Compare within the same setting and scope on the classification leaderboard. Ranks use the full comparison set. See the ranking definitions and budget-specific dataset rules before comparing scores across tables.

Standard · Univariate

Comparison set: 20 configurations

TimEE — Standard Univariate time series classification benchmark results
ConfigurationUnivariate RankUnivariate Accuracy (%)Avg. RankCoverage
TimEE183.764.98128 / 128

TimEE paper, code and checkpoint