felixdivo/kernel_synth_annotated
KernelSynth (annotated) One million synthetic univariate time series, each 1024 points long, drawn from a Gaussian process prior whose kernel is a random composition of up to five base kernels. This is the KernelSynth procedure from Chronos with one addition: the generating kernel is kept alongside each series. The ground-truth structure behind every series is therefore known, which makes the corpus usable for interpretability work rather than only for pretraining.… See the full description on the dataset page: https://huggingface.co/datasets/felixdivo/kernel_synth_annotated.
KernelSynth (annotated)
One million synthetic univariate time series, each 1024 points long, drawn from a Gaussian process prior whose kernel is a random composition of up to five base kernels. This is the KernelSynth procedure from Chronos with one addition: the generating kernel is kept alongside each series. The ground-truth structure behind every series is therefore known, which makes the corpus usable for interpretability work rather than only for pretraining.
Fields
start— constant2000-01-01T00:00:00. The time index is arbitrary and carries no meaning.target— the series itself: 1024float64values sampled from the GP prior.selected_kernel_reprs— the base kernels drawn from the kernel bank, as scikit-learn reprs.kernel_formula— the composed kernel with explicit precedence, e.g.RBF(length_scale=1) * (DotProduct(sigma_0=1) + WhiteKernel(noise_level=1)). scikit-learn's own repr omits parentheses, so(a + b) * canda + b * care indistinguishable there; this field disambiguates them.
Generation
Produced by a script adapted from Chronos' `kernel-synth.py`, with these settings: 1,000,000 series, length 1024, at most 5 base kernels each, seed 1. Kernels are drawn with replacement from the 33-entry Chronos kernel bank and combined pairwise with random + / * operators. Each series draws from its own independent random stream derived from the seed, so the corpus is reproducible and independent of worker count.
One deviation from upstream: a jitter of 1e-8 * mean(diag(cov)) is added to the covariance diagonal to keep it numerically positive semi-definite, since composed kernels are often ill-conditioned.
License and attribution
The data is released under CC-BY-4.0.
It was produced by a script adapted from Chronos' kernel-synth.py (Copyright Amazon.com, Inc., Apache-2.0). That license covers the generator code, not the series it emits, and no Chronos data is contained here — every series is sampled fresh from a GP prior. The method, however, is theirs; please cite the Chronos (1) paper.