Chronos
A time series foundation model proposed by Amazon, belonging to a family of pretrained models and usable for zero-shot time series forecasting.
Timeline (last 90 days)
Time series foundation models (TSFMs) refer to models pretrained on large-scale multi-domain time series data that can be transferred to downstream forecasting tasks via zero-shot or lightweight fine-tuning; representative works include Chronos, TimesFM, Moirai, and others.
Amazon's Chronos discretizes continuous numeric series into tokens by quantile binning, then reuses the language model's Transformer architecture for forecasting.
Time-Series Foundation Models are general-purpose forecasting models pretrained on large-scale, multi-domain time-series data with cross-task generalization capabilities, with representative works including TimesFM, Chronos, and Moirai.
Representative time-series foundation models include Google's TimesFM, Amazon's Chronos, and Salesforce's Moirai.
All Facts (4)
Time series foundation models (TSFMs) refer to models pretrained on large-scale multi-domain time series data that can be transferred to downstream forecasting tasks via zero-shot or lightweight fine-tuning; representative works include Chronos, TimesFM, Moirai, and others.
50%UnverifiedAmazon's Chronos discretizes continuous numeric series into tokens by quantile binning, then reuses the language model's Transformer architecture for forecasting.
50%UnverifiedTime-Series Foundation Models are general-purpose forecasting models pretrained on large-scale, multi-domain time-series data with cross-task generalization capabilities, with representative works including TimesFM, Chronos, and Moirai.
50%UnverifiedRepresentative time-series foundation models include Google's TimesFM, Amazon's Chronos, and Salesforce's Moirai.
50%