Placeholder chapter page. Sections are produced by the book-skills pipeline.
Sections
- 19.1 What makes a model a "foundation" model for time series
- 19.2 Pretraining corpora and tokenization (patching vs value tokenization)
- 19.3 Model families: TimesFM, Chronos/Chronos-Bolt, Moirai/Moirai-MoE, MOMENT, Time-MoE, TTM/TinyTimeMixers, Lag-Llama, Toto, UniTS, Timer
- 19.4 Zero-shot and few-shot forecasting, classification, imputation, anomaly detection
- 19.5 Foundation-model embeddings as features for downstream tasks
- 19.6 The evaluation crisis: leakage, non-stationarity, and honest benchmarking (GIFT-Eval, Monash, LOTSA)
- 19.7 When a foundation model helps and when a small model wins
Lab 19
apply a pretrained TSFM zero-shot to a new sensor dataset; compare against a tuned task-specific model and a fair-baseline check.