Introduced on day 27
Zero-shot forecast
A forecast from a pretrained model that is applied to a series it was not trained on, using only that series' history as context and no fitting on it.
In detail
Zero-shot describes how a model is used, not how well it works. The weights are fixed; the series supplies the context window. It still needs a backtest against strong baselines on your origins, and a check that the pretraining corpus did not include the evaluation data.