Time Anchor ModernBERT 32M A compact time series model built on a 32M parameter ModernBERT backbone (33.4M parameters in total with the time series input/output heads). One call returns three things: 📈 Quantile forecast Probabilistic forecasts at any requested quantile levels 🧭 Variable / time impact Normalized contribution of every input variable at every time step 📌 Anchor forecast Forecasts conditioned on known future points you specify Installation predict time anchor accepts a Hugging Face Hub model id or a local checkpoint directory. Source code for the package lives at K Iwa/time anchor. Quantile Forecast Hourly temperature in Tokyo (Open Meteo archive), 64 hour holdout: median MAE 0.57 °C, with every actual value inside the q10–q90 band. Variable / Time Step Impact Per hour contribution of each weather variable over a one week Tokyo context. For each time index , impact across all variables sums to 1. Impact Validation The impact scores were checked against synthetic data where the correct answer is known in advance. Test 1 — do the shares match known mixing weights? The target is built as a weighted sum of three sine waves, and the same three waves are passed in as the…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy