From f560005d21d888c1b2b11d76cf088acc1a33cdd7 Mon Sep 17 00:00:00 2001 From: liangs6212 <80952198+liangs6212@users.noreply.github.com> Date: Thu, 27 Jan 2022 11:03:48 +0800 Subject: [PATCH] Chronos: add `NBEATSForecaster` API (#3880) * nbeats first version * add nbeats_forecaster API * Split automodel and forecaster * fix typo * modified params description * pep8 * add nbeats doc * fix typo and readthedocs error * fix known bugs * fix doc typo * rename NBEATS to NBeats * fix doc typo * pep8 * fix errors in ut * generic not support distributed * pep8 --- .../source/doc/PythonAPI/Chronos/forecasters.rst | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/docs/readthedocs/source/doc/PythonAPI/Chronos/forecasters.rst b/docs/readthedocs/source/doc/PythonAPI/Chronos/forecasters.rst index 41858d1c..da1651c9 100644 --- a/docs/readthedocs/source/doc/PythonAPI/Chronos/forecasters.rst +++ b/docs/readthedocs/source/doc/PythonAPI/Chronos/forecasters.rst @@ -42,6 +42,19 @@ Temporal Convolutional Networks (TCN) is a neural network that use convolutional :show-inheritance: +NBeatsForecaster +---------------------------------------- + +:strong:`Please refer to` `BasePytorchForecaster `__ :strong:`for other methods other than initialization`. + +Neural basis expansion analysis for interpretable time series forecasting (`N-BEATS `__) is a deep neural architecture based on backward and forward residual links and a very deep stack of fully-connected layers. Nbeats can solve univariate time series point forecasting problems, being interpretable, and fast to train. + +.. automodule:: bigdl.chronos.forecaster.nbeats_forecaster + :members: + :undoc-members: + :show-inheritance: + + TCMFForecaster ----------------------------------------