Kazuma Iwase and Tomoyuki Takenawa
Graduate School of Marine Science and Technology, Tokyo University of Marine Science and Technology, 2-1-6 Etchujima, Koto-ku, Tokyo, 135-8533, Japan.
Contributing authors: kazumaiwase676@gmail.com; takenawa@kaiyodai.ac.jp;
Observation data (observe.csv) is what we pre-processed from data downloaded from the JMA (Japan Meteorological Agency) website (https://www.data.jma.go.jp/risk/obsdl/index.php).
There are csv files (dummy_nesw.csv, dummy_tenki_to_kurasu.csv) in the data folder, but they are dummy data. Note that they are not the data used in this research. Dummy data was created by adding an error to the observed data so that the RMSE after 8 hours match the actual forecast data.
You can run some notebooks in notebooks folder on googlecolab.
experiment_models.ipynb: Comparison of various models for predicting Temperature at Mt.Fuji and Precipitation at Hakone. Forecast times can be selected from 2h and 8h ahead.
training_lightgbm.ipynb: LigtGBM training for predicting Temperature at Mt.Fuji and Precipitation at Hakone. Forecast times can be selected from 2h, 7h, 8h and 9h ahead.
training_nofuture.ipynb: LigtGBM(that do not use future forecast data) training for predicting Temperature at Mt.Fuji and Precipitation at Hakone. Forecast times can be selected from 2h, 7h, 8h and 9h ahead.