Published March 6, 2024 | Version v2
Model Open

DeepPhenoMem V1.0: Deep learning modelling of canopy greenness dynamics accounting for multi-variate meteorological memory effects on vegetation phenology

Authors/Creators

  • 1. ROR icon Max Planck Institute for Biogeochemistry

Description

This repository contains code aimed at developing LSTM (Long Short-Term Memory) models (named DeepPhenoMem V1.0) for predicting vegetation phenology. The tasks include training LSTM models, tuning hyperparameters, and predicting with test datasets. 

Contents:

  1. UserManual.md: This document is the guidline to run the LSTM model.  

  2. lstm_train.py: This script is responsible for training the LSTM models. It involves feeding the prepared data into the model and optimizing the model parameters using techniques such as Adam optimization.

  3. lstm_hp_tuning.py: Here, we tune the hyperparameters of the LSTM model to optimize its performance using grid search.

  4. lstm_pred.py: Finally, this script is used for making predictions with the trained LSTM models on unseen test datasets. It loads the saved model weights and applies them to new data to generate predictions.

  5. example: This folder contains an example demonstrating how to run the DeepPhenoMem V1.0 model.

Files

DeepPhenoMemV1.0.zip

Files (9.6 MB)

Name Size Download all
md5:33b067b5ca3153fccc46a8bc87923e47
9.6 MB Preview Download