gwmodeling / challenge Goto Github PK

team members: Ed de Sousa, Rui Hugman, Mike Fienen, Nick Martin, Jeremy White
approach: ensemble of TFN models

Model: NHiTS
Names: Nikolas Benavides Höglund

I'm a bit late to the game, but I would like to give it a chance and provide a contribution for this challenge. The model I'm using is an implementation of NHiTS (link to paper). LUHG = Lund University HydroGeology.

Model: multi-frequency LSTM with MC dropout for uncertainty estimates
Names: Tim Franken

Note: I'm a bit late with my subscription but still plan / hope to get the submission in before the deadline (5/1, 24hCET) if that's ok

Model: LSTM

Members:
Morteza Behbooei
Jimmy Lin
Rojin Meysami

Model: Transformer
Member: Anna Pölz, Ali Obeid, Ahmad Ameen

Geological Survey of Sweden

Model: SGU-HYPE
Names: Anders Retzner

The model is based on the HYPE model (http://www.smhi.net/hype/wiki/doku.php?id=start) but with some added processes to improve groundwater level predictions.

@selinawaang registering the team here.

Model: Random forest ensemble
Name: Antoine Di Ciacca

Model: Hybrid deep-learning
Members

1. Sivarama Krishna Reddy Chidepudi
2. Abel Henriot
3. Nicolas Massei
4. Abderrahim Jardani

Model: Linear Regression with Distributed Lags
Member: Jonathan Kennel

Model: Random Forest model
Name: Jānis Bikše

I worked on this some time ago and was slow to polish it, but just noticed that Team Mirkwood has quite a similar approach. I hope it won't make any problems but definitely, it would interesting to compare the results.

Model: Mixed Effects Random Forest (MERF)
Names: Ayush Prasad (University of Helsinki)

Model: LSTM
Names: Raphael Schneider & Julian Koch

Model: HydroSight - lumped conceptual model
Names: Xinyang Fan, Tim Peterson

Please open a GitHub Issue to register your team.

Model: XX-model
Names: X.X. XYZ

Hi,

I just submitted my results for the challenge and submitted my results.

Model: 1D-CNN Deeplearning model
Team: AmirEtAl
please check and confirm if you have recieved.

Best,
Amir,

Discussed in #34

Im sure Im missing something but the forcing data csv files are constant in time - just one value for all times. Is that right? Like I said Im slow so maybe Im not understanding something...

Hello,

I'm planning to have a crack at this.

Matthew Taylor

Model: LSTM
Names: Max Rudolph, Alireza Kavousi (both Institute of Groundwater Management, TU Dresden, Germany)

Model: Ensemble of shallow learners
Names:

• Max Nölscher