Sign inSign up

omslab/m4

By omslab

Updated about 3 years ago

Multi-Model Machine-learning Metasystem (M4) for operational water supply forecasting

Image
0

1.4K

omslab/m4 repository overview

Overview

The Multi-model Machine-learning Metasystem (M4) is the prototype next-generation mathematical and software model for operational water supply forecasting (seasonal river flow volume prediction) built for and employed by the Snow Survey and Water Supply Forecasting (SSWSF) Program of the US Department of Agriculture (USDA) Natural Resources Conservation Service (NRCS).

Usage

1. Download and extract an example data set (Run control file + training data)
$ mkdir ex
$ cd ex
$ wget https://csip.engr.colostate.edu/data/MMPE_RunControlFile.txt
$ wget https://csip.engr.colostate.edu/data/MMPEInputData_ModelBuildingMode.txt

2. Run M4
$ sudo docker run --rm -it -v $(pwd):/work omslab/m4

Output:

[1] "Running genetic algorithm for optimal selection of input variables and PCA modes for current model..."
[1] "principal components regression:    binary chromosome =  0 0 1 1 1 0 ,    trial cross-validated RMSE =  29765.682"
[1] "principal components regression:    binary chromosome =  0 1 1 0 1 1 ,    trial cross-validated RMSE =  28836.476"
[1] "principal components regression:    binary chromosome =  1 1 0 0 0 1 ,    trial cross-validated RMSE =  30782.966"
[1] "principal components regression:    binary chromosome =  0 1 0 1 0 1 ,    trial cross-validated RMSE =  30612.281"
...

Tag summary

Content type

Image

Digest

sha256:9ad448fd8

Size

511.5 MB

Last updated

about 3 years ago

docker pull omslab/m4