Commit 822199a4 authored by Mike Bedington's avatar Mike Bedington

Marginally more in the readme file

parent e4b11aa0
A set of functions for modelling the river input into
A set of functions for modelling the river input into a FVCOM
river_dict is then a dictionary of the rivers used in the domain
......@@ -6,22 +6,31 @@ river_dict is then a dictionary of the rivers used in the domain
Neural network flow models
--------------------------
There is an option to
WRF data
--------
WRF data used are the variables T2, RAINNC, and TIMES
Th neural network training functions are in nn_functions.py. The python keras package is used for modelling. The nn design used is a Sequential model with two layers (Dense layers, the standard fully connected nn layers), one layer of width and one with a single output node. The loss function used is mean squared error and the Adam optimiser.
The training data is a series of lagged series and window summed . If I could get reccurent nns working properly in Python I probably should use them instead.
An example shell script for using the ncecat program to do this is included.
The WRF grid data comes from a file
WRF data
--------
WRF data used are the variables T2, RAINNC, and TIMES. It is setup to take th hey are taken from monthly netcdf files. An example shell script for using the ncecat program to do this is included.
The WRF grid data comes from a file
CEH river gauge data
--------------------
The UK rivers are modelled on the
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment