AIML Challenges: Difference between revisions

From CSDMS
No edit summary
No edit summary
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
== The AI&ML Challenge Dataset ==
== Earth Surface Dynamics Modeling: AI&ML Training Data ==
=== The Seabed Offshore of NE USA ===
The idea here is to serve datasets that can be used to compare the performance of different
<gallery>
ML methods, such as Neural Networks, Random Forest and their variants. The data can
SeabedStack_v01.png|500px|Example outputs of mapped Seabed Character
also be used for education in the subject.
</gallery>
=== 1. The Seabed Offshore of NE USA ===
The Challenge Dataset may be found at the [[Data:AI&ML_Challenge_Dataset|'''CSDMS Substrates Dataset page''']].
<div class="row" style="background-color:#F5F5F5">
The Zipfile holds Documentation, Training Data and a Starter Python program.
<div class="col-sm-6"><br>
This Dataset may be found at the [[Data:AI&ML_Challenge_Dataset|'''CSDMS Substrates Dataset page''']].
The Zipfile holds Documentation, Training Data and a Starter Python program. The data come from the system 'dbSEABED' ([http://instaar.colorado.edu/~jenkinsc/dbseabed/ '''link''']).


With questions, please email Dr CJ Jenkins at INSTAAR.
Any questions: email Dr Chris Jenkins (chris.jenkins@colorado.edu) at INSTAAR, University of Colorado Boulder.
</div>
<div class="col-sm-6">
<gallery widths="320px" heights="300px" >
SeabedStack_v01.png||Example outputs of mapped Seabed Character</gallery>
</div>
</div>

Latest revision as of 13:21, 16 August 2019

Earth Surface Dynamics Modeling: AI&ML Training Data

The idea here is to serve datasets that can be used to compare the performance of different ML methods, such as Neural Networks, Random Forest and their variants. The data can also be used for education in the subject.

1. The Seabed Offshore of NE USA


This Dataset may be found at the CSDMS Substrates Dataset page. The Zipfile holds Documentation, Training Data and a Starter Python program. The data come from the system 'dbSEABED' (link).

Any questions: email Dr Chris Jenkins (chris.jenkins@colorado.edu) at INSTAAR, University of Colorado Boulder.