DataTest (Module)
Compare calculated data from modules with external or internal references
Overview
DataTest
is predominantly used in Dissolve’s system tests for checking calculated one-, two-, or three-dimensional properties against external reference data. It also allows comparison between datasets within Dissolve, for example when ensuring the correctness and consistency of two different methods calculating the same thing.
Keywords
Test
Keyword | Arguments | Default | Description |
---|---|---|---|
Data1D |
TargetData Data1DFileAndFormat |
– | Filename and format of external one-dimensional data to be read in and tested against the named internal TargetData . This keyword may be specified multiple times in order to test many datasets sequentially. |
Data2D |
TargetData Data2DFileAndFormat |
– | Filename and format of external two-dimensional data to be read in and tested against the named internal TargetData . This keyword may be specified multiple times in order to test many datasets sequentially. |
Data3D |
TargetData Data3DFileAndFormat |
– | Filename and format of external three-dimensional data to be read in and tested against the named internal TargetData . This keyword may be specified multiple times in order to test many datasets sequentially. |
ErrorType |
ErrorType |
Euclidean |
Error metric to use when comparing the datasets within this module instance. |
InternalData1D |
TargetDataA TargetDataB |
– | Target internal datasets TargetDataA and TargetDataB to compare and calculate error. This keyword may be specified multiple times in order to test many datasets sequentially. |
SampledDouble |
TargetData double |
– | Target internal SampledDouble data TargetData and the real value to compare it against. This keyword may be specified multiple times in order to test many datasets sequentially. |
SampledVector |
TargetData ValueFileAndFormat |
– | Filename and format of external value data to be read in and tested against the named internal TargetData , which is a SampledVector . This keyword may be specified multiple times in order to test many datasets sequentially. |
Threshold |
double |
5.0e-3 |
Threshold value for error, above which the test fails. |
Last modified March 10, 2023: Web docs styling (#1350) (43924dc)