8. Output¶
8.1. Output Coordinates¶
.new text file is created after the compensation process; it contains the compensated coordinates in the user-input projection. To convert .new into .cor in order to update initial coordinates, use Export to Cor tool .
If Normal Matrix Inversion was computed, compensated coordinates in .new file are followed by empirical half confidence intervals.
8.2. Subframes Files¶
All .xyz files are updated after compensation with a final comment containing:
transformation between the subframe and the global cartesian frame
measured points (=raw observations) in the global cartesian frame
measured points (=raw observations) in the projection frame (as in .new file).
This transformation is used by Apply Cartesian Transfo tool.
8.3. Computation Data¶
All computation data are stored in the .comp file, alongside project configuration.
This file is a JavaScript source that can be interpreted as JSON once its first line is removed, making it easily machine-readable.
8.4. Computation Report¶
The report can be displayed as a web page by opening the .comp.html file, which takes the data from the corresponding .comp file.
Note
The tables can be sorted by any column and are easy to copy-paste into spreadsheets, as units and other information are not selected.
8.4.1. Project Configuration¶
It contains the list of projects parameters and is equivalent to the Project Settings
form from the graphical user interface (see Project Configuration). It also contains a map of the project.
8.4.2. Computation Information¶
It represents general information about the computation. It is possible to hide or show the information messages with the corresponding buttons.
8.4.3. \(\sigma_0\) Evolution - \(\chi^2\) Test¶
The log scale graph represents the evolution of \(\sigma_0\) along the iterative process.
The \(\chi^2\) test limits are computed from the degrees of freedom of the least squares adjustment and they are also represented in the graph.
If Normal Matrix Inversion was computed, the biggest confidence ellipsoid is also given. This may help to detect a problem of external constraints that does not affect \(\sigma_0\).
8.4.4. Initial Coordinates¶
Initial coordinates are the coordinates of the points in the user-input projection after Parameters Initialization. They contain the point class, comment as infotip, a priori precisions of constraints and the number of active observations on each point.
8.4.5. Observations¶
The observations table contains all the observations in the given input order, with their .obs file name and line number as infotip and several indicators, such as:
distance between points
total \(\sigma\): combination of absolute and relative \(\sigma\)
residual, projected residual and normalized residual (initial residuals if before computation)
Normal Matrix Inversion mode adds more indicators:
a posteriori \(\sigma\): parameters precision propagated to the observation
standard residual: residual normalized by a posteriori \(\sigma\)
redundancy
and in case of sufficient redundancy:
standardized residual: residual divided by its own \(\sigma\)
\(\nabla\) (nabla): biggest non-detectable fault, internal fiability
probable error
Note
Some columns can be folded by clicking on them.
8.4.6. Residual Repartition¶
It is represented by a histogram showing how the residual are distributed by type, normalized residual values and observation distances.
Infotips are present on each bar to display more information.
8.4.7. Pseudo Random Propositions¶
To help fix observations’ a priori \(\sigma\), a factor is proposed for each type of observation (if sufficient redundancy). It is calculated from residuals and redundancy (see the infotips for more information).
The residuals’ repartition by distance helps determine if the absolute or relative \(\sigma\) has to be changed in the .obs file.
8.4.8. Biggest Residuals¶
Biggest residuals are displayed in an observations table containing only suspicious observations, ordered by decreasing residuals, with a maximum of 20 observations.
The anchors on the beginning of the lines link to the observation’s line in the table containing all the observations.
8.4.9. Similarities¶
This section shows the subframes (Cartesian Subframes or Polar Subframes), with vertical error and transformation between the subframe and the global cartesian frame.
8.4.10. Axes¶
This section shows the Rotation Axes with their parametrizations (from left to right: axis orientation, orientation precision, origin) and the list of their targets (circle parameter, number of positions).
8.4.11. Compensated Coordinates¶
The compensated coordinates are given in the user-input projection, with their displacements since initialization.
These coordinates are also exported in .new file (see Output Coordinates).
8.4.12. Points Empirical Confidence Indicators¶
These indicators are available after Normal Matrix Inversion.
They are given in the spherical computation frame, therefore there is no scale error, nor convergence of meridians.
Points empirical confidence indicators are confidence indicators scaled by final \(\sigma_0\) to take into account the potential a priori \(\sigma\) “optimism” or “pessimism” of the observations.
8.4.12.1. Confidence Ellipsoids¶
The confidence ellipsoids at \(1\sigma\) (confidence of 20% because of the three dimensions) are given as lengths and orientations for each half axis.
8.4.12.2. Half Confidence Intervals¶
Half confidence intervals are the projections of the confidence ellipsoids on each direction (68% confidence), corresponding to a posteriori \(\sigma\) of points coordinates. These \(\sigma\) are also exported in .new file (see Output Coordinates).
8.4.13. Points Displacements Simulation¶
This output is available only for the Monte-Carlo Simulation mode.