-
We are cooperatively working on a number of models and the commit history is proving to be very useful. However, the user is always reported as "anon". Would it be possible, for SQLite DBs (where ther…
-
As discussed in the Spine Engine meeting, it could be interesting to create a new project item to embed a Jupyter Notebook that could be used both for data cleaning or for data visualisation.
The not…
-
In order to test SpineOpt against the RTS test system and also to increase SpineOpt's visibility, we should make it possible to run RTS test system in SpineOpt and include SpineOpt in https://github.c…
-
# Description of the bug
In the Spine DB Editor, the parameters *upward_reserve* and *downward_reserve* for the node entity should be parsed as boolean rather than string when the user enters the v…
-
The MIP optimality gap is a very important metric for juding the quality of a result and, in particular, is important to know when comparing the results from two or more runs.
This is also useful f…
-
The proposed database re-structure might need a reworking of current DB editor layout. Currently the concept of objects and relationships is baked into the design with seperate trees and parameter val…
-
Here's a mock-up for creating new entities. It was motivated by the need to support abstract classes (there is a need to show what is the subclass in order to distinguish entities with the same name b…
-
Currently the `type` and `default_type` fields in `parameter_value` and `parameter_definition` tables are set to null if the value is string, float or boolean. It works for now, but would break if we …
-
We can dump spine dbs into a JSON and back, mostly. Which means we could use a JSON file instead of a SQLite file for version control purposes.
The first way of doing this that came to mind was to …
-
## Current issues
- visual clutter for complex models
- layout is unpredictable/updates while working on it
## Resolutions/Feature request
- aggregate nodes to simplify the view, also useful whe…