@karllark had a potentially great idea. Instead of using peak intensity as the fitting parameter, use integrated intensity as the varying parameter. This is somewhat more complex to specify in the model, but may greatly reduce the cost of "tying" parameters together, for example when fitting two segments which contain the same line.
This can be done by changing to line profile models that use integrated intensity instead of central intensity as one of the parameters. See the notebook in #158 for an example of how to do this.
@karllark had a potentially great idea. Instead of using peak intensity as the fitting parameter, use integrated intensity as the varying parameter. This is somewhat more complex to specify in the model, but may greatly reduce the cost of "tying" parameters together, for example when fitting two segments which contain the same line.