Ryan Hammonds

Thanks for the follow up @rdago. All of this makes sense. I ran tests and don't think one model is consistently more numerically stable than the other. There's tradeoff in...

cfa8a2b adds `skewed_gaussian_function`. It takes the same parameters as `gaussian_function`, with an additional alpha parameter to control the skew. This implementation multiplies the normal gaussian by a skewed cumulative distribution...

I ran into a similar issue, and lowering gamma worked for me. However, an alternative I found was to use scipy.signal.resample to downsample. My timeseries had 800 values and downsampling...

Cool, this looks like it track's important/relevant metadata for new models. What is the plan for handling parameter guess procedures? I guess that's kinda related to #291. I guess for...

I can merge it but it looks like something in the CI pipeline is failing. I'm in favor of deprecating support for conda / conda forge. I've never used conda...

Sounds good. I imagine people who try to install from conda forge do so because of the instructions on the readme. But maybe I'm underestimating how popular conda forge is....

I added you as a maintainer to the conda forge repo. I re-triggered the CI and the error being thrown is: `ModuleNotFoundError: No module named 'matplotlib'`. Seems like a test...

Hi @TomDonoghue, this looks super useful. I was able to run the example you provided. The only thing that confused me at first was the shape of spectrograms. The docstring...

This looks really cool/useful! In the last case you gave, could the super class be SpectralModel? Or what is the benefit of using the base class + the type magic...

Cool! I agree with you that this is useful and in the direction of being more flexible. The distinction between the base and model/api is nice. I think it would...