That would be a rough estimate of how much room for reduction there is.

... it's made a 40MB exe ...
This does not include the weights file which is ~300MB. The weights file contains the model data used to encode and decode the audio and is a necessity, and the model is not interchangeable. You have to use the same model to decode any encoded file.

Quote from: SimBun on 2023-09-17 09:30:58

The section you highlighted is the code that computes and rounds the rating, so if you wanted to change the logic it would go there.
If you run the following code it'll show you everything you could reference in your calculation - although likely it's just length_seconds that would be of interest.

Finally I've been thinking about averaging the rating weighted on track length (sometimes I do have instrumental only tracks which can be >3 mins on which i defined a rating of 1)
I tried to play with length_seconds and with the sum() function to get total album length but without succes so far, still investigating 

sum(length_seconds) as name_of_calculated_column

  select album,
         [album artist],
         date,
         round(avg(rating), 2) albumrating,
         sum(length_seconds) as total_album_length_seconds
  from Playlist
  where playlist_index = active_playlist()
        and rating <> 1
  group by 1, 2, 3;

The thing about a signal that is a single sine, is that you can predict them by two coefficients of which one is = -1.
You have a predictor x(N+1) = x(N) * 2 cos q - x(N-1) and with q fit to pi* 2f/F where f is the sine's frequency and F (>2f) is the sampling frequency, then you are all good - provided that the format can offer high enough precision for that 2 cos q.

• Create up to 10 themes from the current configuration (name have to be unique)
• Choose a theme amongst a list and change the current configuration
• Update a theme with the current configuration
• Rename a theme (cannot duplicate name)
• Delete a theme