Luc Mcgrady

Good point. Hopefully by using a priority queue and iterating through the cards in the order that they are reviewed, changing the review priorities will be possible.

I think that's better? I think it was treating new cards as review cards at certain points. ``` left: [0, 13, 23, 30, 58, 81, 83, 81, 86, 88, 88,...

Well on my computer at least: Current: ``` optimal_retention time: [274.85 ms 278.54 ms 282.14 ms] ``` This pr: ``` optimal_retention time: [35.805 ms 36.197 ms 36.589 ms] change: [-87.228%...

https://github.com/Luc-Mcgrady/anki-10k-notebooks/blob/3d38effd7f5b0c2a72eb57079f1c719b77e11d11/fatigue.ipynb here's a Notebook I set up to test that. I'm not sure if its flawed though.

The notebook uses FSRS-5 Non-recency. https://github.com/open-spaced-repetition/srs-benchmark/blob/main/result/FSRS-5.jsonl It does seem wrong. ``` Average All cards per day 23.86522309711286 Average Loss = 0.8822175277504851 ```  ``` "RMSE(bins)": 0.114211, "ICI": 0.100382, "AUC": 0.56959},...

> You've forgotten to update the label-mappings. In current implementation, `again` is mapped to zero, while `hard`, `good` and `easy` are mapped to one. If the user abused `hard`, `hard`...

As has been explained in this thread and for reasons that were explained to me at length which I have since somewhat forgotten, this option ignores **cards** which are **introduced**...

"Ignore cards introduced before"?

"Ignore introduced before"?? little less clear

> such as preloading the next card's resources? I've made an attempt at this in the above linked commit. (https://github.com/Luc-Mcgrady/anki/commit/6d35ce61bed1a9b07c8c442f1eca61618873a407)