Skip to content

Latest commit

 

History

History
44 lines (28 loc) · 2.08 KB

File metadata and controls

44 lines (28 loc) · 2.08 KB

Throughput Monitor

Tries to gamify Anki by breaking your study session into batches of time (ex: 5 minutes) through a countdown bar.

Using these batches, we

  1. Show a rolling average of how much you study in the given batch
  2. Shows how many cards you've studied in the current batch compared to your average
  3. If you surpass the rolling average, a flame will appear to show you that you're on fire!

We use this to also show how long it will take to complete your review at your current pace.

Settings

Countdown bar settings

  • timebox size (in seconds) of a study batches to consider for

  • countdown_colors color of the bar as the countdown gets closer to 0

  • invert_timer decides if countdown bar fills from left->right or right->left

  • countdown_timer_as_percentage whether or not to display the time left in batch or just the % time passed

Study time left settings

  • include_all_study_time_for_day whether to include the entire day's worth of study in the bar or just the current review

  • show_time_till_end show the exact time until you finish all cards

Point bar settings

  • penalize_idle if you got 0 points in batch, whether or not we should count it

  • exponential_weight decay for exponential weighted average

  • goal_offsethow many more reviews than the exponential weighted average you hope to get this round

  • initial_throughput_guess initial goal when you just started on a deck with no prior study history

  • bonus_points_by_card_type get different amount of points based off if this card is (new, learning, due, relearn)

  • number_batches_to_keep number of batches to use to calculate moving average

  • threshold_for_batch how many answers needed in a batches to be considered for the moving average

  • points_as_percentage whether or not to show % towards reaching moving average in batch

  • points_as_number whether or not to show how many points you have in the current batch

  • show_flame whether or not to show a flame when doing better than average

  • flame_height height in pixels of the fire (width is automatically adjusted to maintain aspect ratio)