I have designed the following chart to visualize my daily progress in wanikani. It shows the number of items in the different states and the number of kanji (white line). Unfortunately I began only one month after my start in wanikani, so I am missing the ‘historical’ data of my beginnings. I hope this will be an impressive chart when I reach the mystic level 60 one day.
The chart is generated in a LibreOffice Calc file, which I manually update after each wanikani session (at least after the last session of the day).
I haven’t checked the api myself, but you can ask WK about all past reviews all the way back to sometime in 2017. So those data aren’t lost.
Edit: someone else did something similar, you can check their code to see how to access it
[WebApp] WaniKani Timemachine!
Way back in December I saw a post sharing a Google spreadsheet where someone had done something the same, but automated. I copied their spreadsheet setup, and it has been going great so far. With Google Sheets you can get your own JS code to run every set interval (i.e. day) and fetch the latest data from WaniKani using the API – which reminds me, I need to update it to work with API v2 before v1 gets the plug pulled!
(A quick search shows me that the creator of the setup was probably Kumirei, but I couldn’t find the original post.)
Here’s a couple of screenshots of the graphs:
I have used this, its really well-made and real useful! Have a look
Can you tell me where to find it?
It’s in @Naphthalene’s link
This is the direct link:
I like it, but I would flip it so apprentice is on the bottom of the stack…enlightened, burned on top. The reason is, at equilibrium flow, each SRS level should stay approximately the same. Some go out, some go in, at about the same rate you’re doing lessons. So I should see a ‘layer cake’ of flat lines and it will be visually easier to see anomalies.
(you can sort of see that already, the apprentice layer keeps the same thickness more or less, just it’s harder to judge when it’s sloping like that on top.)
Thanks. I had not seen that @Naphthalene had edited his reply, adding the link. The link has helped me to fill the gap between my start at wanikani and my start of data collection.
Yes, that’s right. I have now flipped the stack and I see that I should have more masters (or enlightened) items.
It should* settle out in the following proportions:
* but it won’t, because failed items. Enlightened will always be a bit thin, and guru/master a bit fat. Apprentice varies way too much to judge because it depends so much on how irregularly you do the lessons.
Whoa, that’s exceptionally cool. I’d never seen that before.
Did… did you just sit down in October and do one and a half thousand reviews in one go?
Yeah pretty much.
That’s what happens when you don’t do any for seven months.
Heh, I’m just ticking over seven months since my last review too, though I’m not intending to do them all in one go. There’s 3885 of them.
They really should count as immediately burned if you still remember after that long, regardless of current stage.
The hills are alive… with the sound of kanji…
With the songs they have sung for one thousand six hundred and sixteen years?
The first version of the timemachine (which is still online, just type the URL without the “2” at the end) has apprentice on top and the second version on the bottom. Ideally this is something that we should be able to adjust ourselves (hide/show certain SRS level, sort them differently, show number of review items at different times, etc). But the timemachine app seems to have been abandoned, so we can’t do this.
I would actually dig a version of this that wasn’t number of items in each stage, but number of reviews I did that day in each stage. The data is all there in WaniKani heatmap, but would take a lot of clicks and manual entry, which I haven’t gotten up the energy to do yet.
I think it would definitively debunk that “workload” graph that’s been going around since forever irritating the ^&)* out of me. (it is actually A graph, in fact it’s THIS graph. It just isn’t a ‘workload’ graph)