ZBaduk - LeeLa Zero and KataGo from your webbrowser

spook · **#21**

mhlepore wrote:

Second, the winrate plot is super useful, but as I play through my game, sometimes the top choices are not shown. Is this a bug? Did I not let it analyze long enough? (sorry if this info is right in front of my eyes)

I must admit that this point isn't entirely clear for us.

Perhaps this answers your question:

The chart is a plot of the winning percentage (always from black's perspective) according to LeeLa Zero.
The Y-axis shows the winrate, and the X-axis shows the move number.

Each node in the tree however could have 0, 1 or multiple proposed moves by LeeLa Zero, which one is used to draw the chart ? LeeLa Zero will prefer to play the variation with most playouts. For that reason, that is also the move for which the winrate will be shown in the chart.

Well, what if there are branches ? For which of the selected nodes does ZBaduk plot the winrate?
Zbaduk always shows the winrate of the TOP branch. (the main/first line of play).
That means, if there are multiple moves played out not all of them will be plotted as charts.
The variations (branching) will not be displayed at all, only the main line.

Well, that's not entirely true. On top of that also the selected node will be displayed.
So, when you move through a variation which is not the main line, then also that variation will be plotted in the chart as a yellow color, but only for the selected node and the path that leads to it. At that point 2 lines will be plotted, the black line is still the main line, and the yellow line represents the selected variation. So, that you can see visually in the chart whether this is an improvement to the main line or not.

I use this chart especially to quickly detect where the game was decided (i.e. a big fluctuation of the winrate).

The moves which are actually shown on top of the goban, are the ones which are also shown in the table with stats.
The winrate and playouts is displayed on the board. But the colors (green, yellow, red) indicate which move has most playouts.
Again, because that would be the choice of LeeLa Zero.

---

I hope this helps, but please ask away if there are more questions.

spook · **#22**

ZBaduk now support LeeLa Zero version 0.17.
I also updated the weights to the most current "best network".

spook · **#23**

Just installed an update which should result in a more fluent experience for the "Analyze full game" button.

Enjoy.

spook · **#24**

Created basic help files:

https://github.com/bvandenbon/zbaduk-docs/blob/master/README.md

johnsmith · **#25**

How do I stop zooming-in the moves when I'm on laptop?

spook · **#26**

johnsmith wrote:

How do I stop zooming-in the moves when I'm on laptop?

@johnsmith sorry for the inconveniences of this.
ZBaduk switches to a kind of "mobile" mode based on the resolution of your computer and the size of the goban.
It is real convenient for mobile phones, but it can be annoying for laptops.

A better way would be to look at the "user agent" of your webbrowser.
And that's exactly what we will do in the next release.
I promise you that I will make this a priority for our next build.

As for now, an easy workaround is to use the zoom-features of your webbrowser.
i.e. if you put your webbrowser in 80% zoom that could be enough.
For most webbrowsers the zoom can be altered with key combination: "Ctrl +/-"

I hope this helps.

(PS: In mean time, I added this information to the manual as well)

johnsmith · **#27**

Thanks spook, I'm looking forward to seeing this.

spook · **#28**

Just installed a new version with lots of small changes:

There's been a lot of discussion in LZ forums about the definition of "the best move". Apparently many tools (and so did ZBaduk) just take the move with the highest playouts. But actually that's not entirely correct. LeeLa Zero actually sends us an "order" property which should be closer to the truth. To determine this "order", it will often rely on the "LCB" winrate.

So, from now on, our statistics table will be ordered by that "order", and we added and renamed the columns as well.

While going through this we added several small performance tweaks, bug fixes, reconnection improvements which will surely make a difference if you have a weak wifi signal on your phone,

Added just a little transparency to the numbered stones, in case you select a sequence in the stats table. Which makes it a little more beautiful.

For @johnsmith better resizing/autozoom decisions: takes your useragent in account in its choice.

Added contact info on several pages.

And a lot more ...

Enjoy !

spook · **#29**

Sorry for the interruptions that you may have experienced today.
We are in the middle of an update to an experimental LeeLa Zero version.
This may allow us to support handicap games in the near future, score estimations, ...

spook · **#30**

It's time for an update.

In the near future there will be a new release that supports KataGo as well as LeeLa Zero.
To prepare us for this big release, I did a small one today.

For those who want to purchase a commercial membership, our payment page will highlight more payment methods (e.g. visa, mastercard, ...),
even though they may not show up in all countries from what I've heard. --> please do not hesistate to contact me. (zbaduk361 at gmail dot com)

There is a better connection recover mechanism. This is for people with poor internet connections.
There have been some minor performance tweaks as well.

And finally, some minor cosmetics:

PS: we would like to translate our entire service to Korean.
Possibly also Chinese and Japanese. - Please let us know if you know somebody.

The next release will probably be all about KataGo.

spook · **#31**

You may have noticed that in the last weeks, the value of cryptocurrencies like Bitcoin has increased again.
This is good news for companies like NVidea and AMD who create the hardware for these miners.
And it is also good news for cloud service providers. It is not such a good news for ZBaduk though.

Our main cloud service provider has decided to raise its prices by 50%.
(Let's hope that Bitcoin soon crashes again, the only good use for a GPU is baduk mining.)

That does mean that the costs of running ZBaduk will also increase by 50% for now.
In the course of next week, ZBaduk will have to increase prices, as well.
(And we are considering to either limit or totally stop free use of the AI servers).

Q&A:

Why are we doing this ?
From its start ZBaduk has never made profit.
But we have to protect us against bigger losses.

What does it mean if you already have payed for a membership ?
Nothing will change for existing customers.
(You can see the expiry date of your membership in your "Account" page).

What does it mean for new customers ?
Prices will raise soon. If you already had plans to become a paying member,
you may want to do it before we increase prices.

spook · **#32**

Installed another update just a couple of minutes ago.

Again a lot of small changes:
Perhaps the most notable of all changes is the way ZBaduk displays color codes.

In previous versions it was actually the number of playouts that was used to color the moves. For technical reasons, that is also how other tools do it (e.g. Lizzie). But it can be confusing. Anyway, from now on, ZBaduk uses the LCB. ZBaduk refers to this value as the "decision" value.

Attachment:

statistics.png [ 52.93 KiB | Viewed 5603 times ]

spook · **#33**

First of all, last week have been very heavy on our servers,
due to the promotions we did at the European Go Congress.
Next week we will introduce a new server which has more calculation power.

In mean time again a small release, with a little nifty feature:

Attachment:

impact.png [ 16.27 KiB | Viewed 5468 times ]

Basically, it's an evaluation of the last played move.

spook · **#34**

Today's release puts a lot of effort in supporting KataGo as well as LeeLa Zero.
ZBaduk keeps a separate book keeping for each bot, but also merges the statistics to create an average value.
And we also show these 3 lines in the charts.

The chart will show up to 3 lines now. Red, Blue and Grey. Red and Blue are the 2 bots, and Grey is the average of them.
Because these 3 series are very close to each other, we decided to zoom the chart initially.

Attachment:

multiple chart series.png [ 32.42 KiB | Viewed 5403 times ]

We did have to add a lot of normalization for LCB values, to keep the values between 0 and 100,
but anyway, when the LCB value has a low amount of visits, then we will use the winrate.

We did a tweak for LeeLa Zero which is pretty cool. When you play a move which is not proposed by LeeLa Zero,
and you move back up, and put it in "analyze this position" mode, it will actually force LeeLa Zero to consider your "creative" move.
And in return the statistics of that move will also be shown on the board.

Attachment:

tengen.png [ 42.31 KiB | Viewed 5403 times ]

If you play a really long variation, and that variation surpasses the main line in length.,
then after some time the variation will automatically be promoted and will become the main line,
'cause it makes sense in 99% of times.

The game list, now has an upload button, which also supports uploading multiple files.

Attachment:

2019-08-06_015046.png [ 62.81 KiB | Viewed 5403 times ]

There's also a fix for people who may have experienced login issues due to their e-mail address containing uppercase characters.
All of these issues should be fixed.

Cyan · **#35**

Problem with your averaging: both bots has H16 as top choice but B11 is chosen overall. H16 is shown in orange while B11 is in blue.

johnsmith · **#36**

Hi Author, I have 2 questions.
1. I am not able to upload sgf and review full game, even though I bought a package with paypal. Is that an expected behaviour? Uploaded sgf doesn't contain any moves. So I have to relay move by move.
2. The number of playouts on the board circle is not the sum of LZ and KataGo playouts. Is that expected? For example, on the board I see 35k playouts for a move, but in the table below - LZ has 6k and KataGo has 7k playouts.

spook · **#37**

Cyan wrote:

Problem with your averaging: both bots has H16 as top choice but B11 is chosen overall. H16 is shown in orange while B11 is in blue.

You are absolutely right. Something is wrong here.
It's hard to compare apples with oranges, but ZBaduk is trying to mix them to create multi-fruit juice.

What I am trying to say is, that Leela Zero and KataGo estimations are not easy to aggregate/merge/average. Some bots are more sensitive/optimistic than others when they generate a winrate. And sometimes only one bot rates a move, which makes it even more difficult to come up with an average.

One thing is for sure: We took a wrong turn when we tried to calculate a weighed average, that only makes things worse. Right now, ZBaduk uses the number of playouts, and often it does not really matter. But in examples like the one you posted, it screws things up.

A better approach is to benchmark the bots, and to find out how sensitive they are (a standard deviation) for moves they actually agree on. The result of this benchmark is just a simple standard deviation, which can be used to rescale future results. Once the results are all in the same range (i.e. same sensitivity) we can just take a simple average of them.

Attachment:

2019-08-13_020422.png [ 21.03 KiB | Viewed 5282 times ]

In this example there is a conversion factor of 1.1966 for converting KataGo scores to the range of leela zero. (which is the result of a quick benchmark).
It certainly is better than what we had before. So, that's probably what ZBaduk will use in its next release.

spook · **#38**

johnsmith wrote:

Hi Author, I have 2 questions.
1. I am not able to upload sgf and review full game, even though I bought a package with paypal. Is that an expected behaviour? Uploaded sgf doesn't contain any moves. So I have to relay move by move.

Thank you for contacting me, it is certainly *not* expected behavior.
Could you send me the SGF file ? (here, or in a private message, or by e-mail: zbaduk361@gmail.com ).

Is it a handicap game file by any chance ?

johnsmith · **#39**

Hi, thanks for answering. I don't remember the exact game, but it wasn't working for any game downloaded from gokifu, or just saved to sgf from Zen 7. It wasn't a handicap game.

spook · **#40**

I just installed another update of ZBaduk.
This release contains several small fixes for issues.
One of them is a rather important one.

(in)visible game records:

Apparently there was a bug that was hiding some of the game records of users.
In fact, you may have experienced issues uploading files to the server previously,
and it is possible that those files now suddenly appear.
For that matter, we also improved the responsibility of the game record list.

better merging of results

The 2nd major change, is how results of katago and leela zero are merged.
In short: it's getting better.
Longer version: it uses a conversion rate to scale the winrates of katago just a little.

ui settings

There is a configuration panel (it's in the "More..." dropdown list).
It contains some settings on how winrates/lcb values are displayed.

Some prefer winrates to be values in the range of [0-100] , but others actually prefer values in the range of -50 to +50.
And some prefer winrates to be expressed according to the active player, others prefer it to be from black's perspective.
And then some people prefer the rating chart to be zoomed (e.g. 40% to 60%), others prefer it to show the full range (e.g. 0 to 100)
All of this can now be configured.

Attachment:

ui settings.png [ 57.3 KiB | Viewed 5067 times ]

ZBaduk - LeeLa Zero and KataGo from your webbrowser

Who is online