It works with my tablet and with my phone
Thanks
| Life In 19x19 http://lifein19x19.com/ |
|
| New android app "BadukAI" http://lifein19x19.com/viewtopic.php?f=18&t=17784 |
Page 2 of 27 |
| Author: | go4thewin [ Sat Sep 19, 2020 2:43 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Would it be possible to release a version with the 15b katago net, or even the 10b one? |
|
| Author: | go4thewin [ Sat Sep 19, 2020 4:03 pm ] |
| Post subject: | Re: New android app "BadukAI" |
@akigo I adjusted the private file with winrar (using zip?) to use the 10b katago (kg.bin.gz) private.mp3: https://drive.google.com/file/d/1-1PCX9 ... p=drivesdk but i can't sign it. Maybe some time in the future when you are not too busy, can you sign and upload? Thanks so much!!! |
|
| Author: | Vargo [ Sat Sep 19, 2020 10:41 pm ] |
| Post subject: | Re: New android app "BadukAI" |
@akigo : I uninstalled and then downloaded the new version. It's still the same, (the message "KG is dead/crashed" comes instantly) But even without KG, it's a great app. |
|
| Author: | akigo [ Sun Sep 20, 2020 6:41 am ] |
| Post subject: | Re: New android app "BadukAI" |
@Vargo,go4thewin,Uberdude Obviously OpenCL is somewhat different on each device type and I can't analyse all these cases since I have only one device here at my disposal. Therefore I checked again with the latest cpu version of katago and lightvector seems to have done a lot here recently since it has become much faster. So it's usable now and I decided to switch to the cpu version in BadukAI. Please download the APK again (version 0.3 now) and check if katago is working for you now. @go4thewin From your admirable work to patch the APK I understand that the need for a 10 blocks network is really urgent, so I have included both the 10b and the 20b network in the new version and added a switch in the settings to choose one of them. |
|
| Author: | go4thewin [ Sun Sep 20, 2020 7:58 am ] |
| Post subject: | Re: New android app "BadukAI" |
Thanks so much! Works great on all my devices! edit: 32 bit version even works on old chromebook, wow! katago now works on windows, mac, linux, android, chromebook, ios and ogs, what wonderful work by these programmers |
|
| Author: | Vargo [ Sun Sep 20, 2020 8:12 am ] |
| Post subject: | Re: New android app "BadukAI" |
It works with my tablet and with my phone Thanks |
|
| Author: | Uberdude [ Sun Sep 20, 2020 9:57 am ] |
| Post subject: | Re: New android app "BadukAI" |
KataGo works on phone now, thanks.
|
|
| Author: | Pio2001 [ Sun Sep 20, 2020 3:18 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Hi, Thank you for this app ! It is the first time that I can use a go AI since Leela 11  (other softwares won't run in Linux Mint Debian since Leela 11).Is it normal that KG runs faster than LZ ? If I understand properly, KG runs in CPU mode while LZ runs in GPU mode. But I get around 20 visits/s with KG and around 10 visits/s with LZ on a Samsung Galaxy S7 smartphone. |
|
| Author: | akigo [ Sun Sep 20, 2020 4:02 pm ] |
| Post subject: | Re: New android app "BadukAI" |
@Pio2001 KataGo will always run on CPU, LeelaZero will use the Neural Processing Unit (called DSP) on Qualcomm devices and use the CPU (with some optimizations) on other chipsets. As the LZ 15b model is much bigger than the KG 10b model (and also a lot stronger), the number of playouts, that you observed, seems plausible. |
|
| Author: | Vargo [ Sun Sep 20, 2020 8:22 pm ] | ||
| Post subject: | Re: New android app "BadukAI" | ||
Same test (H3, komi 7.5, Exynos 990) W: BadukAI 0.3 (KG 20b, 5sec/move) v B: CrazyStone Deep Learning 5d KG 20b and LZ 40b both win at H3, I'll try at H4 
|
|||
| Author: | Vargo [ Sun Sep 20, 2020 9:40 pm ] | ||
| Post subject: | Re: New android app "BadukAI" | ||
Another game : H4, komi 7.5, Exynos 990 W: BadukAI 0.3 (KG 20b, 5sec/move) v B: CrazyStone Deep Learning 5d W+1 PS at move 302, B passed, considering the game was over with W+1. W wanted to continue, I stopped the game, to avoid 50+ useless moves.
|
|||
| Author: | And [ Mon Sep 21, 2020 7:44 am ] |
| Post subject: | Re: New android app "BadukAI" |
Lazy Baduk is installed but your program is not - incompatibility. tablet on MediaTek MT8321, 4x1.3 GHz ARM Cortex-A7. Android 7.0. will not work? limitation on android version? is the limitation in the program? |
|
| Author: | akigo [ Mon Sep 21, 2020 10:46 am ] |
| Post subject: | Re: New android app "BadukAI" |
@Vargo thanks for conducting and sharing the games. As far as KataGo's behaviour at the end of the second game is concerned, the reason is as follows: I configured KataGo to use Tromp-Taylor rules (for the sake of consistency within the app since LeelaZero always uses Tromp-Taylor rules, it doesn't know any other ruleset). With these rules your final score is "stones on the board" + "territory". They have the advantage that they can be formulated very concisely (ideal for AI). But a "disadvantage" is that you can continue playing until the board is full without hurting your score: Playing inside your safe territory doesn't cost (the stone will live, so +1 for the additional stone and -1 for the lost territory) and playing inside your opponent's safe territory doesn't cost either (the stone will die, so no plus here, but also no minus for lost territory). So when there is nothing productive left to do, all moves and the pass are of equal value. I assume that in this case KataGo just plays one of these candidates randomly which is mostly an additional move. |
|
| Author: | akigo [ Mon Sep 21, 2020 11:14 am ] |
| Post subject: | Re: New android app "BadukAI" |
@And The reason for the incompatibility is that the tablet you mentioned has 32 bit operating system whereas BadukAI-APK currently only supplies 64 bit binaries. I will try to provide a 32 bit build too. But for this to succeed all the 3rd party software that I use must be available/buildable for 32 bit, so I can't promise anything at the moment. But I will try ... |
|
| Author: | And [ Mon Sep 21, 2020 1:30 pm ] |
| Post subject: | Re: New android app "BadukAI" |
it would be great! users have many more 32-bit devices, many will say thank you! |
|
| Author: | lightvector [ Mon Sep 21, 2020 7:56 pm ] |
| Post subject: | Re: New android app "BadukAI" |
Vargo wrote: W wanted to continue, I stopped the game, to avoid 50+ useless moves. akigo wrote: I configured KataGo to use Tromp-Taylor rules (for the sake of consistency within the app since LeelaZero always uses Tromp-Taylor rules, it doesn't know any other ruleset). With these rules your final score is "stones on the board" + "territory". They have the advantage that they can be formulated very concisely (ideal for AI). But a "disadvantage" is that you can continue playing until the board is full without hurting your score: Playing inside your safe territory doesn't cost (the stone will live, so +1 for the additional stone and -1 for the lost territory) and playing inside your opponent's safe territory doesn't cost either (the stone will die, so no plus here, but also no minus for lost territory). So when there is nothing productive left to do, all moves and the pass are of equal value. I assume that in this case KataGo just plays one of these candidates randomly which is mostly an additional move. @Vargo @akigo - Just so you know, it should be fairly rare that KataGo prolongs the game by a lot. Usually play some solid connecting moves to make some of its groups absolutely connected, and then proceed to focusedly capture all the dead groups. After that, it will pass, and will usually *not* attempt to play any further moves that would waste time. For example, you can take a look at games like this for typical behavior: https://online-go.com/game/27067104 Or this one, which has more dead stones: https://online-go.com/game/27066163 - after the last dame is finished, kata spends almost all moves to cleanly capture the dead stones (required by tromp-taylor) and interposes one or two solid connection moves besides that, but no more, and then readily passes. The reason KataGo behaves like this is because I added a *very* slight bias to prefer "orderly cleanup and then pass without wasting time" moves as a way to discriminate between moves at the end of the game that otherwise have no intrinsic preference or difference. It's only a very slight bias, but enough to produce this kind of behavior when all the moves still lead to the same outcome, rather than cleanup and passing being at the whim of the neural net's random noise preferences among all the possible moves. Unless you've disabled or changed it, the default config also has KataGo play faster if the opponent is passing a lot, so this cleanup also often doesn't take long. 50+ moves would be very atypical, although maybe it could happen in some case. But very, very atypical. |
|
| Author: | Vargo [ Mon Sep 21, 2020 11:20 pm ] | |||
| Post subject: | Re: New android app "BadukAI" | |||
@akigo @lightvector Very clear explanations, thanks. Two more games (BadukAI03_KG20b is White, 5sec/move, komi 7.5, Exynos990): H9 : Gnugo 3.8 is Black. Result W+R H6 : CrazyStoneDL 5d is Black. Result B+1.5
|
||||
| Author: | Sneegurd [ Tue Sep 22, 2020 5:45 am ] |
| Post subject: | Re: New android app "BadukAI" |
KG init was almost instantly here. Huawei Mate 20 X |
|
| Author: | And [ Tue Sep 22, 2020 7:09 am ] |
| Post subject: | Re: New android app "BadukAI" |
KataGo 10 blocks are stronger than Leelaz 15 blocks: KataGo g170e 10 block d204M - Leelaz 15b 0c4ade79 2 sec/move (GT 610) 152:48 (76-24 KataGo Black, also White) |
|
| Author: | go4thewin [ Tue Sep 22, 2020 8:15 am ] |
| Post subject: | Re: New android app "BadukAI" |
Wow that means that katago 10b is stronger than some of the 40b Leela nets from last year when Leelazero .17 was first released (time parity)? That's crazy. If I remember right, bubblesid's 15b was ahead of the 40b nets for quite some time on consumer machines. That's ridiculous how strong katago is. I think the 10b with 1 playout was stronger than zen6 on breakawa's 5d setting. |
|
| Page 2 of 27 | All times are UTC - 8 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|
