Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python?code
> I just realized - you mentioned that you are using Python to run it - if > you use PyPy instead, you should get about 3x speedup or more. That was a great suggestion! With PyPy michi now takes less than one minute per move. Maybe I'll play it some more.. mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python?code
> You'll need to start reading/changing the code - the second point is > about the N_SIMS variable. Well, I don't like changing other people's software, but if you wish... > What kind of computer are you playing on? > It might be that on Windows Python cannot use multiple processes... Windows 7 on a I5 processor, and I'm quite sure multiprocesses is OK (I see 5 python processes running when michi is thinking). Maybe it's multicore usage not working properly. I will investigate that. BTW, this morning I managed to finish the game. I won by 7 points (88-81 area), that means either me or michi won depending on Komi. Let's say this was a practical draw, meaning our level should be the same, and that confirms michi must really be 6K on KGS. I noticed michi got on playing for a long time after the game was settled, despite my continual passing. For a 6K program that was a bit strange. mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python?code
> The correct parameter is 'gtp', not '--mode gtp'. That worked! It was necessary to insert full path of python executable followed by full path of michi, then "gtp". Full path of michi only + gtp throws again the same error (not a win32 application). Other things: - when playing (michi) against GnuGo (that was winning) I inserted a wrong move, then typed "help" hoping to get help about getting the move back. The program crashed. - at the moment I'm playing myself against michi. It takes more than 3 minutes per move (13x13). Are there severals levels of play to set up? mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python?code
> It should be enough to just enter the path to the program, make sure to > pass the "gtp" parameter. Try opening the shell window to investigate > if you hit trouble. After opening the shell window it works: michi.py is the right command to let it work (of course I have the Python interpreter installed). In GoGui, it does not. Selecting "michi.py" from "new program window" throws an "invalid win32 application". The same happens selecting "michi.pyc" (the compiled program, that also works inside the shell window). Selecting python executable plus full path of michi.py throws "malformed response". Nothing changes adding --mode gtp to all the commands above. >> - can't download patterns.spats. Download stops at 247 bytes. > > Oops, thanks for the heads-up. Fixed. Now it works! Thanks. mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python code
Petr Baudis ucw.cz> writes: > that inspired me to try out something similar in Go over a few evening > recently: > > https://github.com/pasky/michi Very nice work, dr. Baudis! But: - have you any idea how to get it to work with GoGui under Windows (I guess it was developed under Linux)? - can't download patterns.spats. Download stops at 247 bytes. mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Imago - Go board optical recognition
... and again... > - Pisa's organizers agreed on letting us take pictures (or movies, we'll > decide later) of some games. One of the players will also contribute (he > wants to take the pictures himself); others will probably join later on. Pisa was a success. At least, it was on the software side; not so much on the hardware's. The first game, my colleague took pictures with his camera, and everything went allright - the reconstruction was completed in a matter of few minutes. The second game, I also took pictures with my camera, but one of the players kept bumping the table and even my tripod, creating so many errors it required about twenty minutes to complete the reconstruction - we'll discard this game. The third game, we recorded a video with an Ipad mounted on a tripod: everything went right and we'll proceed with the reconstruction as soon as we'll have enough frames extracted. The fourth game, we tried a new program, PhotoClock, capable of taking a picture each time one of the players pressed the keyboard (it worked as a clock): that proved uneffective, as the players' arms were present in each picture (the contrary of what we hoped), and we'll have to try again with a pair of mouses instead. In the whole, we're very satisfied. The paper will be ready at the end of the week, but without a LaTeX template or at least the number of chars per page we won't be able to redact it. We hope the problem will be solved soon! mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Imago - Go board optical recognition
Petr Baudis ucw.cz> writes: > I think a presentation of paper describing your system would find a very > interested audience. > If the system goes through sufficient prior testing, I think deploying > it on EGC2015 would be truly awesome and I would be glad to support it. Hi again! I'm glad to let everyone know that: - version 2.1 of PhotoKifu was released last week; we're now working on version 2.5. - Pisa's organizers agreed on letting us take pictures (or movies, we'll decide later) of some games. One of the players will also contribute (he wants to take the pictures himself); others will probably join later on. - in the next days we'll make a movie of PhotoKifu analyzing a game we recorded two years ago at the David tournament, and will post it on YouTube; - the paper is on its way, and hopefully will be ready at the end of March. mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Imago - Go board optical recognition
Marc Landgraf gmail.com> writes: > Out of curiosity...In the picture you linked > ( http://i59.tinypic.com/10cnu5c.jpg ), how does your program read > the position in the top left, considering the illegal stone there? > > Or does it not have any Go rules knowledge and leaves the interpretation > to the user? In that case it may create .sgf with illegal moves in it. > Considering during a regular game, you will actually snapshot illegal > positions a few times (when you take a picture between placing the stone > and removing the captures) I'm really wondering how your program handles it. Of course it is advisable to take every picture after the removal of the stones possibly captured; the program expects such stones to be removed from the goban. Otherwise, two cases are possible. First one, the player forgot to remove the stone (or did not make in time): in such a case the program detects the error and asks the user what to do next. Second, a suicide move was played (willingly or not). In such a case the program checks the rules and, if suicide is not allowed, again asks the user what to do next (otherwise it gets on). But as this particular problem does not concern the main algorithm, we have not made a final decision. There are other options on the table (for example automatic error correction/stone addition). mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Imago - Go board optical recognition
Hi mr. Baudis! > we must have missed PhotoKifu when surveying the available software. We're getting used to that. In Italy Go is completely unknown and everything related goes unnoticed. Months ago I showed the program to a colleague of mine, who shares many of my hobbies, and he asked "What is this? Chinese checkers?". > There will be a scientific conference at EGC 2015 as well: > > http://pasky.or.cz/iggsc2015/ > > I think a presentation of paper describing your system would find a very > interested audience. That's a fantastic new! After getting no reply from Sibiu last year we were afraid a second conference would never occur. We'll complete the paper soon and will attend the conference. We're already checking the accomodations in Liberec! > we aimed to first test the software on series of events - in (well lit) > Go clubs and on a larger tournament - starting the testing around now > to have enough time to test it sufficiently and notice things like sun > suddenly coming out of clouds, bumping the table, Go server streaming > issues etc. We did an extensive test in December 2012, at the Firenze Go tournament (http://www.eurogotv.com/tournament/showresults.php?toernooicode=T121208B). We realised our program (then version 1.0) only worked fine under good conditions - well lit environment, high point of view, limited number of "disturbances" in the pictures. We encountered dim lights (gobans' surfaces looked almost grey in the morning, completely grey in the afternoon), average points of view (40°-50° of elevation), and up to 60-70 pictures per game spoiled by fingers, hands, arms (sometimes of both players in the same picture, for example: http://i59.tinypic.com/10cnu5c.jpg). It took us a long time, but eventually we solved all these problems. We are now capable of analyzing without errors, taking just a bunch of milliseconds, even this kind of pictures: http://i61.tinypic.com/ak9zdv.jpg (this is probably the worst kind, speaking of "sun suddendly coming out of the clouds") > We also wanted to start talking around now to wms (KGS author) about > a possibility of extending kgsGTP computer program interface of KGS > to demonstration games. I would recommend you to reserve enough time > to working out these things too. Firenze's go players did ask us for such a thing (live feed on KGS). We're certainly going to work that out as soon as VideoKifu will be ready. > I think a good time to start discussing this with them [EGC's organizers] > would be after a first successful real-world tournament test (even on a > small scale like single board). Of course we'll let you know about the Pisa testing. We're now contacting the organizers and will likely be able to take pictures of 3 or 4 games, and print the Kifus immediately afterwards (in a matter of minutes, we hope). Greetings and thanks for your interest! mr. Andrea Carta ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
Re: [Computer-go] [ANN] Imago - Go board optical recognition
> Tomas Musil (a student of mine), has created a state-of-the-art open > source Go board optical recognition software. We have focused on > completely automatic runs, so it automatically detects the board corners > and then the stones on the board, and the precision seems pretty good > at least in reasonable lighting conditions. > > My personal dream would be if we added video capability and further > improved speed + reliability in time for EGC2015 (in Czech Republic) > and were able to deploy it there to transfer large number of top boards. > But this will depend on how much time Tomas will have after the summer > (and we didn't actually check with EGC2015 organizers yet), so it's > still more of just a dream. Hello mr. Baudis! Musil's work is remarkable indeed. I and my fellow colleague, prof. Mario Corsolini, studied the thesis and found it of the greatest interest, both on the theoretical side and the practical one. BTW, Musil seems to believe that "we have not found any other work that we can meaningfully compare our results with"; but since november 2012 we have developed and distributed PhotoKifu, a program aimed to reconstruct whole Go games by means of a series of photograph. We're now releasing version 2.1 (a paper is also in progress) and are working on version 2.5, which will implement Open CV instead of the external (and slow) suite Image Magick. We're also working on VideoKifu, a program that will reconstruct a game from a live video feed; we could not develop VideoKifu before because of the slow Image Magick suite, but Open CV will allow that. At the moment (version 2.1) we achieve on single pictures the same - very good - results Musil got, and take 1 to 4 seconds for picture depending on its size. But our program is optimized for whole games, for which we have 100% stones recognition, even in the presence of hands and whole arms between the camera and the goban (of course the stones we're looking for must still be visible), given the pictures are not too small (I estimate at least 1024 x 768). That requires about 1/10 of second for each picture plus image optimization either by Image Magick or Open CV. Of course, we too are interested in EGC 2015. We hoped to attend the scientific conference at EGC 2013, but could not find reliable informations. Maybe Musil will now be able to contact EGC 2015's organizers and catch their interest: it would be a remarkable feat if we could both go and talk about (and, of course, demonstrate) two programs making use of different approaches. For the moment we're planning to do such a thing during the big international Pisa tournament at the beginning of March, but the EGC 2015 would be the ideal stage. Let us know if you're interested, of course! mr. Andrea Carta, BSc(IT) ___ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go
