Page 2 of 3

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Wed Oct 09, 2019 7:33 pm
by Dann Corbit
Same as last time for me.
I could see three last time and three were vague.

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Wed Oct 09, 2019 8:02 pm
by Daniel Shawul
10. Dann Corbit ?

I am suprized there is no mcts in my DNA :)

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Wed Oct 09, 2019 8:06 pm
by chrisw
Daniel Shawul wrote:
Wed Oct 09, 2019 8:02 pm
10. Dann Corbit ?

I am suprized there is no mcts in my DNA :)
Nope, Dann Corbit not in this batch.
I downloaded the texts more than two months ago, so that might account for MCTS?

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Wed Oct 09, 2019 10:25 pm
by chrisw
Daniel Shawul wrote:
Wed Oct 09, 2019 8:02 pm
10. Dann Corbit ?

I am suprized there is no mcts in my DNA :)
That prompted me to go bug hunting, and indeed, no MCTS, but, plenty of MCTS concatenated with the word that preceded or followed it.
Your dictionary, of the first few hundred entries, is showing up 'scorpiomcts' with the space char removed. But the text database has no 'scorpiomcts' but it does have 'scorpio mcts'. Very weird, I'm wondering if I did some sort of special case handler for upper case mnemonics somewhere, problem is I wrote this code more than two months ago and kind of forgotten what I did.
You have 'cutechesscli because I strip '-' chars and 'atleast' because that seems to be Daniel-ese(!), so it seems to be some kind of special case bug.

Sorry I missed your key word :(


Counter({'search': 1051, 'should': 886, 'because': 867, 'better': 661, 'engine': 535, 'different': 530, 'problem': 507, 'number': 481, 'doesnt': 394, 'probably': 383, 'alphabeta': 375, 'before': 371, 'result': 365, 'method': 357, 'something': 340, 'version': 320, 'engines': 310, 'anyway': 301, 'without': 297, 'scorpio': 291, 'thread': 280, 'results': 278, 'though': 277, 'really': 271, 'already': 264, 'evaluation': 264, 'position': 255, 'threads': 252, 'actually': 248, 'pieces': 247, 'random': 246, 'example': 245, 'winboard': 238, 'instead': 225, 'second': 222, 'positions': 221, 'standard': 213, 'possible': 212, 'bitbases': 210, 'thought': 204, 'network': 202, 'similar': 200, 'enough': 199, 'people': 198, 'variance': 197, 'single': 196, 'algorithm': 194, 'understand': 193, 'another': 190, 'reason': 182, 'estimate': 180, 'approach': 179, 'problems': 178, 'things': 177, 'change': 177, 'effect': 176, 'memory': 173, 'previous': 171, 'infact': 171, 'square': 169, 'thanks': 168, 'values': 168, 'against': 166, 'checkers': 163, 'difference': 160, 'anything': 159, 'mentioned': 158, 'parallel': 158, 'simple': 157, 'performance': 156, 'trying': 154, 'processors': 153, 'captures': 153, 'always': 152, 'player': 152, 'current': 150, 'capture': 148, 'between': 148, 'formula': 146, 'nebiyu': 146, 'generation': 145, 'exactly': 145, 'stockfish': 144, 'tested': 144, 'atleast': 142, 'question': 142, 'tables': 141, 'making': 141, 'factor': 140, 'pretty': 134, 'average': 134, 'implementation': 133, 'working': 131, 'variants': 131, 'opponent': 130, 'players': 129, 'getting': 128, 'please': 127, 'hardware': 124, 'tactics': 124, 'methods': 124, 'scores': 124, 'ofcourse': 123, 'generate': 122, 'faster': 122, 'difficult': 122, 'started': 121, 'pruning': 121, 'function': 121, 'variant': 121, 'reduction': 121, 'everything': 120, 'policy': 120, 'through': 120, 'compared': 117, 'advantage': 116, 'nothing': 114, 'important': 113, 'discussion': 112, 'searched': 111, 'qsearch': 111, 'reduce': 111, 'usually': 111, 'larger': 110, 'squares': 110, 'branching': 109, 'iteration': 108, 'however': 107, 'either': 107, 'numbers': 107, 'clearly': 107, 'needed': 106, 'almost': 106, 'otherwise': 106, 'uniform': 106, 'default': 105, 'playing': 105, 'processor': 103, 'cluster': 103, 'someone': 102, 'indeed': 102, 'improvement': 101, 'unless': 101, 'correct': 101, 'windows': 99, 'command': 99, 'matter': 99, 'saying': 99, 'winning': 99, 'algorithms': 99, 'learning': 98, 'simulations': 98, 'selection': 97, 'scheme': 97, 'depths': 96, 'looking': 96, 'except': 96, 'havent': 95, 'material': 94, 'shared': 94, 'general': 94, 'neural': 92, 'testing': 92, 'little': 92, 'completely': 92, 'ordering': 91, 'becomes': 91, 'believe': 91, 'support': 90, 'information': 90, 'program': 89, 'scaling': 88, 'history': 88, 'multiple': 88, 'tactical': 88, 'montecarlo': 88, 'searching': 88, 'biased': 88, 'expect': 87, 'played': 87, 'training': 87, 'distribution': 87, 'others': 87, 'itself': 86, 'computer': 86, 'improve': 86, 'required': 86, 'reduced': 86, 'promotion': 86, 'rollouts': 83, 'implement': 83, 'definition': 83, 'increase': 82, 'weights': 82, 'illegal': 82, 'option': 81, 'around': 81, 'compare': 81, 'bayeselo': 81, 'separate': 80, 'machine': 79, 'following': 79, 'stored': 79, 'parameters': 79, 'reductions': 79, 'anyone': 78, 'become': 78, 'format': 78, 'changes': 78, 'progress': 78, 'directly': 77, 'during': 77, 'implemented': 77, 'expected': 77, 'finish': 77, 'abdada': 76, 'bigger': 76, 'original': 76, 'whatever': 75, 'sequence': 75, 'referee': 75, 'talking': 74, 'averaging': 73, 'message': 73, 'programming': 73, 'solution': 73, 'instance': 73, 'simply': 73, 'networks': 72, 'normal': 72, 'singular': 72, 'calculate': 72, 'update': 72, 'margin': 72, 'available': 71, 'heuristic': 71, 'estimates': 71, 'million': 70, 'wanted': 70, 'certain': 70, 'constant': 70, 'daniel': 69, 'rating': 69, 'significant': 69, 'calculated': 69, 'specific': 69, 'forgot': 69, 'requires': 69, 'static': 68, 'especially': 68, 'strength': 68, 'significantly': 67, 'worked': 66, 'assume': 66, 'benefit': 66, 'starting': 66, 'iterations': 66, 'couple': 65, 'having': 65, 'changed': 65, 'options': 65, 'slower': 65, 'higher': 65, 'strong': 65, 'unlike': 65, 'generator': 65, 'shallow': 65, 'visits': 65, 'rather': 65, 'thinking': 64, 'iterative': 64, 'currently': 64, 'require': 64, 'figure': 63, 'bitboards': 63, 'splitting': 63, 'generated': 62, 'issues': 62, 'suggested': 62, 'interesting': 62, 'compile': 61, 'inside': 61, 'subtree': 61, 'sometimes': 61, 'adding': 61, 'effort': 61, 'points': 61, 'taking': 60, 'missing': 60, 'control': 60, 'perform': 60, 'quickly': 60, 'posted': 60, 'compression': 60, 'endgame': 59, 'latter': 59, 'sample': 59, 'necessary': 59, 'overhead': 59, 'minmax': 59, 'finished': 59, 'maximum': 59, 'selective': 59, 'consider': 59, 'versions': 58, 'mobility': 58, 'unbiased': 58, 'handle': 57, 'according': 57, 'tuning': 57, 'searches': 57, 'happens': 57, 'smaller': 57, 'selected': 57, 'easier': 57, 'system': 56, 'window': 56, 'feature': 56, 'server': 56, 'towards': 56, 'explanation': 56, 'interested': 56, 'sounds': 55, 'decided': 55, 'running': 55, 'double': 55, 'basically': 55, 'download': 55, 'giving': 55, 'simulation': 55, 'protocol': 54, 'questions': 54, 'generating': 54, 'regular': 53, 'analysis': 53, 'including': 53, 'amount': 53, 'optimization': 53, 'mistake': 52, 'starts': 52, 'attack': 52, 'additional': 52, 'source': 52, 'egbbdll': 52, 'minimax': 52, 'prefer': 52, 'features': 52, 'leaves': 52, 'select': 52, 'loosing': 52, 'lczero': 52, 'considered': 52, 'situation': 52, 'chance': 51, 'actual': 51, 'shouldnt': 51, 'wouldnt': 51, 'whether': 51, 'further': 51, 'obvious': 51, 'decide': 51, 'counts': 50, 'strategy': 50, 'equivalent': 50, 'parameter': 50, 'extensions': 50, 'probability': 50, 'distributed': 50, 'comparison': 50, 'belive': 50, 'models': 50, 'tournament': 49, 'perfect': 49, 'sequential': 49, 'internal': 48, 'probing': 48, 'remember': 48, 'determine': 48, 'solved': 47, 'somewhere': 47, 'called': 47, 'obviously': 47, 'improved': 47, 'programmers': 47, 'access': 47, 'global': 47, 'ratings': 47, 'sampling': 47, 'explain': 46, 'yourself': 46, 'reducing': 46, 'process': 45, 'exploration': 45, 'forced': 45, 'forward': 45, 'shorter': 45, 'calculation': 45, 'programing': 45, 'missed': 44, 'automatically': 44, 'pointed': 44, 'produce': 44, 'captured': 44, 'prediction': 44, 'cutoff': 44, 'highlight': 44, 'database': 44, 'longer': 43, 'script': 43, 'alphazero': 43, 'correctly': 43, 'equally': 43, 'serious': 43, 'searcher': 43, 'return': 43, 'factors': 43, 'assumption': 43, 'counter': 42, 'attacks': 42, 'provided': 42, 'couldnt': 42, 'failed': 42, 'experiment': 42, 'dynamic': 42, 'follow': 42, 'common': 42, 'bitbase': 42, 'hashing': 42, 'reversi': 42, 'legality': 42, 'multiply': 42, 'collisions': 42, 'myself': 41, 'happen': 41, 'useful': 41, 'cutechesscli': 41, 'properly': 41, 'beginning': 41, 'latency': 41, 'include': 41, 'efficient': 41, 'answer': 41, 'opponents': 41, 'understood': 41, 'compute': 41, 'showed': 40, 'display': 40, 'finally': 40, 'knowledge': 40, 'forget': 40, 'recursive': 40, 'definately': 40, 'bounds': 40, 'allows': 40, 'turned': 39, 'realized': 39, 'storing': 39, 'differences': 39, 'choose': 39, 'easily': 39, 'research': 39, 'computing': 39, 'measure': 39, 'deeper': 39, 'reasons': 39, 'parent': 39, 'weight': 39, 'loaded': 38, 'tensorflow': 38, 'behaviour': 38, 'mention': 38, 'complicated': 38, 'discuss': 38, 'discussions': 38, 'everyone': 38, 'percentage': 38, 'depending': 38, 'allowed': 38, 'calculating': 38, 'proportioning': 38, 'purpose': 37, 'removed': 37, 'nullmove': 37, 'cutoffs': 37, 'useless': 37, 'assuming': 37, 'argument': 37, 'remove': 37, 'probes': 37, 'changing': 36, 'modified': 36, 'backup': 36, 'programs': 36, 'wonder': 36, 'sending': 36, 'clicks': 36, 'somehow': 36, 'definitely': 35, 'initial': 35, 'besides': 35, 'suspect': 35, 'children': 35, 'syzygy': 35, 'slightly': 35, 'comparing': 35, 'evaluate': 35, 'optimal': 35, 'switch': 35, 'speedup': 35, 'moving': 35, 'addition': 35, 'saving': 35, 'affect': 35, 'ultima': 35, 'language': 35, 'hashkey': 35, 'researches': 35, 'davidson': 35, 'bidding': 35, 'stackoverflow': 35, 'experience': 34, 'choice': 34, 'deepening': 34, 'replace': 34, 'immediately': 34, 'closer': 34, 'appropriate': 34, 'improvements': 34, 'extend': 34, 'multiplayer': 34, 'middle': 34, 'matrix': 34, 'estimation': 34, 'gradient': 34, 'representation': 34, 'minimal': 33, 'compiled': 33, 'android': 33, 'highest': 33, 'complete': 33, 'output': 33, 'regard': 33, 'setting': 33, 'mostly': 33, 'authors': 33, 'effective': 33, 'operator': 33, 'stupid': 33, 'existing': 33, 'specially': 33, 'passed': 33, 'length': 33, 'chameleon': 33, 'trouble': 32, 'library': 32, 'noticed': 32, 'xboard': 32, 'variable': 32, 'positional': 32, 'section': 32, 'paranoid': 32, 'blocks': 32, 'programmer': 32, 'computation': 32, 'pattern': 32, 'prunings': 32, 'decrease': 32, 'assumed': 32, 'computed': 32, 'loading': 31, 'project': 31, 'weaker': 31, 'former': 31, 'endgames': 31, 'stronger': 31, 'earlier': 31, 'humans': 31, 'details': 31, 'matches': 31, 'limited': 31, 'formulas': 31, 'address': 31, 'sharing': 31, 'impossible': 31, 'transposition': 31, 'strcpy': 31, 'counting': 30, 'uploaded': 30, 'updated': 30, 'opening': 30, 'visited': 30, 'learned': 30, 'checked': 30, 'suppose': 30, 'extension': 30, 'necessarily': 30, 'registers': 30, 'indexing': 30, 'accept': 30, 'crafty': 30, 'deviation': 30, 'functions': 30, 'machines': 29, 'hopefully': 29, 'binary': 29, 'resolve': 29, 'alphago': 29, 'likely': 29, 'anywhere': 29, 'absolutely': 29, 'safety': 29, 'systems': 29, 'future': 29, 'elsewhere': 29, 'special': 29, 'explained': 29, 'peters': 29, 'manner': 29, 'looked': 29, 'implementing': 29, 'reduces': 29, 'complex': 29, 'elostat': 29, 'widening': 29, 'payoff': 29, 'matters': 28, 'apparently': 28, 'batching': 28, 'applied': 28, 'trained': 28, 'comment': 28, 'provide': 28, 'achieve': 28, 'operations': 28, 'increment': 28, 'selectivity': 28, 'entries': 28, 'particular': 28, 'defined': 28, 'communication': 28, 'collision': 28, 'bitsmove': 28, 'confused': 27, 'success': 27, 'install': 27, 'reading': 27, 'minutes': 27, 'upload': 27, 'terminal': 27, 'aggressive': 27, 'accuracy': 27, 'brought': 27, 'reported': 27, 'unfortunately': 27, 'perfectly': 27, 'suddenly': 27, 'screwed': 27, 'suitable': 27, 'checks': 27, 'ponder': 27, 'playouts': 27, 'linear': 27, 'keeping': 27, 'included': 27, 'logistic': 27, 'separately': 27, 'confusion': 27, 'report': 27, 'objective': 27, 'writing': 27, 'clicking': 27, 'voting': 27, 'passing': 27, 'variables': 27, 'zobrist': 27, 'highly': 26, 'directory': 26, 'afraid': 26, 'planes': 26, 'inferior': 26, 'written': 26, 'statement': 26, 'opposite': 26, 'proper': 26, 'minimum': 26, 'introduced': 26, 'handled': 26, 'regarding': 26, 'recently': 26, 'efficiency': 26, 'target': 26, 'console': 26, 'course': 26, 'occupancy': 26, 'singularity': 26, 'remains': 25, 'managed': 25, 'tensorrt': 25, 'website': 25, 'scorpios': 25, 'caching': 25, 'software': 25, 'despite': 25, 'described': 25, 'hanging': 25, 'ignore': 25, 'originally': 25, 'requirement': 25, 'thousands': 25, 'mistakes': 25, 'displayed': 25, 'returns': 25, 'assumes': 25, 'repeat': 25, 'replaced': 25, 'suicide': 25, 'highlighting': 25, 'delete': 25, 'misunderstood': 25, 'processes': 25, 'promote': 25, 'waiting': 24, 'optimized': 24, 'wasted': 24, 'increased': 24, 'package': 24, 'surprized': 24, 'randomly': 24, 'statistics': 24, 'meaning': 24, 'quality': 24, 'imagine': 24, 'related': 24, 'intermediate': 24, 'classification': 24, 'detect': 24, 'reasonable': 24, 'bother': 24, 'critical': 24, 'response': 24, 'relative': 24, 'dropped': 24, 'design': 24, 'parallelization': 24, 'within': 24, 'differently': 24, 'tricks': 24, 'alternative': 24, 'proportion': 24, 'overall': 24, 'confidence': 23, 'supports': 23, 'google': 23, 'confusing': 23, 'happened': 23, 'scratch': 23, 'backend': 23, 'treeht': 23, 'modification': 23, 'claimed': 23, 'claims': 23, 'proven': 23, 'multiplication': 23, 'unique': 23, 'curious': 23, 'difficulty': 23, 'structure': 23, 'surprised': 23, 'probed': 23, 'dependent': 23, 'checkmate': 23, 'decision': 23, 'opinion': 23, 'behavior': 23, 'register': 23, 'modify': 23, 'checking': 23, 'suggestion': 23, 'probabilities': 23, 'likelihood': 23, 'pondering': 23, 'spartan': 23, 'latest': 22, 'precision': 22, 'comparable': 22, 'linked': 22, 'recall': 22, 'causing': 22, 'expand': 22, 'months': 22, 'benefits': 22, 'looses': 22, 'combination': 22, 'massive': 22, 'hashtable': 22, 'asking': 22, 'unnecessary': 22, 'applications': 22, 'combined': 22, 'outcome': 22, 'suggest': 22, 'destination': 22, 'realize': 22, 'somewhat': 22, 'forums': 22, 'largest': 22, 'patterns': 22, 'programers': 22, 'situations': 22, 'michel': 22, 'inverse': 22, 'device': 22, 'bitboard': 22, 'lognormal': 22, 'define': 22, 'playout': 22, 'sorting': 22, 'expanded': 22, 'together': 22, 'libraries': 21, 'calculations': 21, 'divide': 21, 'optimize': 21, 'awesome': 21, 'seconds': 21, 'recent': 21, 'branch': 21, 'controls': 21, 'settings': 21, 'understanding': 21, 'mobile': 21, 'coming': 21, 'depends': 21, 'request': 21, 'houdini': 21, 'surely': 21, 'clicked': 21, 'promotions': 21, 'speculative': 21, 'variances': 21, 'correlation': 21, 'enpassant': 21, 'verification': 21, 'proportions': 21, 'adapter': 21, 'interms': 20, 'increasing': 20, 'thinks': 20, 'egbbdlldll': 20, 'scorpiomcts': 20, 'followed': 20,

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Thu Oct 10, 2019 12:58 am
by Daniel Shawul
Haha even my mis-spellings are there "doesnt", "interms","looses" -- i get corrected on these even when i had to write papers.

How do you even extract all these data ?
This is the kind of programming required for a data science job and definitely very useful in the real world.
We, chess programmers, tend to pefer the lowest level language and optimize -- even when we are doing neural networks now.
This makes us miss the big picture i guess, so many fun things to learn in AI at the moment, but i am only learning how to squeeze
more performance out of a tensorrt backend... :(

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Thu Oct 10, 2019 10:18 am
by chrisw
Daniel Shawul wrote:
Thu Oct 10, 2019 12:58 am
Haha even my mis-spellings are there "doesnt", "interms","looses" -- i get corrected on these even when i had to write papers.
apostrophes get stripped away with all the punctuation (the texts were originally processed for a "data science" NLP project, where removals are 'normal'), so might be my software, and not your spelling!

How do you even extract all these data ?
This is the kind of programming required for a data science job and definitely very useful in the real world.
We, chess programmers, tend to pefer the lowest level language and optimize -- even when we are doing neural networks now.
This makes us miss the big picture i guess, so many fun things to learn in AI at the moment, but i am only learning how to squeeze
more performance out of a tensorrt backend... :(
Well, you do it in Python. I found Python quite a struggle at first because I really was applying my C-mindset to it, and that is just not the way to go. Also is the struggle to even decide to start on something very new, but, like most things, once you start doing it, it all gets really quite easy (plus your thought processes have changed enough to be thinking the new language). There is masses to learn, but I figure we are not literate if we don't understand ML and big data, just being able to code is not enough for the new literacy. By literate I mean: able to work out how the world works.
I tried to get through most of the main fields, image recognition, web scraping, different kinds of nets, viusualistion of networks, NLP, yesterday I started on a short intro to GeoData, that includes mapping the world to 2D, discovering there are available algorithms for all the projections (Mercator and so on), discovering there are all manner of location datasets which one can map to the world map or some subset of it. Datasets store location info, and library geopandas will plot maps from geodata csv's.
I find the google "how can I plot wordclouds" function very useful! Most things that can be done have been done, libraries exist, example code exists, actually quite good tutorials exist, lost of competitions exist (competitions are very good for getting your brain imagining interest stuff to do). Just hurl yourself at the thing, have some ideas, enjoy making connections and you can have a lot of fun. Plus keep up with the way the world is changing, what is and isn't currently possible.
I know what you mean about tensorrt backends. I did all the NN chess training stuff, MCTS-UCT, working program, but realised there was no way for a one or two man team to be able to reach and keep up with Leela developers who form basically a moving target. Hence my switch to wide band knowledge seeking. It also means I can have relatively coherent discussions with rather bright youngest son who is in bleeding edge neurology research doing some amazing stuff. One's kids are not allowed to surpass the patriarch! Well, they do, but I can try.

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Thu Oct 10, 2019 10:25 am
by chrisw
Correctly guessed so far:

7. bob
8. Laskos
9. xxxx
10. xxxx
11. Graham Banks
12. Daniel Shawul

leaving 9 and 10.

First hint: in no particular order: amongst other things, one is or was a chess programmer and one is or was a chess engine tester.

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Thu Oct 10, 2019 11:10 am
by Guenther
May be I am '9'? A pity my eyesight is too bad for the smaller letters...

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Thu Oct 10, 2019 11:47 am
by Paloma
09. Ingo Bauer
10. Christophe Theron

Re: Guess the Poster CCC-WordCloud Fun Competition - Part 2

Posted: Thu Oct 10, 2019 12:01 pm
by chrisw
Guenther wrote:
Thu Oct 10, 2019 11:10 am
May be I am '9'? A pity my eyesight is too bad for the smaller letters...
Well done, indeed you are no 9