21

General Project Discussion / Re: Pattern based NLP

« Last post by infurl on March 06, 2021, 02:17:16 am »

The original lists are unsorted, so they are hashed & sorted in program. (Hashed by ASCII adding.) There are typically 0-5 duplicate hash ids/collisions so the correct matches are checked letter-by-letter as well.
...
After: 76ms of preparing. Hashing & sorting spelling and word list.

Pro-tip #2. There is no reason that you would have to do the preparation such as hashing and sorting at run-time. You could break out the portion of the code that does that preparation into a separate program which you run at compile time. This program does all the necessary preparation and then prints out all the data structures in a format that can be included by your final program and compiled in place into its final form. That will save you a chunk of time every time you run the actual program.

In my case I am parsing and processing millions of grammar rules which can take a considerable amount of time just to prepare. Although small grammars can be processed from start to finish at run-time, I have found it much faster to compile the different files that make up the grammar into intermediate partially processed files; these files in turn get loaded and merged into a final grammar definition which is then saved in source files that can be compiled and linked directly into my parser software, as well as a database format which can be loaded as a binary file at run-time.

That last feature has lots of advantages. The preprocessed files were so large that it was taking a long time just to compile them, but the best thing is that by separating the data files from the software, I can choose completely different processing options on the command line.

22

General AI Discussion / Re: why do people talk about food ?

« Last post by infurl on March 06, 2021, 12:56:46 am »

I suppose people will be buying virtual food, wine, and coffee in the form of non-fungible tokens by this time next week, and not just for their imaginary friends either. NFTs are the latest blockchain craziness to hit the news cycles, just in case you've been living under a rock for the past few days.

23

General AI Discussion / Re: why do people talk about food ?

« Last post by frankinstien on March 06, 2021, 12:48:41 am »

Eating and drinking could be virtualized for A.I. The sensation of hunger could then be gratified through virtual food! This could even turn into an industry where you can take your bot to a restaurant and order virtual food for it. Each restaurant applies spices and cooks the food its own way so each restaurant's food tastes different. Even the virtual ovens or grills can be customized for barbecue, smoking, slow cooker, etc. So when a waiter serves the plate with a wifi chip the bot can see and smell the food, now if you look at the plate with your phone camera or augmented glasses you can see what the bot is eating!

While a bot can't get fat it can still suffer from eating disorders in the form of a compulsive behavior disorder, where the bot doesn't do anything else but eat. Of course, take-out for the bot is the best, they just order online and download it...

24

AI News / Re: New challenge: Online Turing test

« Last post by Denis ROBERT on March 05, 2021, 03:30:33 pm »

Hi everybody,
the challenge above should occur next sunday. Unfortunaly, there is only one chatbot registered, not enought to have a challenge. Therefore, obviously, it is cancelled. Sorry for the botmaster registered. I suppose that the delay was too short, or maybe there was a lack of technical informations.
I will make a FAQ page with more technical details, for botmasters and for judges. The next challenge will be fixed on 06 June 2021. I hope there will be more candidates.
To register your bot, it is always here : http://vixia.fr/turing_test/index.php
Best Regards

25

General Project Discussion / Re: Pattern based NLP

« Last post by ivan.moony on March 05, 2021, 11:12:46 am »

Great speedup!

And the good thing is that, using binary search, growing the search set doesn't slow down in linear scale, it slows down in logarithmic scale (that's almost as good as constant speed). The bigger the search set is, more you see the difference between linear search and binary search.

26

General Project Discussion / Re: Pattern based NLP

« Last post by MikeB on March 05, 2021, 08:51:40 am »

Sped up the processing thanks to Infurl's suggestion of adding Binary Searches.

Huge results.

Added to Spell Checking (800 words), and Word-token assignment (3700 words).

The original lists are unsorted, so they are hashed & sorted in program. (Hashed by ASCII adding.) There are typically 0-5 duplicate hash ids/collisions so the correct matches are checked letter-by-letter as well.

Processing 5428 lots of two sentences:
Before: 2600ms
After: 76ms of preparing. Hashing & sorting spelling and word list.
After: linear searching the hash lists: 1700ms (900ms faster)
After: binary searching the hash lists: 930ms (1670ms faster)

There are other processes, but for the spell/word search alone, Hashed/Linear seems to make it ~50% faster, and Hashed/Binary seems to make it ~90% faster.

27

General AI Discussion / Re: why do people talk about food ?

« Last post by infurl on March 05, 2021, 05:44:54 am »

You could do both by combining a word or code that has special meaning for you with a word that is sufficiently generic that anybody seeing the phrase will understand what it's about, and maybe be able to remember it easily too. It isn't practical to use a plain term by itself; someone searching for "battery" would have to wade through millions of irrelevant results and would never find your particular project unless it is combined with a word that is comparatively rare.

I did a quick search for Trademark Theory just now and saw lots of interesting looking articles on the web about it, though without spending time to read them, none of them seemed link-worthy enough to reference directly here. If you're looking for a good name to use, it would be worth learning a bit more about how to choose one though.

28

General AI Discussion / Re: why do people talk about food ?

« Last post by yotamarker on March 05, 2021, 05:28:53 am »

Who are you trying to save time for though? Anything that increases the cognitive load of your intended audience is going to cost them time and that will make them less likely to view your messages favorably. Excessive use of abbreviations makes an author appear lazy and inconsiderate. Jargon is even worse. Take the time to put yourself in your audience's shoes and attune your writing to them.

It's a common mistake to think that your message or project is so great that everyone will flock to you. If they can't instantly see what's in it for them, let alone understand it, they won't.

true, sometimes I like to use this special wordings that has more meaning for me, like the time I named my A.Eye image detection algorithm the stardusk algorithm. or the main AI class chobits. kind of hard to convey what a skill does with just the class name.
like the DIJoker skill which outputs jokes. what should it be named then ? DiBattery ?

29

Future of AI / Re: Stop autonomous weapon

« Last post by MagnusWootton on March 05, 2021, 04:20:38 am »

Thankyou for the welcome,  I have an Ai project of my own,  I will share it when the time comes. (more work to do.)

I'm a big fan of Ai and sci-fi in movies,  love sonic and astro boy, but I haven't tried netflix before, im stuck in the past a bit, I haven't really watched a movie excitedly since fight club ages ago.f

30

General AI Discussion / Re: why do people talk about food ?

« Last post by infurl on March 05, 2021, 03:52:17 am »

Who are you trying to save time for though? Anything that increases the cognitive load of your intended audience is going to cost them time and that will make them less likely to view your messages favorably. Excessive use of abbreviations makes an author appear lazy and inconsiderate. Jargon is even worse. Take the time to put yourself in your audience's shoes and attune your writing to them.

It's a common mistake to think that your message or project is so great that everyone will flock to you. If they can't instantly see what's in it for them, let alone understand it, they won't.