Re: [agi] Re: Pre-crime in the multiverse - is REAL! This is mind bending.

2021-09-05 Thread Quan Tesla
Oh, one can but dream. Sarcasm tends to work like that. By your argument, accountants could instantly save 50% on the government salary bill by effectively halving the current salaries of all government employees and moving the savings over to the fully unemployed. In the absence of this, more

[agi] Re: working code to compare words!!!

2021-09-05 Thread magnuswootton81
Great results. definitely looks like its working! -- Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T01fa5e447808d368-M4c55323d5be6eb59069fe194 Delivery options: https://agi.topicbox.com/groups/agi/subscription

Re: [agi] Re: Pre-crime in the multiverse - is REAL! This is mind bending.

2021-09-05 Thread Matt Mahoney
On Thu, Sep 2, 2021 at 5:57 AM Quan Tesla wrote: > > "Full employment can be had with the stoke of a pen. Simply institute a six > hour workday. That will easily create enough new jobs to bring back full > employment." > > Would this mean that government employees would have to productively

Re: [agi] Re: Black Winter

2021-09-05 Thread doddy
are you going to develop this open source GPLv3+ licensed *ref*lexive *per* sistent *sys*tem for windows? On Sun, Sep 5, 2021 at 4:12 AM Nanograte Knowledge Technologies < nano...@live.com> wrote: > Great share! Thanks > > -- > *From:* Basile Starynkevitch > *Sent:*

[agi] Re: working code to compare words!!!

2021-09-05 Thread immortal . discoveries
and the new code: input2 = open('lc.txt', 'r', encoding='latin-1').read() preword1words = [] preword2words = [] preword1counts = [] preword2counts = [] betterbe10timesseenatleast1 = 0 betterbe10timesseenatleast2 = 0 for count2 in range(10): window = input2[count2: count2 + 12]

[agi] Re: working code to compare words!!!

2021-09-05 Thread immortal . discoveries
Added double sides :) for DOUBLE accuracy. Found a bug where every word got often a '' with no word, big oops ! Added minimum = 10 counts like the others do. It warms you this is Not going to do well for accuracy. Made all caps lowercased. 1/6th or so More data now! Fixed the algorithm so it

[agi] Re: Base work for very fast, pretty dirty & extremely practical audio recognition

2021-09-05 Thread magnuswootton81
Hi Stefan.   pixel-wise sum of differences?  that sounds cool, I never thought of that!     that would make it really damn fast.    if that wasnt too convoluted that would be an amazing way to do it! I was just going to store the "samplet" itself as the class centres.  but i guess if you sum

Re: [agi] Re: A letter on Foundation Models, presented here without opinion.

2021-09-05 Thread immortal . discoveries
When I went to a tool below it actually told me some interesting stuff > https://www.ultraedit.com/company/blog/community/best-way-to-handle-large-text-files.html Another interesting thing is AGIs while thinking ex. 100 times faster than our brains, can actually work that fast, ex. I myself

Re: [agi] Re: Black Winter

2021-09-05 Thread Nanograte Knowledge Technologies
Great share! Thanks From: Basile Starynkevitch Sent: Saturday, 04 September 2021 18:42 To: agi@agi.topicbox.com Subject: Re: [agi] Re: Black Winter On 8/19/21 2:47 PM, John Rose wrote: For example, dot net has a feature called reflection, as does Java, that