About 10,000, 100,000, 1,000,000, 10,000,000 bytes:
0.44888181407357763
0.433259276532548
0.35976039165384843
0.43851910590941723
must be an issue somewhere
Well, I guess it was not sure at first, hence the large score, it doesn't
change rapidly the more samples it uses see the upper 3.....and it slowly gets
better ya....hmm......So if each correct letter prob it picks is 100%, then
average would be high; 1.00000000....high=better predictor
code is so small now:
for m in range(256):
if m == ord(window[-1]):
list.append(predict[m])
print(round((sum(list) / len(list)), 200))
Now, we have to include program size....so...mine is basically 5,000 bytes not
compressed......Hmm, if I'm compressing 100MBs, or rather trying to get score
1.0, then, we could say if my code was 100MB big then maybe I am storing every
BPC answer (not that this is BPC I'm doing), so 5KBs is like, take 100MB and
minus 5KBs, and that gives you the discount to remove from my 0.44888 score
above I got.....so 100,000,000 - 5,000 = 0.00005....therefore 0.44888 - 0.00005
= 0.44883
RIGHT? Isn't this cool?
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/Td13a829978c4c9f3-Mf59bb4ee48af19e335f636d6
Delivery options: https://agi.topicbox.com/groups/agi/subscription
