Skip to content

Instantly share code, notes, and snippets.

View BooBSD's full-sized avatar

Artem Hnilov BooBSD

View GitHub Profile
@BooBSD
BooBSD / FPTM_best_result.md
Last active October 4, 2025 16:20
Best Result of Fuzzy-Pattern Tsetlin Machine Text Generation

The Fuzzy-Pattern Tsetlin Machine performs character-level text generation using a 64-character context window:

FRANCISCO:
Sir, he may live:
I saw him beat the surges under him, and so heart
What marricg in his for not the king in the world
full for the head of the seath of king of the world
full not the peace of for the change for a would
for his a for for the cannot be a parther by all the
@BooBSD
BooBSD / nanoGPT.md
Last active October 4, 2025 16:22
nanoGPT

The nanoGPT performs character-level text generation using a 64-character context window:

ROMEO:
Uncarvello against what teyn thy husband!
I know you but for my daughters, I come to the
man: to the fresh of him thence from the wap;
If I did play to beat have I spoken'd thee,
But dispatch it to do with the dutyring of him;
When with the lidst encounter widows when the
@BooBSD
BooBSD / ShakespeareTM.md
Last active February 3, 2025 08:52
Shakespeare TM

I am training modified Fuzzy Patterns Tsetlin Machine to generate texts in Shakespeare’s style, character by character, without tokenization with a context window of 64 characters. Here is an example of the generated text:

CAMILLO:
I have stood seem the lady thee honour hearts
What, here, then are enter’d her to the state,
How these thou shalt tell thee thee to him.

KING RICHARD III:
By the not the last thou the king here,