20 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Ethan Shapiro	284a29d7af	f	2024-03-20 19:53:50 -07:00
Ethan Shapiro	3747af9d22	added state saving	2024-03-20 19:52:13 -07:00
Arthur Lu	4fb81317f0	add letter_guess symlink, add model loading into ai.py	2024-03-20 17:31:27 -07:00
Arthur Lu	12601964bd	add eval script for convienience	2024-03-20 12:59:14 -07:00
Arthur Lu	c448e02512	add evaluation to eric's wordle solver (eval.py)	2024-03-20 12:53:40 -07:00
Arthur Lu	848d385482	run model train, abt 3 avg reward	2024-03-20 12:18:15 -07:00
Arthur Lu	f40301cac9	delete gym-wordle, fix some issues in letter_guess gym, add wandb integration	2024-03-19 16:49:01 -07:00
Ethan Shapiro	fc197acb6e	started new letter guess environment	2024-03-19 11:52:10 -07:00
Ethan Shapiro	e799c14ece	new reward scheme	2024-03-18 11:25:14 -07:00
Ethan Shapiro	bbe9a1891c	updated wordle to gymnasium env	2024-03-15 18:19:58 -07:00
Arthur Lu	9172326013	upload wordle env, fix indexing issue in wordle env, attempt to improve reward (no improvement)	2024-03-14 16:47:11 -07:00
Arthur Lu	4836be8121	remove debug prints	2024-03-14 15:00:19 -07:00
Arthur Lu	5672169073	copy the wordle env locally and fix the obs return	2024-03-14 14:49:17 -07:00
Arthur Lu	5ec123e0f1	minor changes	2024-03-13 13:57:23 -07:00
Arthur Lu	e9622b6f68	switch to notebook	2024-03-13 11:04:30 -07:00
ltcptgeneral	83e81722d2	this should probably be working but isn't	2024-03-12 22:14:03 -07:00
ltcptgeneral	320f2f81b7	delete tests	2024-03-12 21:42:59 -07:00
ltcptgeneral	c121415e31	save plot	2024-03-07 00:54:47 -08:00
ltcptgeneral	6cbffcdec1	feasible training times by dropping to bert-base	2024-03-06 21:48:09 -08:00
ltcptgeneral	6a4027afb7	minimal example of bert idea	2024-03-06 21:00:38 -08:00