gen-ai
/
makemore
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • 083d08277a Rename file #13 Hayden Edelson 2023-11-27 22:02:24 -0600
  • c35a675159 Create bigram neural network model Hayden Edelson 2023-11-27 22:01:37 -0600
  • 959ede8752 Building makemore Hayden Edelson 2023-10-23 22:40:24 -0500
  • 82375e4c0a test branch protection Hayden Edelson 2023-10-23 22:39:53 -0500
  • cf585e532d Test branch Hayden Edelson 2023-10-23 22:38:41 -0500
  • f19ed355dc Building bigram model Hayden Edelson 2023-10-23 22:35:47 -0500
  • 2663a1ef3a
         Merge c0938a4b5d into 988aa59e4d #12 Jonas Reinsch 2023-10-19 16:41:47 +0200
  • c0938a4b5d Simplify collection/sorting of all possible characters. #12 Jonas Reinsch 2023-10-19 16:28:25 +0200
  • 5474ccc2f4
         Merge 7e6171987c into 988aa59e4d #10 JohanNorberg 2023-07-23 00:14:07 +0200
  • 7e6171987c Added --input-file-encoding as a command line argument #10 Johan Norberg 2023-07-23 00:11:38 +0200
  • 27c08e0004 Update &nbsp #8 Joshua Batson 2023-01-27 12:40:54 -0800
  • 3adc55b9a6
         Fix link Joshua Batson 2023-01-26 14:41:16 -0800
  • 52c55ce795 Add inducing induction Joshua Batson 2023-01-26 14:39:39 -0800
  • a5ed03427d
         Merge 98a037cf02 into 988aa59e4d #6 iamdoron 2022-12-07 12:51:10 +0200
  • 98a037cf02
         remove duplicate words #6 iamdoron 2022-12-07 12:42:35 +0200
  • 4a7ba9ceca
         Merge 134bdd3943 into 988aa59e4d #5 Johnny Peck 2022-11-20 16:33:15 +0100
  • 988aa59e4d tune description of the repo wrt references master Andrej Karpathy 2022-11-20 18:18:02 +0900
  • 134bdd3943
         'too' -> 'a' #5 Johnny Peck 2022-11-07 17:13:27 -0500
  • f61811b994
         Merge pull request #3 from normanyu/fix-rnn-prev-state Andrej 2022-09-15 08:26:41 -0700
  • bf38625014 Fix bug in RNN where hprev always referred to start. Change so that hprev refers to output of previous cell #3 Norman Yu 2022-09-15 18:10:19 +0800
  • 2f5e8d746e change readme morelm Andrej Karpathy 2022-09-02 20:56:16 +0000
  • c079e1ce76 add a bag of words model that looks suspiciously similar to a transformer ;) Andrej Karpathy 2022-08-21 20:18:20 -0700
  • b697f434bc add an RNN and a GRU language model Andrej Karpathy 2022-08-21 18:54:44 -0700
  • 6694b67d37 generalize makemore into other types of language models, and add bigram LM and an MLP LM Andrej Karpathy 2022-08-21 17:53:52 -0700
  • 50617fa75d fix comment Andrej Karpathy 2022-08-20 01:29:27 +0000
  • d4ede45208 implementation of InfiniteDataLoader sad Andrej Karpathy 2022-08-20 01:24:44 +0000
  • a7c52cd4d0 remove some guardrails for this simple of a use case Andrej Karpathy 2022-08-20 00:37:49 +0000
  • 4e0137ddf6 remove gradient clipping i dont think its needed at this small scale Andrej Karpathy 2022-08-20 00:33:28 +0000
  • 35435ec087 simplify optimizer init and delete code Andrej Karpathy 2022-08-20 00:32:40 +0000
  • d26d9750ee remove weight init, not needed at this scale Andrej Karpathy 2022-08-20 00:30:09 +0000
  • 0a19a59564 add max steps Andrej Karpathy 2022-08-20 00:29:54 +0000
  • 055e7ee48a respect multigpu envs, e.g. cuda:2 designation should work Andrej Karpathy 2022-08-19 22:31:36 +0000
  • 013af92770 big refactor to make easier and api agree with mingpt more Andrej Karpathy 2022-08-19 22:29:58 +0000
  • 054568ec24 add some generated examples of names for fun Andrej 2022-06-09 20:55:39 +0000
  • c3aaadcb16 split out train,test,new separately when reporting on sampling word identity Andrej 2022-06-09 20:55:27 +0000
  • e0a08f234c small tweaks to support the Apple Silicon M1 chip device 'mps'. But this is not yet faster because a lot of ops are still being implemented https://github.com/pytorch/pytorch/issues/77764 , in particular for us the layernorm backward as of today Andrej Karpathy 2022-06-09 12:59:39 -0700
  • 8f79bd0126 first commit Andrej Karpathy 2022-06-09 12:46:25 -0700
  • 180c4f7260
         Initial commit Andrej 2022-06-09 12:29:36 -0700