Commit Graph

  • 73822b8bfa move ex 3.3 solution outside main chapter rasbt 2024-03-10 07:18:24 -05:00
  • da33ce8054 remove redundant unsqueeze in mask rasbt 2024-03-09 17:42:25 -06:00
  • 6ba97adaee add PyTorch version rasbt 2024-03-09 17:35:44 -06:00
  • 1d819c3d9c Merge pull request #56 from eltociear/patch-2 Sebastian Raschka 2024-03-09 10:27:11 -06:00
  • 5ca60321c4 add a100 numbers rasbt 2024-03-09 10:20:08 -06:00
  • 47519f4d14 Update compare-bpe-tiktoken.ipynb Ikko Eltociear Ashimine 2024-03-10 01:11:35 +09:00
  • 29ca41799a use need_weights=False rasbt 2024-03-09 10:09:17 -06:00
  • 5643c88db9 add pytorch mha rasbt 2024-03-08 09:30:55 -06:00
  • 3beaea46ce add lowres figs for better navigation rasbt 2024-03-08 07:18:06 -06:00
  • 404f48aa74 automatically run on gpu or cpu rasbt 2024-03-07 20:14:03 -06:00
  • c5b17c3d67 simplify rasbt 2024-03-07 07:52:24 -06:00
  • f454944d5d add setup recommendations rasbt 2024-03-07 07:32:30 -06:00
  • 083d11fbd0 Merge pull request #55 from rayedbw/patch-4 Sebastian Raschka 2024-03-07 06:31:01 -06:00
  • 99a5e28def rename q,k,v for consistency with chapter 3 rasbt 2024-03-07 06:30:40 -06:00
  • 496079c61e Update mha-implementations.ipynb Rayed Bin Wahed 2024-03-06 23:03:57 +08:00
  • b6fe1a37b3 also add simple wrapper rasbt 2024-03-06 08:38:53 -06:00
  • 571377a2d6 update title rasbt 2024-03-06 08:34:04 -06:00
  • d2835931b7 Merge pull request #54 from rasbt/mha-variants Sebastian Raschka 2024-03-06 08:32:53 -06:00
  • 87fcfd9245 mha variants rasbt 2024-03-06 08:30:32 -06:00
  • d4754f1bdd change dim=1 to dim=-1 rasbt 2024-03-04 18:54:43 -06:00
  • b50c42ffbb Merge pull request #52 from rasbt/use-embedding-dropout Sebastian Raschka 2024-03-04 07:07:46 -06:00
  • e0df4df433 add dropout for embedding layers rasbt 2024-03-04 07:05:06 -06:00
  • 3198363c4f add wording from three to four rasbt 2024-03-04 06:42:58 -06:00
  • 29672da3b0 stride consistency rasbt 2024-03-03 19:37:06 -06:00
  • 742f0a6d29 add missing output in bonus rasbt 2024-03-03 17:29:46 -06:00
  • f526a8d7fb add requirements file for bonus notebook rasbt 2024-03-02 16:54:24 -06:00
  • cc2383c4de remove duplicated exercise code rasbt 2024-03-02 16:44:36 -06:00
  • c071ea73f9 Update DDP-script.py Sebastian Raschka 2024-03-01 18:31:05 -06:00
  • c9dccb0c40 Merge pull request #33 from rayedbw/patch-1 Sebastian Raschka 2024-02-29 20:00:09 -06:00
  • 267e33cfaf remove redundant import rasbt 2024-02-29 19:59:05 -06:00
  • d419c02792 Merge pull request #39 from rayedbw/patch-3 Sebastian Raschka 2024-02-29 12:30:50 -06:00
  • 32087331ae Update Dockerfile Rayed Bin Wahed 2024-03-01 02:10:01 +08:00
  • a94d53a752 Merge pull request #38 from rayedbw/patch-2 Sebastian Raschka 2024-02-29 12:06:05 -06:00
  • c47e434162 Update README.md Rayed Bin Wahed 2024-03-01 01:56:58 +08:00
  • 7d732a5db0 add readme for devcontainer rasbt 2024-02-29 09:00:06 -06:00
  • ee24acd481 Merge branch 'main' of https://github.com/rasbt/LLMs-from-scratch rasbt 2024-02-29 08:31:20 -06:00
  • b827bf4eea remove redundant double-unsequeeze rasbt 2024-02-29 08:31:07 -06:00
  • 3278243dd5 Merge pull request #31 from rayedbw/main Sebastian Raschka 2024-02-29 08:24:29 -06:00
  • fb770ef97c update docker files and docs rasbt 2024-02-29 08:22:53 -06:00
  • 2fb035435e Update ch04.ipynb Rayed Bin Wahed 2024-02-27 23:05:36 +08:00
  • d89aaf319d update folder name rasbt 2024-02-27 08:53:04 -06:00
  • a060f923d3 Merge pull request #32 from rasbt/hparam Sebastian Raschka 2024-02-27 08:52:01 -06:00
  • 87a743076d hparam tuning script rasbt 2024-02-27 08:51:03 -06:00
  • f6266c3756 improve code comments rasbt 2024-02-27 06:40:35 -06:00
  • 45a10dd823 Add devcontainer starter doc Rayed Bin Wahed 2024-02-27 13:04:06 +08:00
  • fa7e659eb3 Add devcontainer Rayed Bin Wahed 2024-02-26 20:24:15 +08:00
  • 78ed2e35bc Add requirements.txt to main repo Sebastian Raschka 2024-02-25 13:32:30 -06:00
  • 3debb2f0df Update README.md Sebastian Raschka 2024-02-25 13:31:32 -06:00
  • 3f186ab072 use .shape instead of .size() for consistency rasbt 2024-02-25 08:47:25 -06:00
  • cdcd73ba7f drop_last=True rasbt 2024-02-25 07:23:38 -06:00
  • 6243726ab3 rename to dataloader v1 rasbt 2024-02-24 07:48:18 -06:00
  • 4e68649f16 comment update rasbt 2024-02-24 06:52:17 -06:00
  • f057156181 use smaller number of tokens to emphasize next token prediction goal rasbt 2024-02-15 20:09:20 -06:00
  • 557ddfc684 make a new example for shortcut connections rasbt 2024-02-15 19:34:12 -06:00
  • 250e6306e2 use attn_scores from sec 3.4 instead of 3.3 rasbt 2024-02-14 20:23:59 -06:00
  • 231a854ae7 use less ambiguous var name rasbt 2024-02-13 07:05:37 -06:00
  • 320f63829f Merge pull request #29 from Intelligence-Manifesto/patch-5 Sebastian Raschka 2024-02-12 07:34:37 -06:00
  • 6a09e7b03a **step 2** Intelligence-Manifesto 2024-02-12 18:32:28 +08:00
  • 1d6f2c9084 rearrange exercise order rasbt 2024-02-11 14:46:05 -06:00
  • 79d90d8147 Merge pull request #28 from rasbt/ch4-exercise-solutions Sebastian Raschka 2024-02-11 11:52:18 -06:00
  • fe332006de ch4 exercise solutions rasbt 2024-02-11 11:51:39 -06:00
  • 103f7826ad use same iter to make figs consistent rasbt 2024-02-11 09:12:52 -06:00
  • 352b83d225 make softmax explicit rasbt 2024-02-11 08:42:21 -06:00
  • 7d86023fc4 make softmax explicit rasbt 2024-02-11 08:41:45 -06:00
  • 5840b4b5f8 update name of last section rasbt 2024-02-11 07:35:07 -06:00
  • e0b6fdbc53 Merge pull request #27 from Intelligence-Manifesto/patch-4 Sebastian Raschka 2024-02-11 07:31:06 -06:00
  • 1278615c25 12 -> 21 Intelligence-Manifesto 2024-02-11 14:17:55 +08:00
  • baa8617921 variable name fix rasbt 2024-02-10 17:53:54 -06:00
  • 496b52f842 format the other GPT architecture sizes rasbt 2024-02-10 17:47:56 -06:00
  • 40477c55b3 add missing ex sol to table rasbt 2024-02-10 10:13:21 -06:00
  • 10aa2d099d add print statements for illustration purposes rasbt 2024-02-10 10:10:14 -06:00
  • cc459b6b5a Merge branch 'main' of https://github.com/rasbt/LLMs-from-scratch rasbt 2024-02-08 20:17:01 -06:00
  • 5d1d8ce511 add shape information for clarity rasbt 2024-02-08 20:16:54 -06:00
  • 24d71784e2 Merge pull request #26 from Intelligence-Manifesto/patch-3 Sebastian Raschka 2024-02-08 17:19:29 -06:00
  • a61583e31e if -> in Intelligence-Manifesto 2024-02-09 04:18:07 +08:00
  • 55ced1d95e Update README.md Sebastian Raschka 2024-02-06 07:38:52 -06:00
  • 3a5fc79b38 add and update readme files rasbt 2024-02-05 06:51:58 -06:00
  • 2b38b63a7a move overview up rasbt 2024-02-04 15:57:03 -06:00
  • bb50de7210 adjust figure width rasbt 2024-02-04 10:12:11 -06:00
  • 1653f6953a adjust figure width rasbt 2024-02-04 10:09:36 -06:00
  • f2fe1d0d85 Merge pull request #24 from rasbt/ch04 Sebastian Raschka 2024-02-04 10:06:56 -06:00
  • b86cc682f0 update links in readme rasbt 2024-02-04 10:05:55 -06:00
  • ec312e581b add chapter 4 code rasbt 2024-02-04 10:02:05 -06:00
  • 16b30ccd5c set stride equal to max len rasbt 2024-02-03 08:50:56 -06:00
  • b9974da94d Merge pull request #22 from Intelligence-Manifesto/patch-2 Sebastian Raschka 2024-02-03 08:46:40 -06:00
  • 78ab68b4d3 missing single quotes Intelligence-Manifesto 2024-02-03 03:41:31 +08:00
  • d261abce4c add forward pass rasbt 2024-01-31 08:00:19 -06:00
  • fcb13fd636 add code backbone ch04 rasbt 2024-01-29 08:13:52 -06:00
  • 3002777dd2 Update README.md Sebastian Raschka 2024-01-28 18:41:02 -06:00
  • c4803bb91a add libmamba rasbt 2024-01-28 12:47:29 -06:00
  • 22442b7012 use torch.softmax instead of F.softmax rasbt 2024-01-28 12:25:49 -06:00
  • eb142491b2 Merge pull request #20 from eltociear/patch-1 Sebastian Raschka 2024-01-28 09:07:49 -06:00
  • 3a3a4ac1f1 Update bpe_openai_gpt2.py Ikko Eltociear Ashimine 2024-01-28 23:26:23 +09:00
  • f6e80a376d Fix link Sebastian Raschka 2024-01-26 20:30:20 -06:00
  • 04c113c54f Merge pull request #19 from Intelligence-Manifesto/patch-1 Sebastian Raschka 2024-01-25 17:41:49 -06:00
  • 004d6a2af9 a repetition of the word "by" Intelligence-Manifesto 2024-01-26 00:43:57 +08:00
  • 0eab244f00 add full toc rasbt 2024-01-24 07:09:35 -06:00
  • f27c9e6135 rename create_dataloader to create_dataloader_v1 rasbt 2024-01-24 07:02:05 -06:00
  • f6896d17ef image resizing rasbt 2024-01-21 20:25:21 -06:00
  • 4c45c37c51 Update README.md Sebastian Raschka 2024-01-21 20:17:32 -06:00