346 Commits

Author SHA1 Message Date
rasbt
beeaf323f1 rename download_and_unzip to make it more specific 2024-05-12 08:36:24 -05:00
Sebastian Raschka
49306b271f add header 2024-05-11 14:37:21 -05:00
rasbt
84edcfaf43 use spam / not spam labels 2024-05-11 13:42:18 -05:00
speed
7b34833ee1 fix 1024 characters to 1024 tokens (#152) 2024-05-11 13:17:07 -05:00
rasbt
c94f24e759 reorder section 6.6 2024-05-11 08:27:07 -05:00
rasbt
db29f5c685 explain how class labels are obtained 2024-05-11 07:42:13 -05:00
Sebastian Raschka
03d3b6ca72 Update README.md 2024-05-11 06:42:05 -05:00
Sebastian Raschka
cf299777b6 Move GPU column so that test accuracy is always visible 2024-05-11 06:41:25 -05:00
Sebastian Raschka
cfa1d0f997 Add LoRA experiments (#151)
* Add LoRA experiments

* Update ch06/02_bonus_additional-experiments/additional-experiments.py
2024-05-10 07:26:41 -05:00
rasbt
774974de97 6 -> 4 2024-05-10 07:02:14 -05:00
rasbt
d8de9377de Merge branch 'main' of https://github.com/rasbt/LLMs-from-scratch 2024-05-10 06:45:14 -05:00
Sebastian Raschka
216dd010f6 fix punctuation and improve explanation 2024-05-09 21:15:09 -05:00
rasbt
dadd0f7ea3 clarify overfitting 2024-05-09 09:09:26 -05:00
rasbt
2b3d86fe9a Merge branch 'main' of https://github.com/rasbt/LLMs-from-scratch 2024-05-09 08:17:43 -05:00
Sebastian Raschka
ad200a4f3f experiments with largest model (#149) 2024-05-09 07:48:56 -05:00
rasbt
75545e4c1b experiments with largest model 2024-05-09 07:40:09 -05:00
rasbt
1638dc8b7f spelling improvements 2024-05-09 07:25:52 -05:00
rasbt
1e34f5a429 add note about worker number 2024-05-08 21:20:43 -05:00
rasbt
1e7d1f3bcb update figure 6.6 2024-05-08 20:46:54 -05:00
rasbt
a31d571625 text -> dataset 2024-05-08 08:14:03 -05:00
rasbt
6cc9cf9f4e make spam spelling consistent 2024-05-08 06:48:28 -05:00
Sebastian Raschka
41ff2ae4c7 Explain hardware requirements 2024-05-07 20:47:06 -05:00
rasbt
7f841611df fix sentence 2024-05-07 07:05:47 -05:00
rasbt
d0d79103d5 add additional lora figure 2024-05-07 07:04:35 -05:00
rasbt
73c29fd9db spelling fix 2024-05-07 06:46:33 -05:00
rasbt
85050ace50 spelling and consistency improvements 2024-05-06 21:02:13 -05:00
rasbt
7082ecac80 formatting improvements 2024-05-06 20:35:51 -05:00
rasbt
0448162fdc show downloads 2024-05-06 07:40:09 -05:00
rasbt
78829f28e9 tokenizing example 2024-05-06 07:16:40 -05:00
rasbt
15d6f29cf8 ch06 csv 2024-05-06 07:16:30 -05:00
rasbt
c6528ede9e ch06 dataset 2024-05-06 06:55:56 -05:00
rasbt
e574d04eba classfication -> classification 2024-05-06 06:50:38 -05:00
rasbt
9457676640 ouput -> output 2024-05-05 12:21:10 -05:00
Ikko Eltociear Ashimine
d361cef65f Update ch06.ipynb (#143)
ouput -> output
2024-05-05 12:18:20 -05:00
Sebastian Raschka
dd31946b2a Appendix E: Parameter-efficient Finetuning with LoRA (#142) 2024-05-05 12:05:17 -05:00
rasbt
a63b0f626c make code more general for larger models 2024-05-05 10:18:46 -05:00
Sebastian Raschka
3328b29521 cosmetics 2024-05-05 08:15:46 -05:00
rasbt
244593ce01 add text-to-token-id fn 2024-05-05 08:05:20 -05:00
Sebastian Raschka
c6fcadb087 Add figures for ch06 (#141) 2024-05-05 07:10:04 -05:00
rasbt
f917fc76fe update link 2024-05-04 08:08:58 -05:00
rasbt
2c06f824aa table-update 2024-05-04 07:58:18 -05:00
rasbt
97106950c1 add description 2024-05-04 07:34:29 -05:00
Sebastian Raschka
004b0614fc Ch06 draft (#138)
* Ch06 first draft

* add utility files
2024-05-03 08:37:58 -05:00
rasbt
9e149417b2 fix swiglu acronym 2024-05-01 20:26:17 -05:00
rasbt
bb59cbc525 link formatting 2024-04-30 06:26:23 -05:00
rasbt
c5886b7865 Merge branch 'main' of https://github.com/rasbt/LLMs-from-scratch 2024-04-30 06:25:37 -05:00
Sebastian Raschka
8d84800bcf use training set len (#137) 2024-04-29 21:56:05 -05:00
rasbt
354bb35726 use training set len 2024-04-29 21:50:07 -05:00
Sebastian Raschka
a5b353667d Rename drop_resid to drop_shortcut (#136) 2024-04-28 14:31:27 -05:00
Sebastian Raschka
d1edfcb63f add roberta option (#135) 2024-04-28 13:57:36 -05:00