 Sebastian Raschka
|
08040f024c
Test code in pytorch 2.4 (#285)
|
1 year ago |
|
Sebastian Raschka
|
72a073bbbf
Remove leftover instances of self.tokenizer (#201)
|
1 year ago |
|
rasbt
|
050c8b7b73
update pr
|
1 year ago |
 Kostyantyn Borysenko
|
76cdf5e299
Fix an incorrect input dimension
|
1 year ago |
|
rasbt
|
98d453b666
update formatting
|
1 year ago |
 James Holcombe
|
05718c6b94
Use instance tokenizer (#116)
|
1 year ago |
|
Sebastian Raschka
|
2de60d1bfb
Rename variable to context_length to make it easier on readers (#106)
|
1 year ago |
|
rasbt
|
3ad442ee90
skip version cell
|
1 year ago |
|
Sebastian Raschka
|
a2cd8436cb
Ch05 supplementary code (#81)
|
1 year ago |
|
Sebastian Raschka
|
ca96abac8a
Set up basic test gh worklows (#79)
|
1 year ago |
|
rasbt
|
1870b4bacd
update stride param
|
1 year ago |
|
rasbt
|
da33ce8054
remove redundant unsqueeze in mask
|
1 year ago |
|
rasbt
|
87fcfd9245
mha variants
|
1 year ago |
|
rasbt
|
d4754f1bdd
change dim=1 to dim=-1
|
1 year ago |
|
rasbt
|
b827bf4eea
remove redundant double-unsequeeze
|
1 year ago |
|
rasbt
|
8860e16e05
<|endoftext|> token in dataset v1
|
1 year ago |
|
rasbt
|
92896d817c
add toggle for qkv_bias
|
1 year ago |
|
rasbt
|
dfe2c3b46f
use blocksize in positional embedding
|
1 year ago |
|
rasbt
|
9e85f13ba9
readability improvements
|
1 year ago |
|
rasbt
|
a7b4880179
small readability updates
|
1 year ago |
|
rasbt
|
4f161bd549
use block size variable in positional embedding layer
|
1 year ago |
|
rasbt
|
31980a6ef1
add ch03 and TOC
|
1 year ago |