hiyouga
|
8cafc7b055
|
video datasets
|
2024-09-05 02:04:17 +08:00 |
liudan
|
d7ba97be48
|
增加了对minicpm3.0的适配'
|
2024-09-04 23:10:05 +08:00 |
|
hiyouga
|
c5e1736470
|
fix test
|
2024-09-04 22:38:26 +08:00 |
|
hiyouga
|
dabad5570b
|
update get template
|
2024-09-04 22:36:20 +08:00 |
hoshi-hiyouga
|
8f441c2b3a
|
Merge pull request #5323 from naem1023/feat/add-dataset-map-batch-size-argument
Add batch size of map function in the preprocessed dataset
|
2024-09-04 22:09:36 +08:00 |
|
hoshi-hiyouga
|
44d6947e55
|
fix #5228
|
2024-09-04 19:10:30 +08:00 |
|
hiyouga
|
ebddce5d4a
|
fix #5252
|
2024-09-04 03:17:54 +08:00 |
|
hiyouga
|
57497135bf
|
add vl_feedback dataset
|
2024-09-04 03:13:03 +08:00 |
|
hiyouga
|
d41d43a7c3
|
fix #5344
|
2024-09-04 03:06:06 +08:00 |
|
hoshi-hiyouga
|
ce7ed6eb99
|
Merge pull request #5346 from hiyouga/lazy_image
[exp] Lazyload for multimodal inputs
|
2024-09-04 03:00:53 +08:00 |
|
hiyouga
|
47ea97fb1b
|
lazy image load
|
2024-09-04 02:27:08 +08:00 |
|
hiyouga
|
59d2b31e96
|
fix #5334
|
2024-09-03 19:09:42 +08:00 |
|
hiyouga
|
69d0acacc3
|
fix #5338
|
2024-09-03 17:45:17 +08:00 |
|
hiyouga
|
22959bcdd3
|
lint
|
2024-09-03 00:46:25 +08:00 |
|
hiyouga
|
a61c8c4890
|
fix #5324
|
2024-09-02 23:56:21 +08:00 |
naem1023
|
209313eeea
|
feat: add batch size of map function in the preprocessed dataset
|
2024-09-02 13:52:47 +09:00 |
|
hoshi-hiyouga
|
99fd9637bd
|
fix trainer predict
|
2024-09-02 10:15:29 +08:00 |
|
hoshi-hiyouga
|
a6c6750e8a
|
remove .cpu()
|
2024-09-02 10:10:53 +08:00 |
|
hiyouga
|
60fc6b926e
|
fix mm inference
|
2024-09-02 01:47:40 +08:00 |
|
hiyouga
|
3a6f19f017
|
tiny fix
|
2024-09-02 01:33:22 +08:00 |
|
hiyouga
|
ce8c5a2647
|
add image num check
|
2024-09-02 01:31:36 +08:00 |
|
hiyouga
|
194064fdae
|
add pokemon dataset
|
2024-09-02 01:02:25 +08:00 |
|
hiyouga
|
a8f8a2ad8a
|
update readme
|
2024-09-01 23:32:39 +08:00 |
|
hiyouga
|
753c5fb36c
|
update wechat
|
2024-09-01 23:30:57 +08:00 |
|
hoshi-hiyouga
|
6f9e455af4
|
Merge pull request #5317 from ByronHsu/patch-1
Add liger kernel link
|
2024-09-01 23:30:12 +08:00 |
|
hiyouga
|
8e49940746
|
add rlhf-v dataset
|
2024-09-01 22:57:41 +08:00 |
|
hiyouga
|
55027282cd
|
tiny fix
|
2024-09-01 21:15:44 +08:00 |
|
hiyouga
|
64cb947c60
|
fix bug
|
2024-09-01 21:07:49 +08:00 |
|
hiyouga
|
9967ccb3ae
|
fix mixed mm inputs and rlhf-v
|
2024-09-01 20:52:47 +08:00 |
Byron Hsu
|
b8a9cb554e
|
Add liger kernel link
|
2024-08-30 17:16:16 -07:00 |
|
hiyouga
|
b5063b4144
|
fix ci (temp)
|
2024-08-31 02:03:56 +08:00 |
|
hiyouga
|
a2a8c0b92c
|
add test mm plugin
|
2024-08-31 01:53:38 +08:00 |
|
hiyouga
|
a025c3df61
|
remove visual_inputs, fix qlora
|
2024-08-31 00:24:51 +08:00 |
|
hiyouga
|
a244f143f4
|
optimize predict vram
|
2024-08-30 23:08:45 +08:00 |
|
hiyouga
|
e08045a946
|
add examples
|
2024-08-30 21:43:19 +08:00 |
|
hiyouga
|
bee1bd43b9
|
tiny fix
|
2024-08-30 03:21:50 +08:00 |
|
hiyouga
|
8b588c7224
|
fix #5307
|
2024-08-30 02:45:40 +08:00 |
|
hiyouga
|
3382317e32
|
refactor mm training
|
2024-08-30 02:14:31 +08:00 |
|
hoshi-hiyouga
|
727e184840
|
Merge pull request #5290 from simonJJJ/qwen2_vl
support qwen2-vl
|
2024-08-30 02:10:36 +08:00 |
|
hoshi-hiyouga
|
a8f22d8895
|
fix bug
|
2024-08-30 02:05:26 +08:00 |
|
hiyouga
|
a7dd7d325e
|
update liger kernel
|
2024-08-29 20:46:08 +08:00 |
|
hiyouga
|
aa1afdc756
|
fix #5292
|
2024-08-29 20:37:47 +08:00 |
|
hiyouga
|
ad72f3e065
|
fix #5295
|
2024-08-29 20:30:18 +08:00 |
|
hiyouga
|
364b757e30
|
fix #5305
|
2024-08-29 20:16:01 +08:00 |
simonJJJ
|
734e019cc1
|
update
|
2024-08-28 20:22:46 +08:00 |
|
simonJJJ
|
aeb85f200b
|
initial-commit
|
2024-08-28 16:51:35 +08:00 |
|
hiyouga
|
0f5a0f64f7
|
update wechat
|
2024-08-27 12:55:23 +08:00 |
|
hiyouga
|
d14edd350d
|
add extra requires
|
2024-08-27 12:52:12 +08:00 |
|
hiyouga
|
f6ae4e75dd
|
tiny fix
|
2024-08-27 12:49:32 +08:00 |
|
hoshi-hiyouga
|
dbe886ae5c
|
Merge pull request #5237 from marko1616/patch-1
Fix mllm api
|
2024-08-27 12:24:43 +08:00 |