hoshi-hiyouga
13e1b7ee2b
[misc] fix grad ckpt ( #6931 )
...
Former-commit-id: c31c63b41109e616997757ec2da6e0ab89ed3b6e
2025-02-13 23:27:51 +08:00
hoshi-hiyouga
cd493b91de
[model] add liger kernel to qwen2_5 vl ( #6930 )
...
* add liger kernel to qwen2_5 vl
* fix patch
* fix patch
Former-commit-id: 797043d29cb85a8f90fabf48976908037f07000e
2025-02-13 23:05:54 +08:00
hoshi-hiyouga
036fb0d561
[misc] fix grad ckpt func ( #6916 )
...
Former-commit-id: e34c3c06da706f80c74c20800f19110e9ad6b82a
2025-02-13 00:17:18 +08:00
hoshi-hiyouga
ff6658ad27
[deps] upgrade vllm ( #6857 )
...
Former-commit-id: 5f38bcaba921dbdee27b4be4709fcec06fa37c9e
2025-02-08 15:02:28 +08:00
Zhangchi Feng
01915eaf40
[model] support audio ( #6701 )
...
* support qwen2_audio
* improve code
* lint
* fix
* fix
* fix
---------
Co-authored-by: hiyouga <hiyouga@buaa.edu.cn>
Former-commit-id: 24c78429489809873a1269a735ea5421340b32a2
2025-02-05 04:59:09 +08:00
hoshi-hiyouga
1fee69f874
[misc] update license year & fix llama pro ( #6814 )
...
* fix llamapro script
* change year
Former-commit-id: e2dc5b952aa22835d5220ba624f44676138b65ac
2025-02-05 01:53:33 +08:00
hoshi-hiyouga
445d643ef3
[model] add mistral small models ( #6786 )
...
Former-commit-id: 94803d8133fbbadff6d224cb6695feb5434fd4fd
2025-02-01 04:31:38 +08:00
hoshi-hiyouga
e8c1979b79
[model] add qwen2.5 vl models ( #6779 )
...
Former-commit-id: 999c7c8fe0caf6b837a1bdc2c6a24fafec327cd8
2025-01-31 03:00:29 +08:00
hoshi-hiyouga
f6779b0e0c
[breaking] support transformers 4.48 ( #6628 )
...
Former-commit-id: 15357cdad953bba1f2d294819f56b9746ed1b891
2025-01-31 01:36:33 +08:00
hoshi-hiyouga
1efe525df7
[model] support yarn ( #6693 )
...
Former-commit-id: 1f47b6186c267de86cbdbd47ba2adbf1f9db7f39
2025-01-18 13:56:09 +08:00
hoshi-hiyouga
788accb601
fix qwen2 moe ( #6684 )
...
Former-commit-id: 7bf09abf1c4d971cda33daed933c75f391e79294
2025-01-17 13:46:09 +08:00
hoshi-hiyouga
9ef85f8fc4
[optim] clean apollo ( #6645 )
...
* clean apollo code
* update readme
Former-commit-id: 7a04021d0461caea2c7b82169839340b7f51f463
2025-01-15 01:42:50 +08:00
zhuHQ
763f9b9df0
[optim] add support to APOLLO ( #6617 )
...
Former-commit-id: d9189f9f0b23ff6929044919208e0e813ca95b1c
2025-01-15 00:24:56 +08:00
Zhangchi Feng
ad119afc58
Support Inference of MiniCPM-V-2.6 and MiniCPM-o-2.6 ( #6631 )
...
* fix template name
* tiny fix
* support minicpm-o-2.6
* support inference of minicpmv
Former-commit-id: 158a127d340d5e4ca23263ffad042f861fd77deb
2025-01-14 17:34:58 +08:00
Zhangchi Feng
201a495154
Support new features of MiniCPM-V ( #6626 )
...
* fix template name
* tiny fix
* support minicpm-o-2.6
Former-commit-id: c3fda5046d835ba4542d525b8d89cd12838e9f4c
2025-01-14 00:26:19 +08:00
codingma
089c7d5e51
add nf4 qlora support on Ascend NPU ( #6601 )
...
* add nf4 qlora support on Ascend NPU
* add transformers version check
* add python>=3.10 requirement description for npu
* tiny fix
---------
Co-authored-by: hoshi-hiyouga <hiyouga@buaa.edu.cn>
Former-commit-id: 03de5ac912336190d6b3583f70b6340ab9cf9cdf
2025-01-13 19:43:36 +08:00
fzc8578
08e8499a98
adapt to new mllm_param
...
Former-commit-id: 291384dea8a5c10f0358a30d124eaf85557548eb
2025-01-11 00:16:34 +08:00
Zhangchi Feng
d5b18ee4a6
Merge branch 'main' into minicpmv
...
Former-commit-id: ed0895a9c13b0ea8a5cace6b060f01d9771816ad
2025-01-11 00:01:36 +08:00
hiyouga
c89d17ab63
refactor mllm param logic
...
Former-commit-id: f6f630a1c96514053176abb12e35a06242e62abd
2025-01-10 15:45:48 +00:00
fzc8578
7138b43873
fix some
...
Former-commit-id: 2ee8ba2f390551af1b865cfa813f5c8b7bbb41c5
2025-01-10 20:27:06 +08:00
fzc8578
aeb4f82ef2
tiny fix
...
Former-commit-id: 84026be06e34239a828a0cc8b1706084afcfa4ea
2025-01-10 20:15:39 +08:00
Zhangchi Feng
f51ac40f0a
Merge branch 'main' into minicpmv
...
Former-commit-id: fc045d7dd871985d621430b5662cba882188a59c
2025-01-10 20:12:07 +08:00
fzc8578
165fe8e219
add some
...
Former-commit-id: 096a6cb67a7dfd14a6e339d96baab78c12d36a87
2025-01-10 20:01:22 +08:00
hiyouga
da542fad18
imporve log
...
Former-commit-id: 47e17dd689840ca9b3c5f34448e5f80265336cca
2025-01-08 09:56:10 +00:00
fzc8578
b9eeaa9706
add some
...
Former-commit-id: 785cc70ff205f5962c3ca67f453589e4a471ba8c
2025-01-06 19:32:39 +08:00
fzc8578
b5ef5059ee
add some
...
Former-commit-id: 79c2d7090cbf364063ea3608814ab18aa27fdc87
2025-01-04 11:11:15 +08:00
Yaser Afshar
fe4546a7bb
Add trust_remote_code parameter and remove True
...
- Introduced a new model parameter `trust_remote_code`
- Set the default value of `trust_remote_code` to `False`
to enhance security
Former-commit-id: 09437763267bc7081159a6878cee9652a2b1ddac
2024-12-17 12:25:12 +00:00
hoshi-hiyouga
fc18db6290
Merge pull request #6364 from hiyouga/hiyouga/control_reenterent_gc
...
[model] support non-reenterent-gc
Former-commit-id: a665ad6178516faf8aaa628d3b2c672ad831d7b7
2024-12-17 19:58:36 +08:00
hiyouga
64bac4bc7e
support non-reenterent-gc & fix #6358
...
Former-commit-id: f319da6937964c5068864f7815efbf3847bd4ad8
2024-12-17 11:41:59 +00:00
hiyouga
bff1b94583
generalized packing & fix #6343
...
Former-commit-id: 2d107d3aefd5af61163056634c8b91fe3cb3e77c
2024-12-17 10:26:19 +00:00
hiyouga
cf8cad8e7e
support qwen2vl train proj only
...
Former-commit-id: 99c62660c67fd2567c094bb49ec7d108e5670730
2024-12-05 10:37:42 +00:00
hoshi-hiyouga
a489f10986
fix visual patch
...
Former-commit-id: 75b586c31acf47d1bd28e04566ffd1d954e45596
2024-11-25 20:06:06 +08:00
hoshi-hiyouga
f48d07cd6c
fix #6136
...
Former-commit-id: 0516e556a71a22b8767b17734adb94eb127e7e6f
2024-11-25 19:43:42 +08:00
hiyouga
253752ccca
add forbidden modules
...
Former-commit-id: df477370dc67315effac1a8f48068c4e9c4067a5
2024-11-23 18:34:15 +00:00
hiyouga
e99031daa4
fix inputs
...
Former-commit-id: 446441fdb020b5a102480251cb8536dd8b3f8f99
2024-11-23 18:26:02 +00:00
marko1616
23fc0c863e
Linter.
...
Former-commit-id: b1e43e56db4d898c0d6d27b703b82a1eb9af70c2
2024-11-23 16:09:04 +00:00
marko1616
3295519099
Tiny fix.
...
Former-commit-id: 8372c5e3771c42f225d7bd80a758af920f80e893
2024-11-23 16:09:01 +00:00
hiyouga
e83cb17f97
support rank0 logger
...
Former-commit-id: c38aa29336f286266553da4909a7267d7ef21f37
2024-11-02 18:31:04 +08:00
hiyouga
3f7c874594
update tests
...
Former-commit-id: 93d3b8f43faf4a81b809d2f7d897e39bdb5475c3
2024-11-02 12:41:44 +08:00
hiyouga
584ce3a105
fix incorrect loss value for vlms
...
Former-commit-id: 30567a1487727473950104718e626ff660f10cbb
2024-10-30 08:56:46 +00:00
hoshi-hiyouga
2179b91acb
Update visual.py
...
Former-commit-id: 0baa7735f64cbef9bd90e1db485c120b4c1c88bd
2024-10-29 22:10:29 +08:00
Kingsley
3053a806e9
Merge branch 'hiyouga:main' into pixtral-patch
...
Former-commit-id: 67f59579d79e97689a4b3cba7101a423c30dab2b
2024-10-29 21:01:25 +08:00
hiyouga
0d8aa6e6ef
use pre-commit
...
Former-commit-id: 21db8ed2f4a0eba203754a92ce0741538e8ee709
2024-10-29 09:07:46 +00:00
hiyouga
163cf2ba5c
update requires
...
Former-commit-id: 77666bd2278a3cfe5b567f4fe285b0f93871d166
2024-10-29 16:10:07 +08:00
Kingsley
5523a6fd2c
Merge branch 'hiyouga:main' into pixtral-patch
...
Former-commit-id: 93a441a6b746e9a933dad8c45553fb5b68bf2b34
2024-10-08 21:04:08 +08:00
hiyouga
4464a6ff5b
tiny fix
...
Former-commit-id: 451d271718a8026056d0f7d7b8ab333391d24ad4
2024-10-08 17:48:56 +08:00
Kingsley
f3ac97a749
Merge branch 'hiyouga:main' into pixtral-patch
...
Former-commit-id: e53f47c0b3de491d4d9b31c995f9cea100f98896
2024-10-01 00:52:31 +08:00
hiyouga
4df090ff48
fix #5542
...
Former-commit-id: fe7ffccdb9a45b31e20ab7e88282a75b45504a97
2024-09-30 23:28:55 +08:00
Kingsley
6729ed2c7e
sync with former
...
Former-commit-id: 9ddb84052e3cc72e21a92b8103caa179a35859c4
2024-09-30 20:27:05 +08:00
Kingsley
94ce8f561f
fix some errors due to inconsistency of model cards
...
Former-commit-id: 2166b9bc6ba35760ff85b63620af9fa0213a4c78
2024-09-30 19:58:34 +08:00
