46 Commits

Author SHA1 Message Date
Yaowei Zheng
5817583630
[deps] bump transformers to 4.49.0 (#8564) 2025-07-07 20:31:50 +08:00
Redwood-Digital
1b549e3199
[docs] add nvidia-container-toolkit to Linux Docker setup instructions (#8557) 2025-07-06 19:37:08 +08:00
wjunLu
d30cbcdfa5
[ci] Add workflow for building NPU image (#8546)
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>
2025-07-04 20:56:59 +08:00
Yaowei Zheng
cecba57b3e
[assets] fix npu docker (#8298) 2025-06-05 19:09:20 +08:00
Yaowei Zheng
81c4d9bee6
[assets] update docker files (#8291) 2025-06-04 23:30:46 +08:00
hoshi-hiyouga
a4048b7bb6
[assets] fix docker images (#8203) 2025-05-28 22:26:05 +08:00
hoshi-hiyouga
55d37dc472
[assets] update Dockerfile (#8201) 2025-05-28 20:20:59 +08:00
hoshi-hiyouga
07f792148e
[assets] fix docker image (#8180) 2025-05-27 19:01:31 +08:00
hoshi-hiyouga
519ac92803
[assets] update docker files (#8176) 2025-05-27 18:15:23 +08:00
Joe Schoonover
b5d667cebf
[docker] patch docker-rocm (#7725)
* Update Dockerfile

* Fix typo

* Fix syntax for /bin/sh conditional

* Add build args to docker-compose

* Change shell to /bin/bash

This is required for "==" syntax in conditional string comparison
2025-04-15 13:36:39 +08:00
rumichi
747e02d60d
[docker] upgrade to torch 2.6 (#7442) 2025-03-23 21:18:08 +08:00
hoshi-hiyouga
5a0fd22c05 [assets] update readme (#7209)
Former-commit-id: cdf8fc647819523eca8139ba62a510c3078b694d
2025-03-07 17:27:49 +08:00
leo-pony
e86cb8a4fa [npu] update cann base image and torch 2.4 (#7061)
* Update base npu container image version:The Python version required for Hugging Face Transformers is >= python3.10

* Fix the bug: arg type of INSTALL_DEEPSPEED shoud been string now.

* Update Ascend CANN, CANN-Kernel and corresponding torch and torch-npu version

* Upgrade torch-npu needs packages' version: torch==2.1.0 and torch-npu==2.4.0.post2

Former-commit-id: acc52e0fe75d59c2581ef45346ebfbffe45f762b
2025-02-25 23:32:01 +08:00
hiyouga
a02a140840 update wechat
Former-commit-id: cd1433650653810f7934c65cb1de91052eb73dcf
2025-01-04 07:59:57 +00:00
shibingli@yeah.net
c76c33ddb1 Add ARG HTTP_PROXY in Dockerfile to support HTTP proxy during image building.
Former-commit-id: f1d76786e094562f6f095a0b56c9c6cd32e2fa5e
2024-12-27 18:31:14 +08:00
shibingli@yeah.net
a37ef0eaae Add ARG HTTP_PROXY in Dockerfile to support HTTP proxy during image building.This commit introduces an ARG parameter named HTTP_PROXY in the Dockerfile. This addition allows for the configuration of an HTTP proxy, facilitating image building in environments with network restrictions.
Former-commit-id: a3a49b1ea477313c979a1649ee6a7f843fe36469
2024-12-27 18:17:17 +08:00
XYZliang
1c1e48a570 Increase shm_size to 16GB in docker-compose.yml to optimize shared memory allocation for large-scale model fine-tuning tasks.
This pull request increases the shm_size parameter in docker-compose.yml to 16GB. The goal is to enhance the LLaMA-Factory framework’s performance for large model fine-tuning tasks by providing sufficient shared memory for efficient data loading and parallel processing.

This PR also addresses the issues discussed in [this comment](https://github.com/hiyouga/LLaMA-Factory/issues/4316#issuecomment-2466270708) regarding Shared Memory Limit error.


Former-commit-id: 64414905a3728abf3c51968177ffc42cfc653310
2024-11-13 10:13:59 +08:00
hoshi-hiyouga
5ad86fc71d Update Dockerfile
Former-commit-id: 5e8a619d1d20048ad912a1f6d9ccdae5740811bc
2024-11-02 20:20:26 +08:00
Valerio Mariani
1ab3ea21ce make base image parametric.
default `BASE_IMAGE` is nvcr.io/nvidia/pytorch:24.02-py3 for retro-compatibility


Former-commit-id: d5b2def7c029d6f808a97cb062340810e0cd075d
2024-10-30 21:53:32 +01:00
hiyouga
e90a1199da tiny fix
Former-commit-id: 3af57795dda5d236200bad4aa3f2e29ae8930fe2
2024-10-11 23:51:54 +08:00
StrangeBytesDev
d1e766898c Add additional install options to Dockerfiles
Former-commit-id: 237e302b5ccf029dec0c7f859c40f122bf92e1c5
2024-09-24 16:54:46 -07:00
HardAndHeavy
a39dc400ed update the ROCm version to 6.2
Former-commit-id: b1f9ccb29f6899406253cdde1f3546d6a7353669
2024-09-12 23:46:33 +03:00
MengqingCao
a620fa8be3 update npu base image
Former-commit-id: b3f4acd1b4636d15afdfde61d77af521fff39b19
2024-08-21 09:12:38 +00:00
hiyouga
884b0bbb4f tiny fix
Former-commit-id: 668654b5adae3f897d5291b81410226e1304eff9
2024-07-26 11:51:00 +08:00
HardAndHeavy
27f42f6319 Add ROCm support
Former-commit-id: c8e18a669adc775f17555cbf06a5ceef6c0d6235
2024-07-25 21:29:28 +03:00
hzhaoy
93ba3bd5b0 fix #4780
Former-commit-id: 642c6d666f3bd00fcdea45c65a6394bcae9c2080
2024-07-12 00:25:48 +08:00
hzhaoy
b3e4793ded fix #4779
Former-commit-id: a8bf1abf0fd39f84748c94ac3ba39eaa53137529
2024-07-12 00:15:15 +08:00
hiyouga
4b1ab6c83d update npu docker
Former-commit-id: 93e6fbb37ddb60ed356d0acc720463a8cd1a8baa
2024-06-30 21:05:31 +08:00
hiyouga
cba994b9dc fix docker flashattn
Former-commit-id: e3141f5f1b435d12c71d8b1fc6ade6e69deead71
2024-06-28 01:28:59 +08:00
hiyouga
5e60c29971 fix docker file
Former-commit-id: 0f421055da37d7f9b3313954fd008c433b3100f5
2024-06-27 20:29:16 +08:00
hiyouga
7c488cea57 tiny fix
Former-commit-id: e44a4f07f09bbee55c10ccee91dd858256c36054
2024-06-27 20:14:48 +08:00
hoshi-hiyouga
37d3adb1f8 Merge pull request #4461 from hzhaoy/feature/support-flash-attn
support flash-attn in Dockerfile

Former-commit-id: 64b131dcfa381045cba6b77ab9e0dbf6a3934e03
2024-06-27 20:05:26 +08:00
hoshi-hiyouga
ffe7cda93d Update Dockerfile
Former-commit-id: 5fcd33fd4f08f5e46c1df541f188d13faefa7c8e
2024-06-27 19:57:40 +08:00
hoshi-hiyouga
030dda5861 Update Dockerfile
Former-commit-id: 9624af1df3e4b260b37ad10e2a9be8e69ab066df
2024-06-27 19:51:25 +08:00
fanjunliang
9e93ff9479 support docker-npu-[amd64|arm64] build
Former-commit-id: bdda0827b35cbb6005f10aa661fe6f3940b598d8
2024-06-27 15:25:12 +08:00
hzhaoy
c662c2e56f add flash-attn installation flag in Dockerfile
Former-commit-id: e19491b0f0446f2fb2154cf14e0b2fbba5b54808
2024-06-27 00:13:30 +08:00
fanjunliang
e7bd3ab6c3 fix torch-npu dependency
Former-commit-id: 8096f94a7db3e357c52906c83cad9f3c360bad7e
2024-06-26 18:21:42 +08:00
MengqingCao
7c7d6614d8 fix docker-compose path
Former-commit-id: 106647a99d36c4a624511edf651cf5f7624c4c9a
2024-06-26 02:15:00 +00:00
hzhaoy
08a221443c support flash-attn in Dockerfile
Former-commit-id: c88b1be9f3dfaf5fe65448dea20fc697b4f257bd
2024-06-25 15:13:07 +08:00
hiyouga
dafc9268bc fix #4419
Former-commit-id: efb81b25ecd5cb9f4cfda8f2da8b159e4ab26a90
2024-06-25 01:51:29 +08:00
hoshi-hiyouga
af5b2b9299 Update docker-compose.yml
Former-commit-id: 721acd87688b3edd41d2e1890971eb387afc1b6e
2024-06-25 00:54:28 +08:00
hoshi-hiyouga
6cd45e95f7 Update Dockerfile
Former-commit-id: 3af936a76d726c73d14ff6cf8c3088bda5b904c3
2024-06-25 00:50:34 +08:00
hoshi-hiyouga
62e63d74ec Update docker-compose.yml
Former-commit-id: 15608d055827e41b44fb8a381b1fe5a1230afcb0
2024-06-25 00:46:47 +08:00
hoshi-hiyouga
cfa2dbefcb Update Dockerfile
Former-commit-id: fce146ab681a5682d3f96254e3863f70aca8a4af
2024-06-25 00:46:08 +08:00
hoshi-hiyouga
f84bce3638 Update Dockerfile
Former-commit-id: dcc2e24f5cd441a242829f2fb090506db507babd
2024-06-24 23:41:35 +08:00
MengqingCao
3b499948a5 update docker files
1. add docker-npu (Dockerfile and docker-compose.yml)
  2. move cuda docker to docker-cuda and tiny changes to adapt to the new path


Former-commit-id: d7207e8ad10c7df6dcb1f5e59ff8eb06f9d77e67
2024-06-24 10:57:36 +00:00