* add SimPO
* fix dataloader
* remove debug code
* add orpo
* fix style
* fix colossalai, transformers version
* fix colossalai, transformers version
* fix colossalai, transformers version
* fix torch colossalai version
* update transformers version
* [shardformer] DeepseekMoE support (#5871)
* [Feature] deepseek moe expert parallel implement
* [misc] fix typo, remove redundant file (#5867)
* [misc] fix typo
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [Feature] deepseek support & unit test
* [misc] remove debug code & useless print
* [misc] fix typos (#5872)
* [Feature] remove modeling file, use auto config. (#5884)
* [misc] fix typos
* [Feature] deepseek support via auto model, remove modeling file
* [misc] delete useless file
* [misc] fix typos
* [Deepseek] remove redundant code (#5888)
* [misc] fix typos
* [Feature] deepseek support via auto model, remove modeling file
* [misc] delete useless file
* [misc] fix typos
* [misc] remove redundant code
* [Feature/deepseek] resolve comment. (#5889)
* [misc] fix typos
* [Feature] deepseek support via auto model, remove modeling file
* [misc] delete useless file
* [misc] fix typos
* [misc] remove redundant code
* [misc] mv module replacement into if branch
* [misc] add some warning message and modify some code in unit test
* [misc] fix typos
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [Hoxfix] Fix CUDA_DEVICE_MAX_CONNECTIONS for comm overlap
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* [Feat] Diffusion Model(PixArtAlpha/StableDiffusion3) Support (#5838)
* Diffusion Model Inference support
* Stable Diffusion 3 Support
* pixartalpha support
* [HotFix] CI,import,requirements-test for #5838 (#5892)
* [Hot Fix] CI,import,requirements-test
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [Feature] Enable PP + SP for llama (#5868)
* fix cross-PP-stage position id length diff bug
* fix typo
* fix typo
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* use a one cross entropy func for all shardformer models
---------
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [ShardFormer] Add Ulysses Sequence Parallelism support for Command-R, Qwen2 and ChatGLM (#5897)
* add benchmark for sft, dpo, simpo, orpo. Add benchmarking result. Support lora with gradient checkpoint
* fix style
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix eval
* hotfix citation
* [zero] support all-gather overlap (#5898)
* [zero] support all-gather overlap
* [zero] add overlap all-gather flag
* [misc] fix typo
* [zero] update api
* fix orpo cross entropy loss
* [Auto Parallel]: Speed up intra-op plan generation by 44% (#5446)
* Remove unnecessary calls to deepcopy
* Build DimSpec's difference dict only once
This change considerably speeds up construction speed of DimSpec objects. The difference_dict is the same for each DimSpec object, so a single copy of it is enough.
* Fix documentation of DimSpec's difference method
* [ShardFormer] fix qwen2 sp (#5903)
* [compatibility] support torch 2.2 (#5875)
* Support Pytorch 2.2.2
* keep build_on_pr file and update .compatibility
* fix object_to_tensor usage when torch>=2.3.0 (#5820)
* [misc] support torch2.3 (#5893)
* [misc] support torch2.3
* [devops] update compatibility ci
* [devops] update compatibility ci
* [devops] add debug
* [devops] add debug
* [devops] add debug
* [devops] add debug
* [devops] remove debug
* [devops] remove debug
* [release] update version (#5912)
* [plugin] support all-gather overlap for hybrid parallel (#5919)
* [plugin] fixed all-gather overlap support for hybrid parallel
* add kto
* fix style, add kto data sample
* [Examples] Add lazy init to OPT and GPT examples (#5924)
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* [ColossalChat] Hotfix for ColossalChat (#5910)
* add ignore and tiny llama
* fix path issue
* run style
* fix issue
* update bash
* add ignore and tiny llama
* fix path issue
* run style
* fix issue
* update bash
* fix ddp issue
* add Qwen 1.5 32B
* refactor tokenization
* [FIX BUG] UnboundLocalError: cannot access local variable 'default_conversation' where it is not associated with a value (#5931)
* cannot access local variable 'default_conversation' where it is not associated with a value
set default value for 'default_conversation'
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fix test data
* refactor evaluation
* remove real data path
* remove real data path
* Add n_fused as an input from native_module (#5894)
* [FIX BUG] convert env param to int in (#5934)
* [Hotfix] Fix ZeRO typo #5936
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* [Feature] Add a switch to control whether the model checkpoint needs to be saved after each epoch ends (#5941)
* Add a switch to control whether the model checkpoint needs to be saved after each epoch ends
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fix style
* fix style
* fix style
* [shardformer] hotfix attn mask (#5945)
* [shardformer] hotfix attn mask (#5947)
* [Feat] Distrifusion Acceleration Support for Diffusion Inference (#5895)
* Distrifusion Support source
* comp comm overlap optimization
* sd3 benchmark
* pixart distrifusion bug fix
* sd3 bug fix and benchmark
* generation bug fix
* naming fix
* add docstring, fix counter and shape error
* add reference
* readme and requirement
* [zero] hotfix update master params (#5951)
* [release] update version (#5952)
* [Chat] Fix lora (#5946)
* fix merging
* remove filepath
* fix style
* Update README.md (#5958)
* [hotfix] Remove unused plan section (#5957)
* remove readme
* fix readme
* update
* [test] add mixtral for sequence classification
* [test] add mixtral transformer test
* [moe] fix plugin
* [test] mixtra pp shard test
* [chore] handle non member group
* [zero] solve hang
* [test] pass mixtral shardformer test
* [moe] implement transit between non moe tp and ep
* [zero] solve hang
* [misc] solve booster hang by rename the variable
* solve hang when parallel mode = pp + dp
* [moe] implement submesh initialization
* [moe] add mixtral dp grad scaling when not all experts are activated
* [chore] manually revert unintended commit
* [chore] trivial fix
* [chore] arg pass & remove drop token
* [test] add mixtral modelling test
* [moe] implement tp
* [moe] test deepseek
* [moe] clean legacy code
* [Feature] MoE Ulysses Support (#5918)
* moe sp support
* moe sp bug solve
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [chore] minor fix
* [moe] init moe plugin comm setting with sp
* moe sp + ep bug fix
* [moe] finalize test (no pp)
* [moe] full test for deepseek and mixtral (pp + sp to fix)
* [chore] minor fix after rebase
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* [chore] solve moe ckpt test failure and some other arg pass failure
* [moe] remove ops
* [test] fix test: test_zero1_2
* [bug] fix: somehow logger hangs the program
* [moe] deepseek moe sp support
* [test] add check
* [deepseek] replace attn (a workaround for bug in transformers)
* [misc] skip redunant test
* [misc] remove debug/print code
* [moe] refactor mesh assignment
* Revert "[moe] implement submesh initialization"
This reverts commit 2f9bce6686
.
* [chore] change moe_pg_mesh to private
* [misc] remove incompatible test config
* [misc] fix ci failure: change default value to false in moe plugin
* [misc] remove useless condition
* [chore] docstring
* [moe] remove force_overlap_comm flag and add warning instead
* [doc] add MoeHybridParallelPlugin docstring
* [moe] solve dp axis issue
* [chore] remove redundant test case, print string & reduce test tokens
* [feat] Dist Loader for Eval (#5950)
* support auto distributed data loader
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* support auto distributed data loader
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix tp error
* remove unused parameters
* remove unused
* update inference
* update docs
* update inference
---------
Co-authored-by: Michelle <qianranma8@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [lora] lora support hybrid parallel plugin (#5956)
* lora support hybrid plugin
* fix
* fix
* fix
* fix
* Support overall loss, update KTO logging
* [Docs] clarify launch port
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* [Hotfix] README link (#5966)
* update ignore
* update readme
* run style
* update readme
* [Hotfix] Avoid fused RMSnorm import error without apex (#5985)
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* [Chat] fix readme (#5989)
* fix readme
* fix readme, tokenization fully tested
* fix readme, tokenization fully tested
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: root <root@notebook-8f919155-6035-47b4-9c6f-1be133b9e2c9-0.notebook-8f919155-6035-47b4-9c6f-1be133b9e2c9.colossal-ai.svc.cluster.local>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fix sync condition (#6000)
* [plugin] add cast inputs option for zero (#6003)
* [pre-commit.ci] pre-commit autoupdate (#5995)
updates:
- [github.com/psf/black-pre-commit-mirror: 24.4.2 → 24.8.0](https://github.com/psf/black-pre-commit-mirror/compare/24.4.2...24.8.0)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [misc] Bypass the huggingface bug to solve the mask mismatch problem (#5991)
* [Feature] Zigzag Ring attention (#5905)
* halfway
* fix cross-PP-stage position id length diff bug
* fix typo
* fix typo
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* unified cross entropy func for all shardformer models
* remove redundant lines
* add basic ring attn; debug cross entropy
* fwd bwd logic complete
* fwd bwd logic complete; add experimental triton rescale
* precision tests passed
* precision tests passed
* fix typos and remove misc files
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* add sp_mode to benchmark; fix varlen interface
* update softmax_lse shape by new interface
* change tester name
* remove buffer clone; support packed seq layout
* add varlen tests
* fix typo
* all tests passed
* add dkv_group; fix mask
* remove debug statements
---------
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [misc] update compatibility (#6008)
* [misc] update compatibility
* [misc] update requirements
* [devops] disable requirements cache
* [test] fix torch ddp test
* [test] fix rerun on address in use
* [test] fix lazy init
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix the merge
* fix the merge
* overlap kv comm with output rescale (#6017)
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* fix the merge
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix the merge
* fix
* fix
* fix the merge
* fix
* [misc] Use dist logger in plugins (#6011)
* use dist logger in plugins
* remove trash
* print on rank 0
---------
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
* fix
* fix
* fix
* fix
* fix the merge
* fix
* fix
* fix
* fix
---------
Co-authored-by: YeAnbang <anbangy2@outlook.com>
Co-authored-by: Haze188 <haze188@qq.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Edenzzzz <wenxuan.tan@wisc.edu>
Co-authored-by: Edenzzzz <wtan45@wisc.edu>
Co-authored-by: Runyu Lu <77330637+LRY89757@users.noreply.github.com>
Co-authored-by: Guangyao Zhang <xjtu521@qq.com>
Co-authored-by: YeAnbang <44796419+YeAnbang@users.noreply.github.com>
Co-authored-by: Hongxin Liu <lhx0217@gmail.com>
Co-authored-by: Stephan Kö <stephankoe@users.noreply.github.com>
Co-authored-by: アマデウス <kurisusnowdeng@users.noreply.github.com>
Co-authored-by: Tong Li <tong.li352711588@gmail.com>
Co-authored-by: zhurunhua <1281592874@qq.com>
Co-authored-by: Insu Jang <insujang@umich.edu>
Co-authored-by: Gao, Ruiyuan <905370712@qq.com>
Co-authored-by: hxwang <wang1570@e.ntu.edu.sg>
Co-authored-by: Michelle <qianranma8@gmail.com>
Co-authored-by: root <root@notebook-8f919155-6035-47b4-9c6f-1be133b9e2c9-0.notebook-8f919155-6035-47b4-9c6f-1be133b9e2c9.colossal-ai.svc.cluster.local>
8.7 KiB
启动 Colossal-AI
作者: Chuanrui Wang, Shenggui Li, Siqi Mai
预备知识:
简介
正如我们在前面的教程中所提到的,在您的配置文件准备好后,您需要为 Colossal-AI 初始化分布式环境。我们把这个过程称为 launch
。在本教程中,您将学习如何在您的服务器上启动 Colossal-AI,不管是小型的还是大型的。
在 Colossal-AI 中,我们提供了几种启动方法来初始化分布式后端。
在大多数情况下,您可以使用 colossalai.launch
和 colossalai.get_default_parser
来通过命令行传递参数。如果您想使用 SLURM、OpenMPI 和 PyTorch 等启动工具,我们也提供了几个启动的辅助方法以便您的使用。您可以直接从这些启动工具设置的环境变量中访问 rank 和 world size 大小。
在本教程中,我们将介绍如何启动 Colossal-AI 来初始化分布式后端:
- 用 colossalai.launch 启动
- 用 Colossal-AI命令行 启动
- 用 SLURM 启动
- 用 OpenMPI 启动
启动分布式环境
为了启动 Colossal-AI,我们需要两类参数:
- 配置文件
- 分布式设置
无论我们使用何种启动方式,配置文件是必须要求的,而分布式设置有可能依情况而定。配置文件可以是配置文件的路径或 Python dictionary 的形式。分布式设置可以通过命令行或多进程启动器传递。
命令行解析器
在使用 launch
之前, 我们首先需要了解我们需要哪些参数来进行初始化。
如分布式训练 中 基本概念
一节所述 ,涉及的重要参数是:
- host
- port
- rank
- world_size
- backend
在 Colossal-AI 中,我们提供了一个命令行解析器,它已经提前添加了这些参数。您可以通过调用 colossalai.get_default_parser()
来获得这个解析器。这个解析器通常与 colossalai.launch
一起使用。
# add these lines in your train.py
import colossalai
# get default parser
parser = colossalai.get_default_parser()
# if you want to add your own arguments
parser.add_argument(...)
# parse arguments
args = parser.parse_args()
您可以在您的终端传入以下这些参数。
python train.py --host <host> --rank <rank> --world_size <world_size> --port <port> --backend <backend>
backend
是用户可选的,默认值是 nccl。
本地启动
为了初始化分布式环境,我们提供了一个通用的 colossalai.launch
API。colossalai.launch
函数接收上面列出的参数,并在通信网络中创建一个默认的进程组。方便起见,这个函数通常与默认解析器一起使用。
import colossalai
# parse arguments
args = colossalai.get_default_parser().parse_args()
# launch distributed environment
colossalai.launch(rank=args.rank,
world_size=args.world_size,
host=args.host,
port=args.port,
backend=args.backend
)
用 Colossal-AI命令行工具 启动
为了更好地支持单节点以及多节点的训练,我们通过封装PyTorch的启动器实现了一个更加方便的启动器。 PyTorch自带的启动器需要在每个节点上都启动命令才能启动多节点训练,而我们的启动器只需要一次调用即可启动训练。
首先,我们需要在代码里指定我们的启动方式。由于这个启动器是PyTorch启动器的封装,那么我们自然而然应该使用colossalai.launch_from_torch
。
分布式环境所需的参数,如 rank, world size, host 和 port 都是由 PyTorch 启动器设置的,可以直接从环境变量中读取。
train.py
import colossalai
colossalai.launch_from_torch()
...
接下来,我们可以轻松地在终端使用colossalai run
来启动训练。下面的命令可以在当前机器上启动一个4卡的训练任务。
你可以通过设置nproc_per_node
来调整使用的GPU的数量,也可以改变master_port
的参数来选择通信的端口。
# 在当前节点上启动4卡训练 (默认使用29500端口)
colossalai run --nproc_per_node 4 train.py
# 在当前节点上启动4卡训练,并使用一个不同的端口
colossalai run --nproc_per_node 4 --master_port 29505 test.py
如果你在使用一个集群,并且想进行多节点的训练,你需要使用Colossal-AI的命令行工具进行一键启动。我们提供了两种方式来启动多节点任务
- 通过
--hosts
来启动
这个方式适合节点数不多的情况。假设我们有两个节点,分别为host
和host2
。我们可以用以下命令进行多节点训练。
比起单节点训练,多节点训练需要手动设置--master_addr
(在单节点训练中master_addr
默认为127.0.0.1
)。同时,你需要确保每个节点都使用同一个ssh port。可以通过--ssh-port设置。
:::caution
多节点训练时,master_addr
不能为localhost
或者127.0.0.1
,它应该是一个节点的名字或者IP地址。
:::
# 在两个节点上训练
colossalai run --nproc_per_node 4 --host host1,host2 --master_addr host1 test.py --ssh-port 22
- 通过
--hostfile
来启动
这个方式适用于节点数很大的情况。host file是一个简单的文本文件,这个文件里列出了可以使用的节点的名字。
在一个集群中,可用节点的列表一般由SLURM或者PBS Pro这样的集群资源管理器来提供。比如,在SLURM中,
你可以从SLURM_NODELIST
这个环境变量中获取到当前分配列表。在PBS Pro中,这个环境变量为PBS_NODEFILE
。
可以通过echo $SLURM_NODELIST
或者 cat $PBS_NODEFILE
来尝试一下。如果你没有这样的集群管理器,
那么你可以自己手动写一个这样的文本文件即可。
提供给Colossal-AI的host file需要遵循以下格式,每一行都是一个节点的名字。
host1
host2
如果host file准备好了,那么我们就可以用以下命令开始多节点训练了。和使用--host
一样,你也需要指定一个master_addr
。
当使用host file时,我们可以使用一些额外的参数:
-
--include
: 设置你想要启动训练的节点。比如,你的host file里有8个节点,但是你只想用其中的6个节点进行训练, 你可以添加--include host1,host2,host3,...,host6
,这样训练任务只会在这6个节点上启动。 -
--exclude
: 设置你想排除在训练之外的节点。当你的某一些节点坏掉时,这个参数会比较有用。比如假如host1的GPU有一些问题,无法正常使用, 那么你就可以使用--exclude host1
来将其排除在外,这样你就可以训练任务就只会在剩余的节点上启动。
# 使用hostfile启动
colossalai run --nproc_per_node 4 --hostfile ./hostfile --master_addr host1 test.py
# 只使用部分节点进行训练
colossalai run --nproc_per_node 4 --hostfile ./hostfile --master_addr host1 --include host1 test.py
# 不使用某些节点进行训练
colossalai run --nproc_per_node 4 --hostfile ./hostfile --master_addr host1 --exclude host2 test.py
用 SLURM 启动
如果您是在一个由 SLURM 调度器管理的系统上, 您也可以使用 srun
启动器来启动您的 Colossal-AI 脚本。我们提供了辅助函数 launch_from_slurm
来与 SLURM 调度器兼容。
launch_from_slurm
会自动从环境变量 SLURM_PROCID
和 SLURM_NPROCS
中分别读取 rank 和 world size ,并使用它们来启动分布式后端。
您可以在您的训练脚本中尝试以下操作。
import colossalai
colossalai.launch_from_slurm(
host=args.host,
port=args.port
)
您可以通过在终端使用这个命令来初始化分布式环境。
srun python train.py --host <master_node> --port 29500
用 OpenMPI 启动
如果您对OpenMPI比较熟悉,您也可以使用 launch_from_openmpi
。
launch_from_openmpi
会自动从环境变量
OMPI_COMM_WORLD_LOCAL_RANK
, MPI_COMM_WORLD_RANK
和 OMPI_COMM_WORLD_SIZE
中分别读取local rank、global rank 和 world size,并利用它们来启动分布式后端。
您可以在您的训练脚本中尝试以下操作。
colossalai.launch_from_openmpi(
host=args.host,
port=args.port
)
以下是用 OpenMPI 启动多个进程的示例命令。
mpirun --hostfile <my_hostfile> -np <num_process> python train.py --host <node name or ip> --port 29500
- --hostfile: 指定一个要运行的主机列表。
- --np: 设置总共要启动的进程(GPU)的数量。例如,如果 --np 4,4个 python 进程将被初始化以运行 train.py。