关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>
PCL-Platform.Intelligence
/
mPanGu-Alpha-53
Not watched
20
13
Fork 20
Code
Releases 0 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
Issues
/
Detail
Labels Milestones
New Issue

#11 GPU方式训练保存的8份checkpoint如何合并然后进行模型推理

Closed
created 1 year ago by abulice · 1 comments
abulice commented 1 year ago

按照GPU训练的方法用8卡机器finetune了一下2.6B的模型，发现保持的ckeckpoint文件是分成8块的，推理的时候用什么脚本可以把8块文件合并成1个呢？

按照GPU训练的方法用8卡机器finetune了一下2.6B的模型，发现保持的ckeckpoint文件是分成8块的，推理的时候用什么脚本可以把8块文件合并成1个呢？
abulice changed title from GPU方式训练保持的8份checkpoint如何合并然后进行模型推理 to GPU方式训练保存的8份checkpoint如何合并然后进行模型推理 1 year ago
taoht commented 1 year ago
Owner

1、分布式加载8片模型参数

    pangu_alpha = PanguAlphaModel(config)
    eval_net = EvalNet(pangu_alpha, pad_token=args_opt.padding_id)
    eval_net.set_train(False)
    model_predict = Model(eval_net)
    
    load_distributed_checkpoint(eval_net, ckpt_file_list, predict_layout)

2、整合保存成单个模型文件

    save_checkpoint(pangu_alpha, "/path/to/mPanGu_integrated.ckpt", integrated_save=True)
1、分布式加载8片模型参数 ``` pangu_alpha = PanguAlphaModel(config) eval_net = EvalNet(pangu_alpha, pad_token=args_opt.padding_id) eval_net.set_train(False) model_predict = Model(eval_net) load_distributed_checkpoint(eval_net, ckpt_file_list, predict_layout) ``` 2、整合保存成单个模型文件 ``` save_checkpoint(pangu_alpha, "/path/to/mPanGu_integrated.ckpt", integrated_save=True) ```
abulice closed this issue 1 year ago
Sign in to join this conversation.
No Branch/Tag Specified
Clear branch or tag
Branches Tags
Labels
Apply labels
Clear labels
No items
No Label
Milestone
Set milestone
Clear milestone
No items
No Milestone
Assignees
Assign users
Clear assignees
No Assignees
2 Participants
Notifications
Due Date

No due date set.

Dependencies

This issue currently doesn't have any dependencies.

Write Preview
Loading…
Cancel
Save
There is no content yet.
Delete Branch '%!s(MISSING)'

Deleting a branch is permanent. It CANNOT be undone. Continue?

No
Yes
Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit