登录
注册
开源
企业版
高校版
搜索
帮助中心
使用条款
关于我们
开源
企业版
高校版
私有云
模力方舟
AI 队友
登录
注册
轻量养虾,开箱即用!低 Token + 稳定算力,Gitee & 模力方舟联合出品的 PocketClaw 正式开售!点击了解详情~
代码拉取完成,页面将自动刷新
仓库状态说明
开源项目
>
人工智能
&&
捐赠
捐赠前请先登录
取消
前往登录
扫描微信二维码支付
取消
支付完成
支付提示
将跳转至支付宝完成支付
确定
取消
Watch
不关注
关注所有动态
仅关注版本发行动态
关注但不提醒动态
205
Star
1.3K
Fork
1.2K
Ascend
/
MindSpeed-LLM
暂停
代码
Issues
3
Pull Requests
32
Wiki
统计
流水线
服务
质量分析
Jenkins for Gitee
腾讯云托管
腾讯云 Serverless
悬镜安全
阿里云 SAE
Codeblitz
SBOM
开发画像分析
我知道了,不再自动展开
更新失败,请稍后重试!
移除标识
内容风险标识
本任务被
标识为内容中包含有代码安全 Bug 、隐私泄露等敏感信息,仓库外成员不可访问
ds2_lite optim 转换报错
DONE
#ICBXV0
缺陷
ruiq
创建于
2025-06-02 16:28
配置为 ``` python /opt/dpcvol/models/work/sparCheck/MindSpeed-LLM/convert_ckpt.py \ --use-mcore-models \ --model-type-hf deepseek2-lite \ --model-type GPT \ --load-model-type optim \ --params-dtype bf16 \ --target-tensor-parallel-size 1 \ --target-pipeline-parallel-size 1 \ --target-expert-parallel-size 8 \ --moe-grouped-gemm \ --spec mindspeed_llm.tasks.models.spec.deepseek_spec layer_spec \ --load-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/ \ --save-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/ ``` INFO:root: exp_avg_sq is saved to /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/iter_0025880/mp_rank_00_007/distrib_optim_exp_avg_sq.pt. INFO:root:Splitting from /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/ done. INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_000/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_001/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_002/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_003/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_004/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_005/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_006/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_007/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:get source weight for target pp_rank: 0 INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 146.83it/s] my_ckpt_convert_deepseek2_lite_optim.sh: line 16: 279283 Killed python /opt/dpcvol/models/work/sparCheck/MindSpeed-LLM/convert_ckpt.py --use-mcore-models --model-type-hf deepseek2-lite --model-type GPT --load-model-type optim --params-dtype bf16 --target-tensor-parallel-size 1 --target-pipeline-parallel-size 1 --target-expert-parallel-size 8 --moe-grouped-gemm --spec mindspeed_llm.tasks.models.spec.deepseek_spec layer_spec --load-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/ --save-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/ 成功split 但是卡在创建文件,缺少model_optim_rng.pt,但是显然model_optim_rng.pt 还没创建
配置为 ``` python /opt/dpcvol/models/work/sparCheck/MindSpeed-LLM/convert_ckpt.py \ --use-mcore-models \ --model-type-hf deepseek2-lite \ --model-type GPT \ --load-model-type optim \ --params-dtype bf16 \ --target-tensor-parallel-size 1 \ --target-pipeline-parallel-size 1 \ --target-expert-parallel-size 8 \ --moe-grouped-gemm \ --spec mindspeed_llm.tasks.models.spec.deepseek_spec layer_spec \ --load-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/ \ --save-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/ ``` INFO:root: exp_avg_sq is saved to /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/iter_0025880/mp_rank_00_007/distrib_optim_exp_avg_sq.pt. INFO:root:Splitting from /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/ done. INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_000/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_001/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_002/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_003/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_004/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_005/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_006/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:Error: File not found at '/opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/iter_0025880/mp_rank_00_007/model_optim_rng.pt'. Convert the model weight first.Skipping... INFO:root:get source weight for target pp_rank: 0 INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. INFO:root:Data loaded successfully. 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 146.83it/s] my_ckpt_convert_deepseek2_lite_optim.sh: line 16: 279283 Killed python /opt/dpcvol/models/work/sparCheck/MindSpeed-LLM/convert_ckpt.py --use-mcore-models --model-type-hf deepseek2-lite --model-type GPT --load-model-type optim --params-dtype bf16 --target-tensor-parallel-size 1 --target-pipeline-parallel-size 1 --target-expert-parallel-size 8 --moe-grouped-gemm --spec mindspeed_llm.tasks.models.spec.deepseek_spec layer_spec --load-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/mg_merge/ --save-dir /opt/dpcvol/models/checkpoints_deepseek_v2_lite/optim_mg_output/ 成功split 但是卡在创建文件,缺少model_optim_rng.pt,但是显然model_optim_rng.pt 还没创建
评论 (
4
)
登录
后才可以发表评论
状态
DONE
TODO
WIP
DONE
CLOSED
REJECTED
负责人
未设置
标签
未设置
项目
未立项任务
未立项任务
里程碑
未关联里程碑
未关联里程碑
Pull Requests
未关联
未关联
关联的 Pull Requests 被合并后可能会关闭此 issue
分支
未关联
分支 (
-
)
标签 (
-
)
开始日期   -   截止日期
-
置顶选项
不置顶
置顶等级:高
置顶等级:中
置顶等级:低
优先级
不指定
严重
主要
次要
不重要
预计工期
(小时)
参与者(2)
Python
1
https://gitee.com/ascend/MindSpeed-LLM.git
git@gitee.com:ascend/MindSpeed-LLM.git
ascend
MindSpeed-LLM
MindSpeed-LLM
点此查找更多帮助
搜索帮助
Git 命令在线学习
如何在 Gitee 导入 GitHub 仓库
Git 仓库基础操作
企业版和社区版功能对比
SSH 公钥设置
如何处理代码冲突
仓库体积过大,如何减小?
如何找回被删除的仓库数据
Gitee 产品配额说明
GitHub仓库快速导入Gitee及同步更新
什么是 Release(发行版)
将 PHP 项目自动发布到 packagist.org
评论
仓库举报
回到顶部
登录提示
该操作需登录 Gitee 帐号,请先登录后再操作。
立即登录
没有帐号,去注册