name | about | labels |
---|---|---|
Bug Report | Use this template for reporting a bug | kind/bug |
qwen/baichuan2/internlm/skywork/iflytekspark/wizardcoder等网络在910B环境上训练失败
模型仓地址:https://gitee.com/mindspore/mindformers/tree/r1.0/research/baichuan2
Ascend
/GPU
/CPU
) / 硬件环境:Please delete the backend not involved / 请删除不涉及的后端:
/device ascend/
失败版本:
CANN:Milan_C15/20240302
MS:r2.2_20240411041520_c4eddc9d731
MF:r1.0_20240411161523_64bb772e6b78
PyNative
/Graph
):Please delete the mode not involved / 请删除不涉及的模式:
/mode pynative
/mode graph
用例仓地址:MindFormers_Test/cases/baichuan2/7b/train/
用例
test_mf_baichuan2_7b_train_belle_8p_0001
网络训练成功
INFO 2024-04-11 23:16:30 - test_mf_baichuan2_7b_train_belle_8p_0001 - train.py:print_err_info_in_log:321 -
[ERROR] 2024-04-11 23:15:51,763 [mindformers/tools/cloud_adapter/cloud_monitor.py:43] wrapper: Traceback (most recent call last):
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/tools/cloud_adapter/cloud_monitor.py", line 34, in wrapper
result = run_func(*args, **kwargs)
File "./baichuan2/run_baichuan2.py", line 169, in main
trainer.finetune(finetune_checkpoint=ckpt, auto_trans_ckpt=config.auto_trans_ckpt, resume_training=resume)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindspore/_checkparam.py", line 1313, in wrapper
return func(*args, **kwargs)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/trainer/trainer.py", line 491, in finetune
is_full_config=True)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/trainer/causal_language_modeling/causal_language_modeling.py", line 104, in train
**kwargs)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/trainer/base_trainer.py", line 739, in training_process
transform_and_load_checkpoint(config, model, network, dataset)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/trainer/utils.py", line 348, in transform_and_load_checkpoint
dst_ckpt_strategy=dst_ckpt_strategy)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/trainer/utils.py", line 604, in transform_ckpt
wait_transform(config, ckpt_dir, world_size)
File "/home/miniconda3/envs/ci/lib/python3.7/site-packages/mindformers/trainer/utils.py", line 654, in wait_transform
raise ValueError(f"Transform failed, find {transform_failed_txts}.")
ValueError: Transform failed, find ['./output/transformed_checkpoint/train/transform_failed_rank_0.txt'].
INFO 2024-04-11 23:16:30 - test_mf_baichuan2_7b_train_belle_8p_0001 - train.py:print_err_info_in_log:323 -
走给寇凯睿
Please assign maintainer to check this issue.
请为此issue分配处理人。
@zhangjie18
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。
感谢您的提问,您可以评论//mindspore-assistant更快获取帮助:
修复问题单引入,取元素前未对src_strategy_file和dst_strategy_file判空,2024/04/11修复PR已合入。
添加判空。
!68166:【r2.2】修复ckpt转换src或dst策略文件为空时bug
ckpe转换不在ut/st中添加用例。
ms 2024/04/12 daily验证已通过。
ms 2024/04/12 daily及之后版本。
回归版本:CANN:Milan_C15/20240302
MS:r2.2_20240412041523_5fe9f78bde8ac3b7
MF:r1.0_20240411161523_64bb772e6b78
回归步骤:参考issue步骤
基本问题:已解决
测试结论:回归通过
回归时间:2024.4.12
登录 后才可以发表评论