Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / yanqiangmiffy/instructglm issues and pull requests

#34 - Problems in train_deepspeed.py with ZeRO stage 1|2|3

Issue - State: open - Opened by zjhJOJO about 1 year ago

#34 - Problems in train_deepspeed.py with ZeRO stage 1|2|3

Issue - State: open - Opened by zjhJOJO about 1 year ago

#30 - 调教后的逻辑能力如何？

Issue - State: open - Opened by daiaji over 1 year ago - 1 comment

#30 - 调教后的逻辑能力如何？

Issue - State: open - Opened by daiaji over 1 year ago - 1 comment

#27 - 请问下train_deepspeed.py 怎么引入lora.pt

Issue - State: open - Opened by AlexXx-Wu over 1 year ago - 1 comment

#27 - 请问下train_deepspeed.py 怎么引入lora.pt

Issue - State: open - Opened by AlexXx-Wu over 1 year ago - 1 comment

#26 - torch.distributed.elastic.multiprocessing.errors.ChildFailedError

Issue - State: closed - Opened by MonkeyTB over 1 year ago - 1 comment

#26 - torch.distributed.elastic.multiprocessing.errors.ChildFailedError

Issue - State: closed - Opened by MonkeyTB over 1 year ago - 1 comment

#22 - RuntimeError: torch.cat(): expected a non-empty list of Tensors

Issue - State: closed - Opened by hrdxwandg over 1 year ago - 7 comments

#22 - RuntimeError: torch.cat(): expected a non-empty list of Tensors

Issue - State: closed - Opened by hrdxwandg over 1 year ago - 7 comments

#20 - 4张 12G的 3060能训练吗

Issue - State: open - Opened by zlszhonglongshen over 1 year ago - 2 comments

#20 - 4张 12G的 3060能训练吗

Issue - State: open - Opened by zlszhonglongshen over 1 year ago - 2 comments

#19 - 运行web_demo_alpaca_lora.py报错，是单纯的显存不够嘛

Issue - State: open - Opened by tianmala over 1 year ago - 1 comment

#19 - 运行web_demo_alpaca_lora.py报错，是单纯的显存不够嘛

Issue - State: open - Opened by tianmala over 1 year ago - 1 comment

#16 - 微调2:BELLE中文指令数据的问题

Issue - State: open - Opened by czhcc over 1 year ago - 1 comment

#16 - 微调2:BELLE中文指令数据的问题

Issue - State: open - Opened by czhcc over 1 year ago - 1 comment

#15 - Lora+DeepSpeed多机多卡的问题

Issue - State: closed - Opened by zyds over 1 year ago

#15 - Lora+DeepSpeed多机多卡的问题

Issue - State: closed - Opened by zyds over 1 year ago

#14 - 怎么能把lora参数merge回原始模型呢？

Issue - State: closed - Opened by AItechnology over 1 year ago - 1 comment

#14 - 怎么能把lora参数merge回原始模型呢？

Issue - State: closed - Opened by AItechnology over 1 year ago - 1 comment

#12 - 最新update的代码中，web_demo推理时报错

Issue - State: open - Opened by feyxong over 1 year ago - 2 comments

#12 - 最新update的代码中，web_demo推理时报错

Issue - State: open - Opened by feyxong over 1 year ago - 2 comments

#10 - 修改README.md中的错别字

Pull Request - State: closed - Opened by SunYanCN over 1 year ago

#10 - 修改README.md中的错别字

Pull Request - State: closed - Opened by SunYanCN over 1 year ago

#9 - 关于多轮对话的疑问

Issue - State: open - Opened by ZeyuTeng96 over 1 year ago

#8 - web_demo_belle生成结果时有大段重复的问题

Issue - State: open - Opened by JiayiFu over 1 year ago - 10 comments

#7 - 请问支持多卡吗，怎么改造？

Issue - State: closed - Opened by hjyMM2018 over 1 year ago - 2 comments

#6 - RuntimeError: expected scalar type Half but found Float

Issue - State: closed - Opened by fulQuan over 1 year ago - 1 comment

#5 - ValueError: 150000 is not in list

Issue - State: open - Opened by superhg over 1 year ago - 5 comments

#4 - ValueError: Please specify `target_modules` in `peft_config`

Issue - State: closed - Opened by MrInouye over 1 year ago - 3 comments

#3 - 24G显存的3090可以训练吗？

Issue - State: closed - Opened by franklyd over 1 year ago - 1 comment

#2 - 请问有训练好的权重可以下载吗？

Issue - State: closed - Opened by EagleChen over 1 year ago - 1 comment