【求助】算力云上运行Image Captioning 任务与模型执⾏VQA任务报错 #280
Labels
No Label
bug
duplicate
enhancement
help wanted
invalid
question
wontfix
No Milestone
No project
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: HswOAuth/llm_course#280
Loading…
Reference in New Issue
Block a user
No description provided.
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
报错内容如下,是通过老师给的镜像建的实例
这个错误通常发生在网络连接不稳定、服务器不可达或请求超时的情况下。解决这个问题可能需要检查网络连接、服务器状态或调整请求的超时设置。同学可以试试在jupyter最前面开一个cell,然后通过如下命令设置一下网站镜像:
!export HF_HOME=/root/autodl-tmp/huggingface-cache/
!export HF_ENDPOINT=https://hf-mirror.com
可以在命令行中直接用(不需要感叹号),或者是在jupyter单开一个cell在前面加!用。
如果有后续的话在后面回我一下喔,我会跟进的~
换了一个报错好像是什么文件错误?
感觉是没有下载好bert-based-uncased模型啊,你试试把下面的代码再跑一遍,是不是之前忽略了报错?

如果没问题的话,麻烦你截图一下左侧的文件,看看autodl-tmp里面有没有bert-based-uncased这个模型?
应该是之前模型没有下载完整,我重新走了下下载代码运行显示这个
使用pip install下载这个没有的包试试
不行呀显示
ERROR: Could not find a version that satisfies the requirement torch.distributed.fsdp (from versions: none)
ERROR: No matching distribution found for torch.distributed.fsdp
试试先更新一下pip包:
pip install --upgrade pip
然后再用镜像源下载试试,看能安装不:
pip install torch.distributed.fsdp -i https://pypi.tuna.tsinghua.edu.cn/simple
显示这个
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
ERROR: Could not find a version that satisfies the requirement torch.distributed.fsdp (from versions: none)
ERROR: No matching distribution found for torch.distributed.fsdp
(base) root@autodl-container-c33846b235-d9deba44:~#
提供的镜像里面,blip使用的环境是名为blip的conda环境,所以:
用base环境运行会显示缺少torch.distributed.fsdp模块,用blip环境运行会显示The size of tensor a (3) must match the size of tensor b (9) at non-singleton dimension 0
出现这个报错的话就不用beam search,把那一行注释掉,用下面的那个方法就可以了,这是版本问题