wenmeng zhou comments

Results 74 comments of


                                            wenmeng zhou

NER 任务的批量推理没起作用

预测结果迟迟没有，是指运行程序一直没有输出？怀疑是在加载输入列表上，可以用pystack-debugger查看下程序运行在什么代码上

RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

use v1.4.2 which has fix this

求助大佬，win系统，SD扩展训练报错，

1. ERROR facechain: failed to install mmcv, make sure to have "CUDA Toolkit" and "Build Tools for Visual Studio" installed 请安装对应工具编译mmcv 2. 代码中accelerate launch改成python

部署到sd webui上，用ui界面进行多次训练和生成，内存一直不释放，会把内存慢慢打爆，我的机器128G内存都不够，还是会爆，是不是哪块没释放内存啊

是在linux上还是windows上？

部署到sd webui上，用ui界面进行多次训练和生成，内存一直不释放，会把内存慢慢打爆，我的机器128G内存都不够，还是会爆，是不是哪块没释放内存啊

请问大概运行每次会上涨多少内存，一般多少次后会oom，我们复现下

您好，能发布一个windows版的部署说明吗？小白搞了三天都没成功报错太多了

pip install mim mom install mmcv-full

Will ConvMAE's imagenet pre training parameters be provided

https://github.com/alibaba/EasyCV/blob/master/docs/source/model_zoo_ssl.md#fast-convmae use fastconv mae

Names of downloaded files are not match

using git clone to download the raw files

Authentication token does not exist, failed to access model unknown which may not exist or may be private. Please login first.

在训练前执行如下代码登录，使用自己的私有仓库 ```python from modelscope.hub.api import HubApi YOUR_ACCESS_TOKEN = '请从ModelScope个人中心->访问令牌获取' api = HubApi() api.login(YOUR_ACCESS_TOKEN) ```

希望能够给modelscope上的仓库放开修改 commit 信息的权限

Sorry for the late reply, we will consider this feature in recent sprint