YUANMU227

FileExistsError: [Errno 17] File exists: './work_dirs/.eval_hook'

1

mmcls正常train过程中，进行多个机器的分布式训练，运行多个epoch之后，会报错： FileExistsError: [Errno 17] File exists: './work_dirs/.eval_hook' 这个.eval_hook应该是中间文件，训练结束后，会自动生成 由于我是多个机器训练，所以怀疑是每个机器训练之后都想生成这个文件，所以产生冲突，导致训练失败 但是之前就没有遇到过这个情况，同样的代码换了数据集和batchsize就会报错这个，请问这是开发的bug，还是我个人设置问题

Can sam-hq perform breakpoint training: continue training on the already trained sam-hq, or continue training on the official sam-hq weights

2

Can sam-hq perform breakpoint training: continue training on the already trained sam-hq, or continue training on the official sam-hq weights？ Modifying the --start_epoch parameter will not allow continued training on...

## problem During the training of sam-hq, the iou output of the val set is very high, 0.98; but in eval mode, the iou of the val set is only...

Tiled Diffusion+Tiled VAE+ControlNet v1.1： ValueError: too many values to unpack (expected 3)

4

hello，I used Tiled Diffusion+Tiled VAE+ControlNet v1.1. But I had an error: ValueError: too many values to unpack (expected 3) what might be the reason? Is the version of my model...

Tiled Diffusion + Noise Inversion + ControlNet Tile输出包含两张图

1

hello,在webui上，Tiled Diffusion + Noise Inversion + ControlNet Tile输出包含两张图，这两张是什么区别呀

https://github.com/mindspore-lab/mindcv/tree/main/configs/convit 使用：python train.py --config configs/convit/convit_tiny_ascend.yaml --data_dir /home/ma-user/work/data --distribute False 报错： RuntimeError: Run task for graph:kernel_graph_37 error! The details refer to 'Ascend Error Message'. 环境： mindspore2.0.0 log： [WARNING] MD(17532,fffd117fa1e0,python):2023-03-09-20:18:49.009.973 [mindspore/ccsrc/minddata/dataset/engine/datasetops/data_queue_op.cc:832] DetectPerBatchTime]...

If this is your first time, please read our contributor guidelines: https://github.com/mindspore-lab/mindcv/blob/main/CONTRIBUTING.md **Describe the bug/ 问题描述 (Mandatory / 必填)** 报错：RuntimeError: Role name '' is invalid. Maybe you are trying to...

modelarts无法使用ascend进行训练，npu使用率为0

1

启动指令为python train.py --config configs/vit/vit_b32_224_ascend.yaml --data_dir /home/ma-user/work/datasets --distribute False --device_target="Ascend" 但是notebook显示未使用npu： 这个是为什么呀，是config需要别的设置，还是说代码里没有使用ascend训练

Thank you for your work, it's very good work. I need to fine-tune the detection model on a custom dataset and have a few questions to confirm: 1. The training...

Great work. When will box-based segmentation inference be released?

8

Box-based guidance is of great significance for saliency segmentation of specific parts. I saw this item in your TODO list. When will this feature be launched?