mmcv版本问题

[zzhhzz666]

作者您好，又来打搅您了。上次跟您说，每次跑几个epoch会爆内存，然后您说，可以使用延续点进行训练，第一次跑了10个epoch， 我延续这个，又跑到了第22个epoch，又断了，再然后就运行不了了。在第23个epoch开始时，就会直接爆，尝试了好几次，也只能处理第一组，然后直接 out of memory. 然后我按照您说的mmcv-full版本泄露问题，按照您给的指令，pytorch1.9.0对应的mmcv-full应该就是1.7.2.我安装的就是1.7.2.然后我从官网上下载1.7.1,但是就不是mmcv-full了，只能是mmcv1.7.1了，然后运行后，一直报ModuleNotFoundError: No module named ‘mmcv._ext..所以我想问下您用的是哪个版本的mmcv——full,或者说有没有其他解决方法。祝万事如意！

[zzhhzz666]

我将batch_size从2设置成了1，可以跑完24个epoch了，但是train的结果不太理想，map最好为0.386。所以想按照您给的设置，跑完24个epoch，但是会出现我前面提到的memory一直上升，我第一次跑10个epoch就断了，按照您说的延续checkpoint继续训练，第二次到22个epoch，然后23就彻底跑不了了，我看前面您23年回答过别人您用的mmcv版本是1.4.5，但是本文给的readme用的不是mmcv-full嘛，本来想再创建个环境，由于用的这个服务器磁盘空间满了，所以创建不了，为了保险才来问您。如有打扰，请见谅！望事事顺利！

[zzhhzz666]

train的时候用mmcv-full==1.7.2，虽然会有内存泄漏，但是降低batch能运行出来，到了inference的时候，会告诉我安装mmcv>=1.3.2, <=1.4.0，就很奇怪。我又卸载了1.7.2安装mmcv-full1.4.0.也会有别的问题 (open-mmlab) zhuzhenhua@ubuntu-SYS-4028GR-TR:~/pointobb-main/PointOBB$ sh test_p.sh

CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'. If your shell is Bash or a Bourne variant, enable conda for the current user with

$ echo ". /home/zhaoxiaoyao/anaconda3/etc/profile.d/conda.sh" >> ~/.bashrc

or, for all users, enable conda with

$ sudo ln -s /home/zhaoxiaoyao/anaconda3/etc/profile.d/conda.sh /etc/profile.d/conda.sh

The options above will permanently enable the 'conda' command, but they do NOT put conda's base (root) environment on PATH. To do so, run

$ conda activate

in your terminal, or to put the base environment on PATH permanently, run

$ echo "conda activate" >> ~/.bashrc

Previous to conda 4.4, the recommended way to activate conda was to modify PATH in your ~/.bashrc file. You should manually remove the line that looks like

export PATH="/home/zhaoxiaoyao/anaconda3/bin:$PATH"

^^^ The above line should NO LONGER be in your ~/.bashrc file! ^^^

/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/init.py:21: UserWarning: On January 1, 2023, MMCV will release v2.0.0, in which it will remove components related to the training process and add a data transformation module. In addition, it will rename the package names mmcv to mmcv-lite and mmcv-full to mmcv. See https://github.com/open-mmlab/mmcv/blob/master/docs/en/compatibility.md for more details. 'On January 1, 2023, MMCV will release v2.0.0, in which it will remove ' Traceback (most recent call last): File "tools/train.py", line 24, in from mmdet import version File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmdet/init.py", line 25, in f'MMCV=={mmcv.version} is used but incompatible. ' \ AssertionError: MMCV==1.7.2 is used but incompatible. Please install mmcv>=1.3.2, <=1.4.0. /home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/init.py:21: UserWarning: On January 1, 2023, MMCV will release v2.0.0, in which it will remove components related to the training process and add a data transformation module. In addition, it will rename the package names mmcv to mmcv-lite and mmcv-full to mmcv. See https://github.com/open-mmlab/mmcv/blob/master/docs/en/compatibility.md for more details. 'On January 1, 2023, MMCV will release v2.0.0, in which it will remove ' Traceback (most recent call last): File "exp/tools/result2ann_obb.py", line 4, in from mmdet.core.bbox import bbox_overlaps File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmdet/init.py", line 25, in f'MMCV=={mmcv.version} is used but incompatible. ' \ AssertionError: MMCV==1.7.2 is used but incompatible. Please install mmcv>=1.3.2, <=1.4.0.

[zzhhzz666]

Traceback (most recent call last): File "tools/train.py", line 273, in main() File "tools/train.py", line 234, in main test_cfg=cfg.get('test_cfg')) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmdet/models/builder.py", line 58, in build_detector cfg, default_args=dict(train_cfg=train_cfg, test_cfg=test_cfg)) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/utils/registry.py", line 212, in build return self.build_func(*args, **kwargs, registry=self) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/cnn/builder.py", line 27, in build_model_from_cfg return build_from_cfg(cfg, registry, default_args) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/utils/registry.py", line 45, in build_from_cfg f'{obj_type} is not in the {registry.name} registry') KeyError: 'PointOBB is not in the models registry' loading annotations into memory... Done (t=0.98s) creating index... index created! Loading and preparing results... Traceback (most recent call last): File "exp/tools/result2ann_obb.py", line 30, in res = coco.loadRes(args.det_file) File "/home/zhuzhenhua/.local/lib/python3.7/site-packages/pycocotools/coco.py", line 319, in loadRes with open(resFile) as f: FileNotFoundError: [Errno 2] No such file or directory: '/home/zhuzhenhua/pointobb-main/PointOBB/work_dir/test_pointobb_r50_fpn_2x_dior/pseudo_obb_result.json'

[Luo-Z13]

Traceback (most recent call last): File "tools/train.py", line 273, in main() File "tools/train.py", line 234, in main test_cfg=cfg.get('test_cfg')) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmdet/models/builder.py", line 58, in build_detector cfg, default_args=dict(train_cfg=train_cfg, test_cfg=test_cfg)) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/utils/registry.py", line 212, in build return self.build_func(*args, **kwargs, registry=self) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/cnn/builder.py", line 27, in build_model_from_cfg return build_from_cfg(cfg, registry, default_args) File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmcv/utils/registry.py", line 45, in build_from_cfg f'{obj_type} is not in the {registry.name} registry') KeyError: 'PointOBB is not in the models registry' loading annotations into memory... Done (t=0.98s) creating index... index created! Loading and preparing results... Traceback (most recent call last): File "exp/tools/result2ann_obb.py", line 30, in res = coco.loadRes(args.det_file) File "/home/zhuzhenhua/.local/lib/python3.7/site-packages/pycocotools/coco.py", line 319, in loadRes with open(resFile) as f: FileNotFoundError: [Errno 2] No such file or directory: '/home/zhuzhenhua/pointobb-main/PointOBB/work_dir/test_pointobb_r50_fpn_2x_dior/pseudo_obb_result.json'

你好，这个看起来是没有成功生成对应伪标签文件的问题，你可以再次用训练好的权重(如ep22)推理一遍

[Luo-Z13]

我将batch_size从2设置成了1，可以跑完24个epoch了，但是train的结果不太理想，map最好为0.386。所以想按照您给的设置，跑完24个epoch，但是会出现我前面提到的memory一直上升，我第一次跑10个epoch就断了，按照您说的延续checkpoint继续训练，第二次到22个epoch，然后23就彻底跑不了了，我看前面您23年回答过别人您用的mmcv版本是1.4.5，但是本文给的readme用的不是mmcv-full嘛，本来想再创建个环境，由于用的这个服务器磁盘空间满了，所以创建不了，为了保险才来问您。如有打扰，请见谅！望事事顺利！

对，readme里会安装mmcv-full，对于

File "/home/zhuzhenhua/.conda/envs/open-mmlab/lib/python3.7/site-packages/mmdet/init.py", line 25, in
f'MMCV=={mmcv.version} is used but incompatible. '
AssertionError: MMCV==1.7.2 is used but incompatible. Please install mmcv>=1.3.2, <=1.4.0.

你可以注释仓库里对应的mmcv版本限制，我的mmcv-full版本是1.4.5

[Luo-Z13]

我将batch_size从2设置成了1，可以跑完24个epoch了，但是train的结果不太理想，map最好为0.386。所以想按照您给的设置，跑完24个epoch，但是会出现我前面提到的memory一直上升，我第一次跑10个epoch就断了，按照您说的延续checkpoint继续训练，第二次到22个epoch，然后23就彻底跑不了了，我看前面您23年回答过别人您用的mmcv版本是1.4.5，但是本文给的readme用的不是mmcv-full嘛，本来想再创建个环境，由于用的这个服务器磁盘空间满了，所以创建不了，为了保险才来问您。如有打扰，请见谅！望事事顺利！

在我目前可运行的环境中，没有装mmcv，可能你需要调整一下

[zzhhzz666]

我将batch_size从2设置成了1，可以跑完24个epoch了，但是train的结果不太理想，map最好为0.386。所以想按照您给的设置，跑完24个epoch，但是会出现我前面提到的memory一直上升，我第一次跑10个epoch就断了，按照您说的延续checkpoint继续训练，第二次到22个epoch，然后23就彻底跑不了了，我看前面您23年回答过别人您用的mmcv版本是1.4.5，但是本文给的readme用的不是mmcv-full嘛，本来想再创建个环境，由于用的这个服务器磁盘空间满了，所以创建不了，为了保险才来问您。如有打扰，请见谅！望事事顺利！

在我目前可运行的环境中，没有装mmcv，可能你需要调整一下

佬，您好！mmcv-full版本的问题我暂时没有什么问题了，我用1.4.0还可以，然后又安装了您的1.4.5，注销了提示。但是现在运行推理指令sh test_p.sh 会出现附图的问题.老是在这里提问，觉得很很影响您。您的邮件肯定很忙，半天出来一个我的问题，应该会影响您吧。如果可以的话，我能加您一个联系方式吗，我的微信是hua15255859221，麻烦您啦！祝您的科研顺利！