版本问题

[Czeni1]

您好，我一开始用了CUDA11.7，pytorch1.13.1，mmcv-full1.5.1，mmdet2.24.1，测试mmdet也显示是安装成功了，之后运行train.py，出现了以下错误： File "tools/train.py", line 237, in main() File "tools/train.py", line 211, in main model.init_weights() File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\base_module.py", line 116, in init_weights m.init_weights() File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\base_module.py", line 105, in init_weights initialize(self, self.init_cfg) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\cnn\utils\weight_init.py", line 613, in initialize _initialize(module, cp_cfg) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\cnn\utils\weight_init.py", line 517, in _initialize func(module) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\cnn\utils\weight_init.py", line 489, in call load_checkpoint( File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\checkpoint.py", line 581, in load_checkpoint checkpoint = _load_checkpoint(filename, map_location, logger) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\checkpoint.py", line 520, in _load_checkpoint return CheckpointLoader.load_checkpoint(filename, map_location, logger) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\checkpoint.py", line 285, in load_checkpoint return checkpoint_loader(filename, map_location) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\checkpoint.py", line 425, in load_from_torchvision model_urls = get_torchvision_models() File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\runner\checkpoint.py", line 132, in get_torchvision_models model_urls = mmcv.load(json_path) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\fileio\io.py", line 57, in load with StringIO(file_client.get_text(file)) as f: File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\fileio\file_client.py", line 1028, in get_text return self.client.get_text(filepath, encoding) File "D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\fileio\file_client.py", line 552, in get_text with open(filepath, 'r', encoding=encoding) as f: FileNotFoundError: [Errno 2] No such file or directory: 'D:\Anaconda3\envs\xxm\lib\site-packages\mmcv\model_zoo/torchvision_0.12.json'

缺少这个torchvision0.12.json文件，我看了目录，确实没有这个文件，于是我想是不是我版本问题，所以我又重新创建了虚拟环境，在自己虚拟环境下安装了CUDA11.3，pytorch1.10.1，mmcv-ful1.5.1，mmdet2.24.1，但在测试mmdet是否安装成功的时候，显示这种错误，但我pip list看了是存在sparse-conv0.0.0。所以我在想是我CUDA的问题吗，一台服务器不能有两个CUDA吗，而且我是在我虚拟环境中安装的（呜呜呜，救救孩子吧，周一开组会汇报，非常感谢非常感谢非常感谢！！！）

Traceback (most recent call last): File "demo/image_demo.py", line 5, in from mmdet.apis import (async_inference_detector, inference_detector, File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\apis__init.py", line 2, in from .inference import (async_inference_detector, inference_detector, File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\apis\inference.py", line 13, in from mmdet.datasets import replace_ImageToTensor File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\datasets__init__.py", line 13, in from .utils import (NumClassCheckHook, get_loading_pipeline, File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\datasets\utils.py", line 11, in from mmdet.models.dense_heads import GARPNHead, RPNHead File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\models\init.py", line 7, in from .dense_heads import * # noqa: F401,F403 File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\models\dense_heads\init__.py", line 44, in from .fsaf_dy_head import FSAFDYHead File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\models\dense_heads\fsaf_dy_head.py", line 11, in from .retina_dy_head import RetinaDYHead File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\models\dense_heads\retina_dy_head.py", line 9, in from .cuda_dynamic_conv_module import DyConv2D File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\models\dense_heads\cuda_dynamic_conv_module.py", line 6, in from .sparse_conv_net import * File "D:\Anaconda3\envs\amm\lib\site-packages\mmdet-2.24.1-py3.8.egg\mmdet\models\dense_heads\sparse_conv_net.py", line 17, in import sparse_conv ImportError: DLL load failed while importing sparse_conv: 找不到指定的程序。

[Cuogeihong]

有可能是你的sparse-conv安装的虚拟环境和最新的虚拟环境不匹配，应该在最新的虚拟环境下执行python setup.py install

[Cuogeihong]

可以尝试在linux服务器上进行实验，会比windows系统下做更好

[Czeni1]

可以尝试在linux服务器上进行实验，会比windows系统下做更好 好嘞好嘞，已经跑起来啦，非常感谢！！！

[Czeni1]

可以尝试在linux服务器上进行实验，会比windows系统下做更好

您好，我还想再问一下，那个AMM模块还有CE-GN是在gfl_dy_head.py这个文件里面嘛？我怎么找不到在哪呀？

[Cuogeihong]

AMM在这里https://github.com/Cuogeihong/CEASC/blob/2abfd1a99f1b0fe1ed3d51588b64549e1584da50/mmdet/models/dense_heads/gfl_dy_head.py#L468-L485 CE-GN在sparse_conv_net.py和cuda_dynamic_conv_module.py

[Czeni1]

AMM在这里

https://github.com/Cuogeihong/CEASC/blob/2abfd1a99f1b0fe1ed3d51588b64549e1584da50/mmdet/models/dense_heads/gfl_dy_head.py#L468-L485

CE-GN在sparse_conv_net.py和cuda_dynamic_conv_module.py

好嘞好嘞，谢谢您，感恩的心！！！