Peter

i want to show the mAP and every class's AP

> @LiangYong1216 You should change the dictionary of class names. i do not know how to change . could you give me a example?i change the CLASSES={"data","fig","hazelnut"}....but no effective.....

> Hi, I have the same problem. Did you solve it? > @LiangYong1216 i set the class_names=[my classes] in the test_ins_vis.py, and the results is right, you can try it

those are my related code: My environment:         cuda:10.1  cudnn:7.6.5  mmcv:0.2.16  mmdet:1.0 pytorch:1.4 torchvision:0.5 my dataset only have 3 classes:          so i set num_classes=4, dataset_type = 'CocoToyDataset' data_root =...

ImportError: /home/liangyong/anaconda3/envs/solov2/lib/python3.7/site-packages/mmdet-1.0.0+0b0dfb3-py3.7-linux-x86_64.egg/mmdet/ops/sigmoid_focal_loss/sigmoid_focal_loss_cuda.cpython-37m-x86_64-linux-gnu.so: undefined symbol: _ZN6caffe26detail37_typeMetaDataInstance_preallocated_32E can you help me??

First: ImportError: /home/liangyong/anaconda3/envs/solov2/lib/python3.7/site-packages/mmdet-1.0.0+0b0dfb3-py3.7-linux-x86_64.egg/mmdet/ops/sigmoid_focal_loss/sigmoid_focal_loss_cuda.cpython-37m-x86_64-linux-gnu.so: undefined symbol: _ZN6caffe26detail37_typeMetaDataInstance_preallocated_32E My environment: cuda10.1 pytorch1.4 torchvision0.5 second: Traceback (most recent call last): File "/home/liangyong/SOLO/mmdet/ops/sigmoid_focal_loss/sigmoid_focal_loss.py", line 5, in from . import sigmoid_focal_loss_cuda ImportError: attempted relative import...

> import os os.chdir('your d2det directory') !rm -r build !python setup.py develop > […](#) > On Tue, Nov 17, 2020 at 9:14 AM LiangYong1216 ***@***.***> wrote: First: ImportError: /home/liangyong/anaconda3/envs/solov2/lib/python3.7/site-packages/mmdet-1.0.0+0b0dfb3-py3.7-linux-x86_64.egg/mmdet/ops/sigmoid_focal_loss/ sigmoid_focal_loss_cuda.cpython-37m-x86_64-linux-gnu.so:...

UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the...

> @WXinlong how to reduce memory? I use single gpu to train model on just 20 pics data, this problem also is here. The computer have more than 20G memory(not...

有人说是预训练文件加载失败，可是也不知道需要怎样处理 ------------------ 原始邮件 ------------------ 发件人: "WXinlong/SOLO"