Pinned issues
Issues
- 1
使用最新的mmdetection框架的mask rcnn使用自己的数据集去训练多种类别的模型时出现了一个问题,比如训练35种类别,训练出来的模型识别跟mmdet/evaluation/functional/class_names.py文件和mmdet/datasets/coco.py中放入的类别的顺序有关,只要有一个放置错误,识别时类别就会反过来,而且我并不能知道这个多类别在文件放置的顺序,这个应该怎么办,是哪里的设置需要修改吗还是一个bug?
#11704 opened by liaoqingjian - 0
For mmdetection, how can i get the prediction result which is processed by NMS?
#11728 opened by FengRongYue - 0
- 1
考虑支持一下yolov8?
#11699 opened by hcnhcn012 - 1
Fine tune Grounding DINO as open set with custom coco
#11724 opened by narchitect - 2
Training a customised Deformable DETR
#11677 opened by saadyousuf45 - 0
mmgrounding dino 微调遗忘问题
#11725 opened by TingqiaoXu - 1
无法debug到模型源码
#11705 opened by whj-tech - 0
- 0
Evaluating after training nuscenes mini dataset
#11721 opened by AbdilahiMJama - 0
[mm grounding dino]
#11720 opened by SHX9610 - 0
Save only detected images
#11719 opened by Syed05 - 0
Training Mask RCNN without BBox
#11718 opened by ruyadisk - 0
fps benchmark tool can't work
#11717 opened by KRL666 - 0
python demo/image_demo.py ...... error : Initializing libiomp5md.dll, but found libiomp5md.dll already initialized. in windows11
#11715 opened by yfq512 - 0
- 0
- 0
python tools/train.py configs/yolox/yolox_tiny_8xb8-300e_coco.py, than report AssertionError:
#11711 opened by realTravisYou - 0
- 0
TypeError: FasterRCNN.__init__() got an unexpected keyword argument 'bbox_head'
#11709 opened by Grushus - 1
权重文件的使用方法
#11706 opened by Leomin12138 - 0
- 0
Bbox feats (embeddings) from ONNX model
#11707 opened by GeorgePearse - 2
KeyError: 'gt_instances' of eval_metric.py
#11696 opened by HaMeow-lst1 - 2
- 0
- 0
How to use the intel camera api
#11702 opened by LLL632669638 - 0
how use tensorboard to show train log file
#11701 opened by yfq512 - 0
How to select which classes to detect on pretrained model?
#11700 opened by SFSimDev - 0
- 1
bbox_overlaps error
#11698 opened by zhaozhen2333 - 1
RuntimeError: `self._train_loop` should not be None when calling train method. Please provide `train_dataloader`, `train_cfg`, `optimizer` and `param_scheduler` arguments when initializing runner.
#11697 opened by wdzwdxy - 0
获取实例分割结果时报错
#11695 opened by yeyujujishou - 0
Compile for inference
#11694 opened by matthost - 0
BUG in DAB-DETR when with_random_refpoints=True
#11692 opened by YangHan-Morningstar - 0
BUG: clip_grad doesn't work
#11691 opened by Pixie8888 - 0
ONNX for RTMdet
#11689 opened by elbaro - 0
Need help in VGG16 Faster R-CNN implementation
#11688 opened by quitsv - 0
How to modify the conv1 of the pre training weights of the Mask RCnn to fit my 5-channel dataset
#11687 opened by MoonDiana17 - 0
./tools/dist_test.sh evaluation value is all zero
#11686 opened by fathur-rs - 0
Config not reusable for different models
#11685 opened by Grushus - 0
Error in the demos with Google Colab
#11684 opened by antoineedy - 2
Hydranet - separate prediction heads per class
#11683 opened by GeorgePearse - 0
Cannot reproduce rtmdet FPS
#11682 opened by mmeendez8 - 0
pip install -v -e . 编译失败
#11681 opened by whj-tech - 0
The same model and the same configuration file, inference _ detector and inference _ mot results are not the same.
#11679 opened by zbl929 - 0
Many CPU cores are unused
#11678 opened by anastasia-spb - 0
Multi-Scaling Training
#11675 opened by MatheusLevy - 0
Error running mask_rcnn .
#11674 opened by guisolsan - 0
Why is everything -1 in the evaluation and 0 in the training
#11673 opened by blue-q