-
I am applying the Faster-RCNN to detect the spots on tongue images. There a more than 200 spots on a tongue. How can i modify the number of bounding boxes in Faster-RCNN at test time?
Following is m…
-
I want to train fast_rcnn, it occurs a problem. But it works well when training faster_rcnn. The dataset is VOC format.
(open-mmlab) root@localhost:~/data/mmdetection# CUDA_VISIBLE_DEVICES=0,1,2…
-
No issue template in General questions. I use the Error report issue Template as follow
Thanks for your error report and we appreciate it a lot.
**Checklist**
1. I have searched related issue…
-
Any ideas why the no of dets is very high? I am testing 200 images ( KITTI dataset ) w…
-
TorchVision: 0.9.1+cu111
OpenCV: 4.5.1
MMCV: 1.0.5
MMDetection: 2.3.0+
MMDetection Compiler: GCC 9.3
MMDetection CUDA Compiler: 11.2
PyTorch: 1.8.1+cu111
PyTorch compiling details: PyTorch buil…
-
**1. 环境（environment）**
- Build OS and Version: Ubuntu 18.04
- RunTime OS Version: Android 7.1
- RunTime DEVICE: ARM
**2. Github版本**
- branch：hotfix_issue_1109
- commit: ebc079cfc5…
-
**Describe the bug**
Attempting to build on Jetson AGX Xavier, intended to work with the deepstream (tensorrt) sdk. Tried CMake with the experimental TensorRT build parameters, failed halfway through…
-
[2022-05-31 02:44:25] [316241c6] [30] urllib3.connectionpool: http://192.168.2.109:8080 "GET /api/v1/allocations/7.ee72787d-c20e-4824-b0da-ef3e3287e708.1/signals/preemption?timeout_seconds=60 HTTP/1…
-
Environment:
Windows 10
GPU
TensorFlow 1.13.1 (install from pip)
Anaconda
IDE: Spyder
python 3.6
I am going to convert matterport's MaskRCNN implementation from keras to tf.keras. It gives th…
-
I train with my own dataset with 1 class. when i trained at 1.0, i just changed anchor scales, num_classes, img_scale; the config convert from 1.0 to 2.8 as bellow. but the AP drops from 73% to 70%. I…