-
I am using vgg16-faster RCNN to train cityscapes. I find that the model trained with mmdetection vgg16 backbone is much lower than the model trained with other faster RCNN implementations (I use https…
-
## ❓ Questions and Help
When training the model on keypoints I noticed that you need to add keypoint_names and keypoint_flip_map as key value pairs respectively in the dictionary in MetadataCatalog…
-
```
>
报错信息
E:\baiduai\PaddleDetection>python -u tools/train.py -c configs/mask_rcnn.yml --use_vdl=true --vdl_log_dir=vdl_dir/scalar --eval -o use_gpu=true
Traceback (most recent call last):
…
-
我这边是在自己的Android项目中直接把ncnn作为一个module加进去的（git submodule add https://github.com/Tencent/ncnn.git），现在想要去掉一些layer但是不知道如何写cmakelists，以下是我现在的cmakelists
cmake_minimum_required(VERSION 3.4.1)
SET(CMAKE_B…
-
## Instructions To Reproduce the 🐛 Bug:
1. Full runnable code or full changes you made:
```shell
python projects/SparseR-CNN/train_net.py --num-gpus 8 \
--config-file projects/SparseR-CNN/conf…
-
If you do not know the root cause of the problem / bug, and wish someone to help you, please
post according to this template:
## Instructions To Reproduce the Issue:
1. what changes you made (`…
-
Hello,
I want to run cocosnet demo with command like
python3 image_translation_demo.py \\
-d CPU \\
-m_trn /cocosnet.xml \\
-is /input_mask.png \\
-ri /reference_image.jpg \\
…
-
**Environment**
- OS: Ubuntu 18.04.1
- GCC : 7.4.0
- PyTorch version 1.1.0
- How I installed PyTorch : pip
- GPU model TITANX x 4 (GM200) (12G)
- CUDA 10.00
I use the config `cascade_rcn…
-
Using albumentations for Augmentation yields labels_field error
Error goes like this:
```
Original Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/torch/utils/d…
-
I added an albumentations pipeline on this config:
```
# VOC Dataset
albu_train_transforms = [
# dict(
# type='ShiftScaleRotate',
# shift_limit=0.0625,
# scale_lim…