-
The lane detector in this book is a bit too slow. This leads to Carla simulations which are not 30fps.
Ideas for improvements:
- train/run a lane detector on resized images (512 x 256 instead of …
-
I have searched related issues but cannot get the expected help.
my environment
GPU : GeForce RTX 3080
cuda version 11.1
pytorch 1.9.0
When I install mmcv-full =1.3.14
there's an import err…
-
I am trying to train Fast-SCNN on two Tesla P40s, but the training results stay at the first epoch, and I feel that the training is not successful. Is this caused by some post-processing operations? …
-
trainについて
画像サイズがクソデカイので、そのままのロードだとtrainに大変時間がかかる。width=2048で、train時間の85%は画像のロードである。したがって、データセットを事前にダウンサイズすると良いと思われる。例えば、width=1024のを事前に準備しておけば、trainの時間は1/4になると思われる。
-
**Motivation**
Since google colab has upgraded to torch 1.9, it would be great if mmcv can provide pre-built packages for torch 1.9.
Thanks!
-
When I used the fastscnn model, and I found somemistake, such as 'RuntimeError: fast_scnn_4x8x80k_lr0.12_cityscapes-f5096c79.pth is a zip archive'. Should I how to solve this problem or the model is …
-
@Tramac .When I run `eval.py` to predict my own images.I met a problem.The code can inference the first image in the folder of `validation` ,but when I inference the second image and then an error oc…
-
I try to train fast-scnn using default configs, in which the train pipeline is set to evaluate every 8000 iterations(total 80000), but after the first evaluation, it seems that the the validation p…
-
## ❓ Question
How to solve aten::floor converter not found?
## What you have already tried
I am trying to convert a jit trace of a Fast SCNN network into TensorRT. I've confirmed that the t…
-
**Checklist**
1. [x] I have searched related issues but cannot get the expected help.
2. [x] The bug has not been fixed in the latest version.
**Describe the bug**
It fails to train Fast-SCN…