-
I have two models: u2net and u2netp, but node in my workflow dont see this models. Why?
This is node mix lab :
![1](https://github.com/user-attachments/assets/4d898383-f257-4792-ae96-e988f63ff886)…
-
GPU: Nvidia 3090
CUDA: 11.2
ffmpeg: 5.0.1
model: u2net
-
Specs of my server:
GPU: two RTX 3090 TIs
OS: Ubuntu 20.04.3 LTS
I'm following the instructions in the README file.
Docker images, git repos, and the weight of UNet are successfully installe…
-
我用了U2Net,按照README里面的命令
./app_u2net --model=../../data/u2net/u2net.trt --size=320 --batch_size=1 --img=../../data/sailboat3.jpg --show --savePath
但这样只能推理一张图,把batch_size=8之后,应该怎么输入图像路径呢?
-
Hey! Loved the paper.
Have you tried doing a comparison to U2Net? As for now, it's considered the state-of-the-art in SOD. I was wondering if you tested against it.
-
First of all,thank you for your well organized and memory friendly code.
Could you kindly tell me how to use your code for training u2net on the custom dataset?
What are the steps?
Thanks in advanc…
-
Hi, I finally got this python project to start loading models.
Realizing that python must be a 64bit version in order to install torch, and many other trial and error. :)
However, now I'm gettin…
-
-
A wonderful project!
but I have some problems on the training :
Why can reducing the channel value reduce the model size? I am confused about the channel and I'm a beginner here, could you tell me s…
-
I'v been looking in previous issues, but I could not find satisfying answer.
I have packed model using model-archiver in docker.
```bash
docker run --rm -it --name mar -v $(pwd)/output:/output …