-
请问下人像分割v2里边的MobileSeg这个使用的是什么算法?只看到了backbone是mobilenetv3,想知道具体是由什么算法哪个网络改进得到的v2版本的人像分割网络
-
### 请提出你的问题 Please ask your question
gpu:3060laptop cudnn:8.1 paddlepaddle版本:2.2.2 cuda版本11.2
Windows anaconda虚拟环境中python版本是3.8.13
使用https://blog.csdn.net/u013171226/article/details/115179480 中2.5…
-
##### System information (version)
hello, The ONNX model I used had five inputs and was successfully converted to the OpenVino model, but ran with an error.
![图片](https://user-images.githubuserc…
-
Sik-Ho Tsang. [Review: Vision Transformer (ViT)](https://sh-tsang.medium.com/review-vision-transformer-vit-406568603de0).
Dosovitskiy A, Beyer L, Kolesnikov A, et al. [An image is worth 16x16 words: …
-
1. is the "mobile" part MobileNetV3?
2. if yes, how do you initialize it before training? random or already pretrained one?
3. why there is some custom function related to checkpoint loading in hel…
-
### Issue Type
Support
### OS
Ubuntu
### OS architecture
x86_64
### Programming Language
Python
### Framework
TensorFlowLite
### Model name and Weights/Checkpoints URL
Mobilenetv3 small tra…
-
**Describe the bug**
When I update timm from 0.5.4 to latest version, I encountered error.
I only used `forward_features` from timm, then append some custom module.
![image](https://user-images.…
-
请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem
(先谢谢大佬们回答)
在训练前验证下载的预训练权重的hmean
和训练后测试训练权重的hmean
- 系统环境/System Environment:Windows10 RTX8000
- 版本号/Vers…
-
训练命令:python tools/train.py -c configs/det/det_mv3_db.yml \
-o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained
报错信息
![image](https://user-images.githubusercontent.co…
-
### Current Behavior
Hi, I'm using WandB as usual to log my program, I had to Run it on a Server which is not suitable for online logging. So I am Using `Offline` mode to handle the situation.
I'm u…