Comments (3)
by the way, i just wonder why some implementations are different with paper?
for example,
rmi_pool_size is 4 in train.sh
but in paper, 3x3 produce best performance...
thanks..
from rmi.
-
The pre-trained model is downloaded from https://hangzhang.org/PyTorch-Encoding/model_zoo/imagenet.html. You may check it, I will also try to load the pre-trained model in this repo.
-
rmi_pool_size=4
(DF
in the paper) is the size of the pooling kernel (also the stride) of the max/avg pooling. It is also the default setting (withrmi_radius=3
,R
in the paper) in most tables. It is not the best setting with the best performance, but it can work with less extra memory cost than the best setting.
Please check https://github.com/ZJULearning/RMI/releases/tag/1.0.0 for pre-trained models.
from rmi.
Thank you for your quick update :)
I will try to train DeepLabv3+ with resnet101 model on camvid
Thank you so much again
from rmi.
Related Issues (20)
- The RMI loss provides negative loss HOT 3
- About setting of baseline HOT 2
- Benchmarking on Cityscapes HOT 3
- Negative RMI loss HOT 1
- Training a DeepLabv3 model with output_stride=16, crop_size=513, and batch_size=8 on two 2080Ti GPUs HOT 3
- Hi! Can the RMI loss be used to measure the similarity of two natural images? HOT 1
- -
- cholesky_cuda: For batch 0: U(1,1) is zero, singular U. HOT 4
- Use CRF as post-process in image segmentation HOT 4
- Sample weights for RMI loss HOT 4
- Comparison with the Potential Energy Loss on Gibbs distribution HOT 2
- RuntimeError: CUDA error: an illegal memory access was encountered HOT 2
- The intuition behind RMI loss HOT 2
- two classification tasks with output channel = 1 HOT 4
- How to weight some examples if loss may be negative HOT 1
- Effect of resizing image & mask without keeping width-height ratio in the data augmentation HOT 3
- Please do not raise issues here. This repo is archived and will no longer be maintained. Some 3rd party links maybe helpful in the content.
- Please do not raise issues here. This repo is archived and will no longer be maintained. Some 3rd party links maybe helpful in the content.
- RMI HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from rmi.