Comments (8)
Hi cencenxy, thank you for your comment.
I added SeriesAdditionalBlock class in model.py.
Please see the updated model.py file.
I appreciate your advice.
I made csinet-pytorch model architecture as written as the CsiNet, so it works exactly same as it.
However, in my code, the test metric was MSE (not NMSE), so the result will be somewhat different.
I updated NMSE function in train.py, and implemented NMSE for the test metric.
Now, you can see NMSE as a test metric.
Thank you!
from csinet-pytorch.
Hi cencenxy, thank you for your comment.
I added SeriesAdditionalBlock class in model.py. Please see the updated model.py file. I appreciate your advice.
I made csinet-pytorch model architecture as written as the CsiNet, so it works exactly same as it. However, in my code, the test metric was MSE (not NMSE), so the result will be somewhat different. I updated NMSE function in train.py, and implemented NMSE for the test metric. Now, you can see NMSE as a test metric.
Thank you!
Thank you for your patient reply and code update. I still have some questions that I don't understand and I would like to ask for your advice. In the code of 'Deep learning for massive MIMO CSI feedback', the author used a lot of '.m' data based on the cost2100 channel. So why is the data in your code a CSV file?
from csinet-pytorch.
As you said, CsiNet authors used COST2100 channel model.
However, I wanted to do an experiment in mmWave channel model, as well as sub 6GHz.
Because COST2100 doesn't support mmWave, I chose the QuaDRiGa as a dataset, whose output is .csv file.
There's no other reason except I wanted to generate mmWave channel data.
Thank you for your question.
from csinet-pytorch.
As you said, CsiNet authors used COST2100 channel model. However, I wanted to do an experiment in mmWave channel model, as well as sub 6GHz. Because COST2100 doesn't support mmWave, I chose the QuaDRiGa as a dataset, whose output is .csv file. There's no other reason except I wanted to generate mmWave channel data.
Thank you for your question.
Thank you very much for your patient reply. Best regards.
from csinet-pytorch.
Thank you! If you have any issues, feel free to comment here!
from csinet-pytorch.
I changed the value of encoded_dim to change the compression ratio, but the value of NMSE didn't change
from csinet-pytorch.
Hi, Andy7129z, thank you for your comment.
I think you changed the variable 'encoded_dim' in model.py line 11.
I'm afraid you were somewhat confused since it was a useless variable.
I revised the code for it so that you can check model.py line 11.
Instead, you might want to change the variable 'dim' in main.py line 37.
Please check if the NMSE changes when you change 'dim' in main.py.
Thank you!
I changed the value of encoded_dim to change the compression ratio, but the value of NMSE didn't change
from csinet-pytorch.
can you give some predefined setting about the dataset, indoor or outdoor and frequency
from csinet-pytorch.
Related Issues (3)
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 csinet-pytorch.