farc
Submission for the 2nd Bandwidth Estimation Challenge at ACM MMSys 2024
Science Score: 75.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 4 DOI reference(s) in README -
✓Academic publication links
Links to: acm.org -
○Academic email domains
-
✓Institutional organization owner
Organization streaming-university has institutional domain (streaming.university) -
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (12.3%) to scientific vocabulary
Repository
Submission for the 2nd Bandwidth Estimation Challenge at ACM MMSys 2024
Basic Info
Statistics
- Stars: 4
- Watchers: 5
- Forks: 1
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Public repository for FARC
Paper Title: Offline Reinforcement Learning for Bandwidth Estimation in RTC Using a Fast Actor and Not-So-Furious Critic (🔗Link)
Submission for the 2nd Bandwidth Estimation Challenge at ACM MMSys 2024
🥈 FARC Ranked 2nd in the final challenge
Installation
Install the required Python packages
bash
pip3 install -r requirements.txt
Download both testbed (🔗Link) and emulated (🔗Link) datasets from and put them into Datafolder
Split them into train and testfolders. We used 98% of the data for training and 2% for testing.
Final structure should look like this:
|-- Data
| |-- emulated_dataset
| | |-- train
| | |-- test
| |-- testbed_dataset
| | |-- train
| | |-- test
Training
To train the model, first run the train_critic.py script to train the critic model.
Once the training for the critic model is done, you can run the train_actor.py script to train the actor model.
Evaluation
To evaluate the model, run the evaluate.py script.
This will evaluate the model on the test subset of emulated dataset and save the results in the figs folder.
If you want to reproduce the figures in the paper, extract the call traces in the test-traces.zip and change the line 18 (data_dir) in eval.py to point to this folder.
This will evaluate the model on these call traces and save the figures in the figs folder.
ONNX Visualization
You can find the visualization of the ONNX model using the Netron tool below:
Citation
If you use this code in your research, please cite our paper:
@inproceedings{FARC,
author = {Çetinkaya, Ekrem and Pehlivanoglu, Ahmet and Ayten, Ihsan U. and Yumakogullari, Basar and Ozgun, Mehmet E. and Erinc, Yigit K. and Deniz, Enes and Begen, Ali C.},
booktitle = {Proceedings of the 15th ACM Multimedia Systems Conference},
doi = {10.1145/3625468.3652184},
publisher = {Association for Computing Machinery},
series = {MMSys'24},
title = {{Offline Reinforcement Learning for Bandwidth Estimation in RTC Using a Fast Actor and Not-So-Furious Critic}},
url = {https://doi.org/10.1145/3625468.3652184},
year = {2024}
}
Owner
- Name: Streaming University
- Login: streaming-university
- Kind: organization
- Location: Turkey
- Website: https://streaming.university
- Repositories: 14
- Profile: https://github.com/streaming-university
Multimedia Streaming Research group at Ozyegin University
Citation (CITATION.cff)
cff-version: 1.2.0
message: If you use this software, please cite both the article from preferred-citation and the software itself.
authors:
- family-names: Çetinkaya
given-names: Ekrem
- family-names: Pehlivanoglu
given-names: Ahmet
- family-names: Ayten
given-names: Ihsan U.
- family-names: Yumakogullari
given-names: Basar
- family-names: Ozgun
given-names: Mehmet E.
- family-names: Erinc
given-names: Yigit K.
- family-names: Deniz
given-names: Enes
- family-names: Begen
given-names: Ali C.
title: Offline Reinforcement Learning for Bandwidth Estimation in RTC Using a Fast Actor and Not-So-Furious Critic
version: 1.0.0
url: https://doi.org/10.1145/3625468.3652184
doi: 10.1145/3625468.3652184
date-released: '2024-04-18'
preferred-citation:
authors:
- family-names: Çetinkaya
given-names: Ekrem
- family-names: Pehlivanoglu
given-names: Ahmet
- family-names: Ayten
given-names: Ihsan U.
- family-names: Yumakogullari
given-names: Basar
- family-names: Ozgun
given-names: Mehmet E.
- family-names: Erinc
given-names: Yigit K.
- family-names: Deniz
given-names: Enes
- family-names: Begen
given-names: Ali C.
title: Offline Reinforcement Learning for Bandwidth Estimation in RTC Using a Fast Actor and Not-So-Furious Critic
doi: 10.1145/3625468.3652184
url: https://doi.org/10.1145/3625468.3652184
type: conference-paper
pages: 388–393
year: '2024'
isbn: '9798400704123'
collection-title: Proceedings of the 15th ACM Multimedia Systems Conference
conference:
name: MMSys '24
publisher:
name: Association for Computing Machinery
address: New York, NY, USA
GitHub Events
Total
- Watch event: 4
Last Year
- Watch event: 4
Issues and Pull Requests
Last synced: 6 months ago
All Time
- Total issues: 1
- Total pull requests: 0
- Average time to close issues: about 3 hours
- Average time to close pull requests: N/A
- Total issue authors: 1
- Total pull request authors: 0
- Average comments per issue: 1.0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
- Gdragon14131113 (1)
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels
Dependencies
- onnxruntime *
- torch *
- tqdm *