This is the working repo for the Optimization Special Interest Group (SIG), which is responsible for solutions to optimize ONNX models, including model compression techniques (e.g. quantization, pruning, distillation). The optimization solutions investigated by this SIG predominantly operate on ONNX core files as defined by the Architecture & Infrastructure SIG and use ONNX operations as defined by the Operator SIG - i.e. ONNX as input, optimization on ONNX file(s), ONNX as output. This SIG is open to techniques developed in any relevant, open-sourced optimization frameworks and capabilities and may sponsor projects that work in specific optimization frameworks. This SIG also invite discussions on making the ONNX standard more amendable to ONNX optimization technologies.
This repo contains all the artifacts, materials, meeting notes, and proposals regarding the Optimization SIG.
Feedbacks and contributions of other optimization tools are welcome and encouraged for discussion at the monthly meetings.
Please sign up at https://slack.lfai.foundation/ and join onnx-optimization channel.
- 4th Thursdays of the month, from 5PM PST to 6PM PST
- Check LFX calender for login link
- Freddy Chiu (Intel)
- Justin Chu (Microsoft)
- SIG lead(s) will drive the monthly Optimization SIG meeting.
- Meeting annoucement will be posted in our Slack channel.
- Specific Optimization-SIG projects may meet at a higher frequency.
- SIG lead(s) may delegate leadership of project meetings to their respective project technical lead(s).
- Monthly Optimization SIG meetings may take place at the begining of a specific project meeting.
- Feedbacks and topic requests are welcome by all.
- Slack channel: https://lfaifoundation.slack.com/archives/C05GKQH0H34
- Mailinglist:
- Documents and artifacts: https://github.com/onnx/sigs/tree/main/optimizations
- Optimization SIG meeting notes are kept in the meeting folder.