jcjohnson / jcjohnson/cnn-benchmarks
openCL branch of caffe reports much higher speeds
- Dominant language
- Python
- Stars
- 2.5k
- Forks
- 402
- PR merge metrics
- No merged PRs in 30d
Description
on [OpenCL-caffe](https://github.com/amd/OpenCL-caffe), there are performance matrices claiming speeds of about 4ms per image for training AlexNet with Radeon R290X, Considering this GPU is much weaker than a GTX 1080, these figures seem very weird compared with the 20ms in your tests.
What's your take on this?
Contributor guide
No contributing guide indexed for this repository
Research direction
Compare the OpenCL-caffe performance matrices for AlexNet training on the Radeon R290X with the benchmark results reporting about 20ms per image. Start by reviewing the benchmark methodology and hardware assumptions described in the issue and linked project. Done means explaining the discrepancy or identifying a reproducible benchmarking error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100