Blueoil provides two features.
- Training a neural network model
- Converting a trained model to an executable binary (or library), which utilize FPGAs for acceleration.
| Type | Status |
|---|---|
| blueoil |  |
| lmnet |  |
| dlk |  |
| docs |  |
See also CI settings.
You can see online documentation with enter.
Check out the Installation and Usage Guide page for getting started.
Note: Currently, Installation page is just in to be written, Please see Setup section to build docker on your development environment.
- GNU/Linux x86_64 with kernel version > 3.10
- NVIDIA GPU with Architecture >= 3.0 (Kepler)
- NVIDIA drivers >= 410.48
- Docker >=1.12 (or >=17.03.0)
- nvidia-docker >= 2.0
The blueoil is run on docker container with original docker image based on NVIDIA's CUDA images (cuda:10.0-cudnn7-devel).
The machine running the CUDA container only requires the NVIDIA driver, the CUDA toolkit doesn't have to be installed.
Please see the detail in the nvidia-docker's prerequisites.
There are some submodules in this repositry, so you should run git submodule update --init --recursive after cloning or git clone --recursive [this repository].
make build
Note: The private repository submodules are set to connect by ssh, if you want to use HTTPS, you should edit URLs in .gitmodules and run git submodule sync before git submodule update --init --recursive command. (see how to edit)
cd docs
make html
You can see generated documents in HTML format under docs/_build/html/ directory on your enviroment.
Also, you can see the deploy-preview online documentation from a Pull Request page that are integrated by netilify.
We can test each opereations of drore_run.sh by using shell script.
expect>= version 5.45
$ make test
You can test specific task.
$ CUDA_VISIBLE_DEVICES={YOUR_GPU_ID} make test-classification
$ CUDA_VISIBLE_DEVICES={YOUR_GPU_ID} make test-object-detection
$ CUDA_VISIBLE_DEVICES={YOUR_GPU_ID} make test-semantic-segmentation
You can also test the modules used in Blueoil.
$ make test-lmnet
$ make test-dlk