This repository is intended to be a Machine Learning written in C++ for I+D purposes. The aim is be able to create a Machine Learning full performance, create a docker image from this code and deploy it into a production server easily.
As a single developer I'm not an expert about this topic, my goal is to share this project with the community and be able to improve the repository to have a full featured Machine Learning in C++.
Do not hesitate to email me at moriorgames@gmail.com if there is any issue with the project.
Create a docker container, install all dependencies and build the project with cmake. This will give us an image with the compiled application ready to run and test.
- To create the docker image
$ docker build -t moriorgames/ml .
- Run the container
$ docker run -td --name machine_learn moriorgames/ml
- Execute the program
$ docker exec -ti machine_learn ./main
- Execute the Tests
$ docker exec -ti machine_learn ./tests/testsuite
- Execute the code coverage locally
$ sh scripts/coverage.sh
I like to preserve information about tagging and pushing imageFile to docker
- First login at dockerhub
$ docker login
username
password
- Tag an existing image
$ docker tag moriorgames/ml moriorgames/ml:commit_sha
- Push the image to the public repository
$ docker push moriorgames/ml:commit_sha
$ docker push moriorgames/ml:latest