This is the Git repository for the source code of the framework used for realizing GTI770-Systèmes intelligents et apprentissage machine course's labs.
This code is only the framework and is incomplete to let the student explore several machine learning algorithms. It is used jointly with multiple datasets, such as GalaxyZoo, Million Song Dataset and Spambase Data Set.
Students need to complete with their own code to solve classification problems automatically using different machine learning algorithms such as KNN, Naive Bayes, SVM, Neural Networks and Decision tree/Random Forests.
This framework has many dependencies, such as OpenCV 3.x.x, scikit-learn and TensorFlow. A best practice consists of running the code using a Docker environment built with all dependencies : Machine Learning Docker Environment. This framework has some code that can be GPU-accelerated using an NVIDIA GPU.
-
Maintained by:
-
Where to file issues:
-
Supported architectures:
[amd64][amd64-nvidia]
- 1.5 GB free hard disk space
- A minimum of a 4-core, 4-thread x86 CPU.
- A minimum of 8 GB of RAM, 16 GB or more is highly recommended.
- PyCharm Professional IDE (optional).
OpenCV and TensorFlow, can be GPU-accelerated using NVIDIA GPU.
The OpenCV version required to run this code is OpenCV 3.3.x+. OpenCV must be compiled for Python3.
Create an Anaconda virtual environment with Python minimum version 3.5 :
conda create --name gti770_env python=3.5
Activate the environment :
source activate gti770_env
Install the requirements :
conda install nb_conda
pip3 install -r requirements.txt
pip3 install git+https://github.com/hlin117/mdlp-discretization
To launch the script in Jupyter :
cd core
jupiter notebook
- Create a branch by feature and/or bug fix
- Get the code
- Commit and push
- Create a pull request
feature/ [Short feature description] [Issue number]
fix/ [Short fix description] [Issue number]
+ Added [Short Description] [Issue Number]
- Deleted [Short Description] [Issue Number]
* Changed [Short Description] [Issue Number]