Collection and Implementation for ML/DL/RL/AI Algorithms

Collection and implementation of a variety of machine learning/deep learning/reinforcement learning code examples (notebooks and scripts), projects, and paper lists on various topics.

Some of the Jupyter notebooks are referenced from Andrew Ng's Deep Learning Specialization , D2L_ai's D2L and the Probabilistic Machine Learning book by Kevin Murphy.

Requirements

If you wish to run my codes directly, your code environment may need to satisfy the following requirements:

• Python 3.8+
• NumPy (pip install numpy or conda install numpy)
• Pandas (pip install pandas or conda install numpy)
• Pytorch (pip install torch or conda install numpy)

A conda environment is recommended:

export PROJECT_DIR=<ABSOLUTE path to the repository root>
git clone https://github.com/clarenceluo78/deep-learning-lookup $PROJECT_DIR
cd $PROJECT_DIR

conda create -n dl-lookup python=3.8
conda activate dl-lookup

conda install pytorch==1.7.1 torchvision==0.8.2 cudatoolkit=10.1.243 numpy=1.19.2 -c pytorch -y
pip install -r requirements.txt

# if the following commands do not succeed, update conda
conda env config vars set PYTHONPATH=${PYTHONPATH}:${PROJECT_DIR}
conda env config vars set PROJECT_DIR=${PROJECT_DIR}

conda deactivate
conda activate dl-lookup

Utility modules

Utility module for example notebooks

A utility function file called util.py is created under lib directory. I use functions from this module whenever possible in the Jupyter notebooks. For more information you can look into the lib folder.