Freddsle/BI_ML_2021

BI_ML_2021

1. Two tasks about KNN in ./l01_knn/code/KNN.ipynb:

   • Fashion-MNIST dataset
   • Diabetes dataset

2. Practice with linear models and gradient descent (LinearRegression, LogisticRegression, Pipeline from sklearn) in l2_linmodels_GD/code/Linear_models.ipynb.

   load_boston dataset from sklearn and pokemon dataset used.

3. Practice with clustering (KMeans, AgglomerativeClustering, DBSCAN, OPTICS, SpectralClustering, MiniBatchKMeans, Birch) in l3_clustering/code/Cluster_tSNE.ipynb.

   load_digits dataset from sklearn and flow cytometry dataset (in l3_clustering/data_folder/flow_c_data.csv) used.

Install requirements with pip

Install requirements: pip install -r ./01_knn/code/requirements.txt

Install and run with poetry

# install poetry
# for details look for https://python-poetry.org/docs/
sudo apt-get install curl
curl -sSL https://raw.githubusercontent.com/python-poetry/poetry/master/get-poetry.py | python3 -

# poetry will be accessible in current session
source $HOME/.poetry/env

# prepare project
git clone https://github.com/Freddsle/BI_ML_2021
cd ./code


# create env
poetry env use python3.10
poetry install

# Run
poetry run python code/file.py

## or for run .ipynb files
poetry run jupyter notebook