bye bye added to app

smol changes

smol changes

smol changes

smol changes

smol changes

smol changes

testing an app

app.py

app.py

update readme

pain

lfs

pls

fixing

changes to git attr and readme

adding git attr

Create main.yml

initial commit

Merge pull request #2 from atodorov284/model_development

Format code for flake8

Format code for flake8

Add extra scripts. Format code. \n Add .vs to .gitignore

added a python script to compute shap values and make a feature importance plot

Update gitattributes file.

Add gitattributes file.

Update README with all info from the development stage.

Workflow fix for flake8 and reproducibility.

Update workflows and style. Setup main for predictions.

Created a predictor class to encapsulate and wrap the loaded models. \n Make a manual prediction to ensure the MLFlow tracking was correct, which is indeed the case. \n Can be accessed through prediction.py.

Added MLFlow data from Habrok. Registered the best models.

Convert parameter space into a range rather than a grid.

Experiments track train and test MSE and RMSE correctly \n Code refactoring

Added docker file

fix access to paths and set them up properly

Put model development pipeline in a .py file. Rewrote README. Made a config folder to store hyperparameter space searches. Co-authored with Chris and Csenge. Several TODOs in the README.

Set up MLFlow. Automatic server setup at localhost 5000. Experiments for decision tree, random forest, and xgboost set up, optimized through Bayesian search.

working on model development (xgboost and decision tree)

Co-authored-by: SzokeCsenge <[email protected]>

Merge pull request #1 from atodorov284/data_engineering

workflow fix

update workflows

added extra materials and update README

added train-test split, added saving mechanisms after the split, removed the normalizer class, added main.py

Cleaning up and modularizing the data pipeline

fixed time shift mechanism in data pipeline

pipeline preprocessing steps improvement/expansion

adding utils.py which was untracked

creating a pipeline

creating a pipeline