Skip to content

Machine Learning Operations system labs in MADE

Notifications You must be signed in to change notification settings

olyamasaeva/Made_MLOps

This branch is up to date with made-mlops-2022/Made-ML12-Masaeva-Olga:main.

Repository files navigation

Made-ML12-Masaeva-Olga

Installation:

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Usage:

For training set:
python ml_example/train_pipeline.py configs/train_config.yaml
For prediction:
python ml_example/predict_pipeline.py configs/predict_config.yaml

Project Organization

├── README.md          <- The top-level README for developers using this project.
├── data
│   ├── metrics        <- Data from some models metrics
│   ├── predicts       <- Data from some model predicts
│   ├── processed      <- The final, canonical data sets for modeling.
│   ├── raw            <- The original, immutable data dump.
│   └── make_dataset.py<- Code for datasets process
│
│
├── models             <- Trained and serialized models, model predictions, or model summaries
│
├── notebooks          <- Jupyter notebooks for EDA and some prototyping
│
├── requirements.txt   <- The requirements file for reproducing the analysis environment, e.g.
│                         generated with `pip freeze > requirements.txt`
├── configs            <- Configs for pipelines
│
├── predict_pipeline.py <- prediction pipeline file
│
├── train_pipeline.py   <- training pipeline file
│
├── setup.py           <- makes project pip installable (pip install -e .) so src can be imported
|
├── custom_logs        <- folder with logging information & code
|
├── enities            <- enities code folder
|
├── features           <- features code folder
|
├── metrics            <- metrics class folder
|
├── model_gen          <- metrics class folder
|
└──  models             <- serialize models storage

About

Machine Learning Operations system labs in MADE

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 96.9%
  • Python 3.1%