Runtime Data Science using fabric code and cookiecutter
├── Makefile <- Makefile with commands for controlling aspects of the project.
├── README.md <- The top-level README for developers using this project.
├── cloud-mle
│ ├── data <- Scripts to download, upload or generate data.
│ ├── scripts <- Scripts to submit the tasks and deploy model to ML Engine.
│ ├── <MODEL_NAME> <- Python code to train and serve the model. For every
│ │ │ new model, create a new folder.
│ │ ├── model.py
│ │ ├── predict.py
│ │ └── train.py
│
├── data
│ ├── external <- Data from third party sources.
│ ├── interim <- Intermediate data that has been transformed.
│ ├── processed <- The final, canonical data sets for modeling.
│ └── raw <- The original, immutable data dump.
│
├── docs <- Place to put extra project-related documents.
│
├── models <- Trained and serialized models, model predictions, or model summaries.
│
├── notebooks <- Jupyter notebooks. Naming convention is a number (for ordering),
│ │ the creator's initials, and a short `-` delimited description, e.g.
│ │ `1.0-jqp-initial-data-exploration`.
│ ├── exploration <- Exploration and visualization code to determine hypothesis and
│ │ better understanding the data.
│ ├── features <- Features engineering to prepare data to model.
│ ├── ingestion <- Data munging modules to read from different data sources/streams,
│ │ transform and load data onto different destinations.
│ └── models <- Models proposed and related components, like metrics functions.
│
├── references <- Data dictionaries, manuals, and all other explanatory materials.
│
├── reports <- Generated analysis as HTML, PDF, LaTeX, etc.
│ └── figures <- Generated graphics and figures to be used in reporting.
│
└── services <- The production services that are to be deployed.