Skip to main content

A library for training and using risk & impactability models on Curia

Project description

PyPI version

Python 3.8

Quality Gate Status Bugs Coverage Maintainability Rating Reliability Rating Security Rating Vulnerabilities

Release

Curia Python SDK

Curia Python SDK is a library for training and using risk and impactability models on Curia.

For detailed documentation, including the API reference, see our docs at https://foundryai.github.io/curia-python-sdk/.

Installing the Curia Python SDK

The Curia Python SDK is built to PyPi and can be installed with pip as follows:

pip install curia

You can install from source by cloning this repository and running a pip install command in the root directory of the repository:

git clone https://github.com/FoundryAI/curia-python-sdk.git
cd curia-python-sdk
pip install .
Supported Operating Systems

Curia Python SDK supports Unix/Linux and Mac.

Supported Python Versions

Curia Python SDK is tested on:

  • Python 3.8
Curia Permissions

Curia Python SDK will utilize the Curia Platform when training models and generating predictions. You will need access to the platform with appropriate permissions to fully utilize the SDK.

Running tests

Curia Python SDK has unit tests. To run the tests:

python setup.py pytest
Building Sphinx docs

Curia Python SDK has Sphinx docs. To build the docs, run:

cd doc
make html

To preview the site with a Python web server:

cd docs/_build/html
python -m http.server 8000

View the docs by visiting http://localhost:8080

Curia API Token

To use the Curia Python SDK you will need a Curia API Token. To access your API Token visit https://app.curia.ai/settings.

Use gnu-sed

Visit https://medium.com/@bramblexu/install-gnu-sed-on-mac-os-and-set-it-as-default-7c17ef1b8f64 to see how to install gnu-sed for consistency in fixing swagger imports export PATH="/usr/local/opt/gnu-sed/libexec/gnubin:$PATH"

Using the Curia Python SDK

from curia.session import Session
from curia.risk import RiskModel
from curia.synthetic_data import generate_data

# Create synthetic data (demo/testing purposes only)
(X_train, X_test, _, _, y_train, y_test, _, _, _, _) = generate_data(binary_outcome=True)

# Create a session
curia_session = Session(api_token="YOUR_API_TOKEN")

# Instantiate a model
model = RiskModel(
    session=curia_session, 
    name="your-model-name",
    project_id="YOUR_PROJECT_ID",
    environment_id="YOUR ENVIRONMENT_ID"
)

# Train a model on the Curia Platform
model.train(features=X_train, label=y_train)

# Get predictions from your model on the Curia Platform
predictions = model.predict(features=X_test)

Cut your own release (not recommended)

Sometimes we may have updates that are currently in the development environment, but not accessible in production yet. When this happens, the SDK will not reflect the latest changes in develop. To be able to use new features of the API, you can cut a new release of the SDK and use the latest version.

BE CAREFUL - MAKE SURE YOU KNOW WHY YOU ARE DOING THIS

To cut a new release based on the Dev API, run make build-api-dev. In order to get some sed commands to work on Mac, you may need to install and use Gnu-Sed

First, update the version in src/curia/__init__.py Then build the source distribution: python setup.py sdist bdist_wheel Finally, upload the new distribution to pypi: python -m twine upload dist/* You will be prompted for a username and password.

  • Your username is __token__
  • Your password is stored in 1Password in the Curia vault in the PyPi Curia Project secure document

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

curia-0.0.101.tar.gz (141.7 kB view details)

Uploaded Source

Built Distribution

curia-0.0.101-py2.py3-none-any.whl (444.1 kB view details)

Uploaded Python 2 Python 3

File details

Details for the file curia-0.0.101.tar.gz.

File metadata

  • Download URL: curia-0.0.101.tar.gz
  • Upload date:
  • Size: 141.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.7.1 importlib_metadata/4.10.0 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.9

File hashes

Hashes for curia-0.0.101.tar.gz
Algorithm Hash digest
SHA256 0777f25810a9a6a760283dee52b73354f2cb06f3b51e692333f515ef6a3415bb
MD5 e5a955c4639841655e2cfa7e1235036a
BLAKE2b-256 9cf1d70c6191826ca645405ebd1f2ccef0d261881fa729535025d727200f649e

See more details on using hashes here.

File details

Details for the file curia-0.0.101-py2.py3-none-any.whl.

File metadata

  • Download URL: curia-0.0.101-py2.py3-none-any.whl
  • Upload date:
  • Size: 444.1 kB
  • Tags: Python 2, Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.7.1 importlib_metadata/4.10.0 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.9

File hashes

Hashes for curia-0.0.101-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 289b52a4a06e18596fe40774a3445b055d496b0606be4e0987c023da040d09ea
MD5 899ebf091a87718e768e22f3d034a1b0
BLAKE2b-256 b78ecc0ff03c5574cce2ed0edd0a878c8a7a041fc8637e2d4236d04bcff86f7b

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page