Skip to main content

A Powerful Serverless Pre-Learning and Post-Learning Analysis Toolkit

Project description

KXY: A Powerful Serverless Analysis Toolkit That Takes Trial And Error Out of Machine Learning Projects

License PyPI Latest Release Downloads



From PyPi:

pip install kxy

From GitHub:

git clone & cd ./kxy-python & pip install .


All heavy-duty computations are run on our serverless infrastructure and require an API key. To configure the package with your API key, run

kxy configure

and follow the instructions. To request a demo and get a trial API key, contact us.


The Docker image kxytechnologies/kxy has been built for your convenience, and comes with anaconda, auto-sklearn, and the kxy package.

To start a Jupyter Notebook server from a sandboxed Docker, run

docker run -i -t -p 5555:8888 kxytechnologies/kxy:latest /bin/bash -c "kxy configure <YOUR API KEY> && /opt/conda/bin/jupyter notebook --notebook-dir=/opt/notebooks --ip='*' --port=8888 --no-browser --allow-root --NotebookApp.token=''"

where you should replace <YOUR API KEY> with your API key and navigate to http://localhost:5555 in your browser.

To start a Jupyter Notebook server from an existing directory of notebooks, run

docker run -i -t --mount src=</path/to/your/local/dir>,target=/opt/notebooks,type=bind -p 5555:8888 kxytechnologies/kxy:latest /bin/bash -c "kxy configure <YOUR API KEY> && /opt/conda/bin/jupyter notebook --notebook-dir=/opt/notebooks --ip='*' --port=8888 --no-browser --allow-root --NotebookApp.token=''"

where you should replace </path/to/your/local/dir> with the path to your local notebook folder and navigate to http://localhost:5555 in your browser.


Higher ROI Machine Learning Projects

The kxy package utilizes information theory to takes trial and error out of machine learning projects.

From the get-go, the achievable performance analysis of the kxy package tells data scientists whether their datasets are sufficiently informative to achieve a performance (e.g. , maximum log-likelihood, and classification error) to their liking in a classification or regression problem, and if so what is the best performance that can be achieved using said datasets. No need to train tens of models to know what performance can be achieved.

The model-free variable selection analysis provided by the kxy package allows data scientists to train smaller models, faster, cheaper, and to achieve a higher performance than throwing all inputs in a big model or proceeding by trial-and-error.

Once a model has been trained, the kxy improvability analysis quantifies the extent to which the trained model can be improved without resorting to additional features. This allows data scientists to focus their modeling efforts on high ROI initiatives. No need to implement tens of fancy models on specialized hardware to see whether a trained model can be improved.

When a classification or regression model has successfully extracted all the value in using the features to predict the label, the kxy dataset valuation analysis allows data scientists to quickly quantify the performance increase (e.g. , maximum log-likelihood, and classification error) that a new dataset may bring about. No need to train or retrain tens of models with the new datasets to see whether the production model can be improved.

Model Audit

From understanding the marginal contribution of each variable towards the decision made by a black-box regression or classification model, to detecting bias in your trained classification and regression model, the kxy toolkit allows data scientists and decision markers to fully audit complex machine learning models.

Modern Financial Machine Learning

From non-Gaussian and memory-robust risk analysis, to alternative datasets valuation the kxy toolkit propels quants from the age of Gaussian distributions/linear regression/LASSO/Ridge/Random Forest into the age of modern machine learning, rigorously and cost-effectively.

Project details

Download files

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

Files for kxy, version 0.2.4
Filename, size File type Python version Upload date Hashes
Filename, size kxy-0.2.4-py3-none-any.whl (84.9 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size kxy-0.2.4.tar.gz (43.9 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page