We just published our Python SDK to enable model development in Python with the G2M platform API as a compute engine. Feel free to check it out here.
Why use the SDK?
This SDK is intended for analysts who are comfortable with Python but don’t want to manage their own compute resources. By using the Analyzr SDK you can focus on model building and tweaking while letting the Analyzr API take care of the cloud infrastructure required to run your managed cluster.
How does it work?
To train your model you will first log into the Analyzr API:
>>> from analyzrclient import Analyzer
>>> analyzer = Analyzer(host='insert_your_host')
>>> analyzer.login()
You can then start training your model:
>>> res = analyzer.propensity.train(df, algorithm='distributed-xgboost-classifier')
For more info
See these examples for more information on modeling parameters available. See also our SDK documentation. The SDK can handle additional pre-processing seamlessly such as homomorphic encryption (if you need to protect your data), SMOTE pre-processing (when dealing with low-incidence outcome), or outlier removal (comes in handy when clustering a dataset). The API will handle hypertuning on your behalf. If you want to specify hypertuning parameters explicitly you can do so using the param_grid, scoring, and n_splits attributes.
How Can We Help?
Feel free to check us out and start your free trial at https://app.g2m.ai or contact us below!
