AMPs in fully airgapped environments

Some AMPs are built to work in environments that does not have outbound network access. Learn how to set up the environment for AMPs to work.

  • Host your internal PyPI server.
  • Create your Custom AMP Catalog or download an AMP project from the GitHub repository listed in the below table. After downloading, upload it as a ZIP file during the project creation process.

Steps

  • Each AMP has a requirements.txt file, which lists all the Python packages it needs. This file is present in the root of the public GitHub repository.
  • You must ensure that all the packages listed in the requirements.txt are available on your custom-hosted PyPI server.
  • Change the PIP_INDEX_URL environment variable to point to the custom-hosted PyPI server URL.

Below are the list of AMPs that can be used in the airgapped environments without outbound network access.

Title Label Short Description Git URL
Churn Modeling with scikit-learn churn-prediction Build an scikit-learn model to predict churn using customer telco data. Churn Modeling with scikit-learn
Deep Learning for Anomaly Detection anomaly-detection Apply modern and deep learning techniques for anomaly detection to identify network intrusions. Deep Learning for Anomaly Detection
Canceled Flight Prediction canceled-flight-prediction Perform analytics on a large airline dataset with Spark and build an XGBoost model to predict flight cancellations. Canceled Flight Prediction
Getting Started with the CML API apiv2 Demonstration of how to use the CML API to interact with CML. Getting Started with the CML API
AutoML with TPOT automl-with-tpot AutoML using TPOT, distributed with Dask. AutoML with TPOT
Train Gensim's Word2Vec gensim-w2v Demonstration of how to train Gensim's Word2Vec for a non-language use case. Train Gensim's Word2Vec
Continuous Model Monitoring continuous-model-monitoring Demonstration of how to perform continuous model monitoring on CML using Model Metrics and Evidently.ai dashboards. Continuous Model Monitoring
Distributed XGBoost with Dask on CML dask-on-cml How to perform distributed training of an XGBoost model using Dask on CML. Distributed XGBoost with Dask on CML