Enterprise Machine Learning Platform For Teams That Ship
Skymind is an ML platform that guides engineers through the entire workflow of building and deploying machine learning models for enterprise applications on JVM infrastructure. It can be installed on any environment: cloud, on-premises, bare-metal, or hybrid systems. For those that rely on Cisco, Skymind ships with Cisco Hyperflex hardware.
Extract, transform, and load (ETL) data from storage or streaming sources for use in model training and in production. Native integrations include Spark, Hadoop, Kafka, Cassandra, and Amazon EMR.
Use the built-in, Java-based framework to develop, train, and test models in a version-controlled and collaborative environment. Engineers working alongside data scientists can import Python-based notebooks and models such as Keras and TensorFlow for use in JVM environments.
Expose developed models and transformations to enterprise applications via REST API endpoints or the Java client. Monitor and report on performance and resource usage. Retrain, test, and promote models into production.
Skymind lets engineers work in the language they know best (Java), in existing enterprise environments (JVM), in their interface of choice (CLI or notebook).
import skil_clientuploads = client.upload("tensorflow_rnn.pb") new_model = DeployModel(name="recommender_rnn", scale=30, file_location=uploads.path) model = client.deploy_model(deployment_id, new_model) ndarray = INDArray(array=base64.b64encode(x_in)) input = Prediction(id=1234, prediction=ndarray, needsPreProcessing=false) result = client.predict(input, "production", "recommender_rnn")
Get platform and ML support from the creators of Deeplearning4J—the most popular deep learning library for Java—and one of the top contributors to Keras, one of the most popular deep learning libraries for Python.