The ML Core Snap Pack accelerates the building, training, and testing of your machine learning model. With the ML Core Snap Pack, data scientists can still work in Python and Jupyter Notebook environments while also taking advantage of the ease of use, speed, and drag-and-drop simplicity of SnapLogic. AutoML capabilities are embedded in the ML Core Snap Pack to allow citizen data scientists to build their own machine learning models with minimal effort. The ML Core Snap Pack enables you to:
- Rapidly train, test, and cross-validate your model with a visual drag-and-drop interface.
- Leverage state-of-the-art ML algorithms based on mature open source libraries.
- Execute Python scripts remotely to leverage libraries such as TensorFlow, Keras, and others.
- Cut down on hand-coding for non-strategic, routine tasks in the machine learning lifecycle.
- Fine-tune and iterate on your model faster.
The ML Core Snap Pack includes the following Snaps:
Cross-Validator – Classification: Cross-validate classification data models using algorithms and choose the best possible algorithm for the dataset.
Cross-Validator – Regression: Cross-validate regression data models using algorithms and choose the best possible algorithm for the dataset.
Predictor – Classification: Predict the unlabeled data in a classification dataset using a data model.
Predictor – Regression: Predict the unlabeled data in a regression dataset using a data model.
Trainer – Classification: Train/generate a data model for a classification dataset using algorithms.
Trainer – Regression: Train/generate a data model for a regression dataset using algorithms.
Remote Python Script: Execute Pythons scripts remotely on the Python server.
AutoML: Automate the process of training a large selection of candidate machine learning models by providing minimal inputs.
Clustering: Perform exploratory analysis by identifying hidden groupings in data
Build a linear regression model with a drag-and-drop approach
The ML Core Snap Pack enables data scientists to configure models using Snaps – that is, through dragging and dropping. For example, the Predictor – Regression Snap has several state-of-the-art prediction algorithms baked into it (e.g., a linear regression algorithm). If a data scientist wants to build a linear regression model, they can do so simply by dragging this Snap onto SnapLogic’s Designer canvas. This minimizes the amount of coding you have to do when creating your ML model.
The ML Snap Packs are included in SnapLogic Data Science, an extension of the Intelligent Integration Platform that provides a visual drag-and-drop approach to developing and deploying machine learning models. Check out our other ML Snap Packs: ML Data Preparation Snap Pack and ML Analytics Snap Pack.
Learn more about the ML Core Snap Pack in the blog post, “SnapLogic November 2018 Release: Revolutionize your business with intelligent integration.”