Kraken AutoML

Automation Where You Need It…

So you can focus on the work that makes an impact.

Truly Automated Machine Learning

Why spend extra time and money to build your machine learning models from scratch when Kraken can work data magic in no time at all?

The backbone of BigSquid.ai is its Kraken AutoML capabilities—this means your business can implement and iterate on its machine learning models faster and easier than ever before. Because Kraken hosts and automates the machine learning pipeline, you get access to decades of data science experience at any time, from anywhere with an internet connection.

The Secret Recipe For Success With AutoML

We’re not real wizards, but we are wizards of data science. We believe you should be able to ask your question and get answers without spending months and money reinventing the wheel. That’s why Kraken does a lot of the heavy lifting for you.

Kraken AutoML Automates the Process of Building Machine Learning Models With…

DATA PRE-PROCESSING

Kraken AutoML uses a variety of data science techniques to pre-process data when building models including null handling, cardinality, encoding, feature scaling, and more.

MODEL EXPLORATION

Kraken AutoML automatically explores the best in breed ML model approaches for the dataset you choose to analyze. With the click of a button, you’ll see model results from hundreds of approaches, and we’ll even show you how they perform.

ALGORITHM SELECTION

After applying hundreds of ML models against your data, Kraken AutoML will automatically select the best performing algorithm for you. But don’t worry, you can have your choice of any other algorithm that Kraken explored along the way. 

AUTOMATIC THRESHOLD TUNING

Kraken AutoML selects the correct threshold value to optimize the F1 scores of binary classification models. That optimization enables you to get a more realistic sense of how your binary models may perform in “the real world”.

HYPERPARAMETER OPTIMIZATION

With a single click, automatic Hyperparameter Optimization ensures that your model is precisely tuned for optimal predictive results.

AUTOMATIC HOLDOUT

Kraken AutoML uses both multi-fold cross validation and automatic holdout data when training your model to provide a reliable estimate of “real world” performance.

90% of Kraken users build and deploy their own machine learning model within 30 days of signup.

Same day access, no credit card required