Advanced ML and API approaches¶
Topic | Description |
---|---|
Track ML experiments with MLFlow | Learn how to programmatically build a model with DataRobot, and then export and host the model in AWS SageMaker. |
Customize lift charts | Leverage popular Python packages with DataRobot's Python client to recreate and augment DataRobot's lift chart visualization. |
Select models using custom metrics | This AI Accelerator demonstrates how one can leverage DataRobot's Python client to extract predictions, compute custom metrics, and sort their DataRobot models accordingly. |
Tune blueprints for preprocessing and model hyperparameters | Learn how to access, understand, and tune blueprints for both preprocessing and model hyperparameters. |
Fine-tune models with Eureqa | Apply symbolic regression to your dataset in the form of the Eureqa algorithm. |
Migrate a model to a new cluster | Download a deployed model from DataRobot cluster X, upload it to DataRobot cluster Y, and then deploy and make requests from it. |
Feature Reduction with FIRE | Learn about the benefits of Feature Importance Rank Ensembling (FIRE)—a method of advanced feature selection that uses a median rank aggregation of feature impacts across several models created during a run of Autopilot. |
Creating Custom Blueprints with Composable ML | Customize models on the Leaderboard using the Blueprint Workshop. |
Prepare and leverage image data with Databricks | Import image files using Spark and prepare them into a data frame suitable for ingest into DataRobot. |
Gather churn prediction insights with the Streamlit app | Use the Streamlit churn predictor app to present the drivers and predictions of your DataRobot model. |
Perform multi-model analysis | Use Python functions to aggregate DataRobot model insights into visualizations. |
Enrich data using the Hyperscaler API | Call the GCP API and enrich a modeling dataset that predicts customer churn. |
Predict lumber prices with Ready Signal and time series forecasts | Use Ready Signal to add external control data, such as census and weather data, to improve time series predictions. |
Build a model factory with Python multithreading | How to use the Python threading library to build a model factory. |
Predict flight delays starter use case | Designed for DataRobot trial users, experience an end-to-end DataRobot workflow using a use case that predicts flight delays. |
Perform statistical tests with DataRobot and Apache Airflow | Review an example workflow for carrying out statistical tests, notify stakeholders of any issues via Slack, and generate automated compliance documentation with the test results. |
Export model insights and visuals | Review examples for taking a DataRobot project and exporting its model insights as both machine-readable files and plots in various file formats |
Dimensionality reduction using t-SNE | Review examples for taking a DataRobot project and exporting its model insights, in various file formats, as both machine-readable files and plots. |
Create synthetic training data | Review an example workflow for carrying out statistical tests, notifying stakeholders of issues via Slack, and generating automated compliance documentation with the test results. |
View event logs | Change the output of the User Activity Monitor to drop an entire column of output or change the contents of a column in a way that preserves the anonymity of the column but maintains consistency for reporting. |
Run LIME with DataRobot models | Apply Local Interpretable Model-agnostic Explanations (LIME) to models built and deployed with DataRobot. |
Select robust features by permutation importance | This accelerator introduces an approach to select robust features, use multiple seeds for cross validation, add dummy features to compute the median permutation importance, and then select the most robust dummy features. |
Generate prediction intervals via conformal inference | Designed for DataRobot trial users, experience an end-to-end DataRobot workflow using a use case that predicts flight delays. |
Create partial dependence plots | Create one-way and two-way partial dependence plots (PDP), and Individual Conditional Expectations (ICE) insights using DataRobot. |
Reinforcement learning in DataRobot | Implement a model based on the Q-learning algorithm. |
GCP-based enrichment of modeling datasets | Demo the usage of the Google Cloud Natural Language API for sentiment analysis to enrich a customer churn dataset. |
Steel plate defect object detection | Train a highly accurate and robust machine learning model capable of detecting and classifying any-sized scratch present in steel plates. |
Hyperparameter optimization workflow | Build on the native DataRobot hyperparameter tuning by integrating the hyperopt module into DataRobot workflows. |
Updated June 24, 2024
Was this page helpful?
Great! Let us know what you found helpful.
What can we do to improve the content?
Thanks for your feedback!