Predictive AI: Build, Operate, and Govern with the DataRobot AI Platform | Flight Delays Use Case
Complete all phases of building, operating and governing a Predictive AI solution following the starter Flight Delays Use Case. This starter use case showcases essential DataRobot capabilities, but is not comprehensive. Learn about the complete capabilities of the DataRobot AI Platform at https://www.datarobot.com/platform.
Watch the video as an observer to understand the Predictive AI workflow, or follow along with the data files provided. The workflow you see here is under 1 hour of machine time and about 10-15 minutes of human time. Everything you see here would reasonably take 4 to 8 weeks for a senior data scientist working with an ML engineer to complete. The DataRobot AI Platform offers you extreme acceleration of your AI use cases.
Resources:
Trial Access: https://www.datarobot.com/trial/
Training Data: https://s3.amazonaws.com/datarobot_public_datasets/ai_accelerators/flight_delays/flight_delays_training.csv
Scoring Data: https://s3.amazonaws.com/datarobot_public_datasets/ai_accelerators/flight_delays/flight_delays_scoring.csv
Python Notebook Tutorial: https://github.com/datarobot-community/ai-accelerators/blob/main/advanced_ml_and_api_approaches/Flight%20Delays%20-%20Starter%20Use%20Case%20for%20New%20DataRobot%20Users/Flight%20Delays%20-%20Starter%20Use%20Case%20for%20New%20DataRobot%20Users.ipynb
Contents:
In this video you'll see an end-to-end demonstration of how to use the DataRobot AI Platform for Predictive AI Solutions across the Build, Operate and Govern stages.
What you'll see in the demonstration is the wrangling process to generate the model training data. Then we'll use DataRobot's automation to create multiple models that we can compare side by side. Third we'll select one model and send it through the model registration process. Once registered we'll configure the deployment and deploy that model. Then we'll interact with that deployment by sending some new data through it as a scoring job and retrieving the future predictions.
The use case that I'll be using to show you the process is predicting flight delays. Please note that this is the same sample use case that is pre-loaded for free trial users.
Stay connected with DataRobot!
► Blog: https://blog.datarobot.com/
► Community: https://community.datarobot.com/
► Twitter: https://twitter.com/DataRobot/
► LinkedIn: https://www.linkedin.com/company/datarobot/
► Facebook: https://www.facebook.com/datarobotinc/
► Instagram: https://www.instagram.com/datarobot/
Complete all phases of building, operating and governing a Predictive AI solution following the starter Flight Delays Use Case. This starter use case showcases essential DataRobot capabilities, but is not comprehensive. Learn about the complete capabilities of the DataRobot AI Platform at https://www.datarobot.com/platform.
Watch the video as an observer to understand the Predictive AI workflow, or follow along with the data files provided. The workflow you see here is under 1 hour of machine time and about 10-15 minutes of human time. Everything you see here would reasonably take 4 to 8 weeks for a senior data scientist working with an ML engineer to complete. The DataRobot AI Platform offers you extreme acceleration of your AI use cases.
Resources:
Trial Access: https://www.datarobot.com/trial/
Training Data: https://s3.amazonaws.com/datarobot_public_datasets/ai_accelerators/flight_delays/flight_delays_training.csv
Scoring Data: https://s3.amazonaws.com/datarobot_public_datasets/ai_accelerators/flight_delays/flight_delays_scoring.csv
Python Notebook Tutorial: https://github.com/datarobot-community/ai-accelerators/blob/main/advanced_ml_and_api_approaches/Flight%20Delays%20-%20Starter%20Use%20Case%20for%20New%20DataRobot%20Users/Flight%20Delays%20-%20Starter%20Use%20Case%20for%20New%20DataRobot%20Users.ipynb
Contents:
In this video you’ll see an end-to-end demonstration of how to use the DataRobot AI Platform for Predictive AI Solutions across the Build, Operate and Govern stages.
What you’ll see in the demonstration is the wrangling process to generate the model training data. Then we’ll use DataRobot’s automation to create multiple models that we can compare side by side. Third we’ll select one model and send it through the model registration process. Once registered we’ll configure the deployment and deploy that model. Then we’ll interact with that deployment by sending some new data through it as a scoring job and retrieving the future predictions.
The use case that I’ll be using to show you the process is predicting flight delays. Please note that this is the same sample use case that is pre-loaded for free trial users.
Stay connected with DataRobot!
► Blog: https://blog.datarobot.com/
► Community: https://community.datarobot.com/
► Twitter: https://twitter.com/DataRobot/
► LinkedIn: https://www.linkedin.com/company/datarobot/
► Facebook: https://www.facebook.com/datarobotinc/
► Instagram: https://www.instagram.com/datarobot/
