Give your models a competitive edge with our Features API

Access an aggregation of real-time, real-world data with prebuilt ML features so you can can rapidly prove which types of events impact specific locations, improving your models accuracy and AI capabilities.

Sign upTalk to an expert
Our Process

Enable Rapid Model Training with Better than Synthetic Data

Rapidly prove events impact your business with our Features API which goes beyond Synthetic data limitations. Then unlock the how and why they will impact your business with our Events API.

  • STEP 1:

    Gain Relevancy with Beam

    Get greater relevancy and take the guess work out of what is impacting your demand with Beam. Quickly perform a demand analysis collectively or by individual location.

  • STEP 2:

    Train Models with Features API

    Rapidly and cost effectively prove events impact your locations of interest with our Features API built on live real-world data, which is better than synthetic data.

  • STEP 3:

    Gain Explainability with Events API

    Once you have proven the accuracy of events on your demand, you can now seamlessly bring in specific events for more granular explainability on demand patterns.
Book a demo
Forecast ready Features

A library of features at your disposal

For training your forecasting models, you often want to know the total impact of different types of events around a location. Think of Features API as a library of features. Maybe you want to know about all sports games happening within a specific radius around your location – we've got you covered. Or maybe you want to know the average event size happening around your store – we have that. Do you want to know how public holidays, school holidays or observances are impacting your location – it's all there.
Show me how
CurlPythonPython SDK
curl -X POST https://api.predicthq.com/v1/features/ \
  -H "Accept: application/json" \
  -H "Authorization: Bearer $ACCESS_TOKEN" \
  --data @<(cat <<EOF
{
  "active": {
    "gte": "2019-11-16",
    "lte": "2019-11-17"
  },
  "location": {
    "geo": {
      "lat": "37.78428",
      "lon": "-122.40075",
      "radius": "2.6mi"
    }
  },
  "phq_attendance_conferences": {
    "stats": [
      "min",
      "max"
    ]
  },
  "phq_attendance_sports": {
    "stats": ["count", "std_dev", "median"],
    "phq_rank": { 
      "gt": 50
    }    
  },
  "phq_attendance_concerts": true,
  "phq_rank_public_holidays": true
}
EOF
)
Beam API

Discover the most relevant Features for your model

Events vary in type, size, location, duration, and impact. Use the Beam API within your own environment to perform a demand analysis. The output from this demand analysis will be enable you to understand what types of events impact a specific location the most. This can then be used to request the event categories which have the greatest impact and relevance, to train and improve your models
Learn More in Docs

Create a new API Token

Create Token
Immediate updates

Get your forecast model updated live

We know getting updates to models live in production can be time consuming and expensive. Building features for your model often involves downloading large data volumes into a data lake and managing that data. Then you need to write code to aggregate and create features on top of the data lake. Instead of downloading all of the events data into a data lake use the Features API to get the features you need for your model. Reduce your feature engineering effort and get your models updated in a fraction of the time.
Try our Data Science Guides
Research and development

Accelerate your model R&D quickly and cost effectively

Incorporating intelligent event data into your models will increase forecast accuracy for many businesses. We want you to be able to see the value quickly. During research and development, Data Science teams can spend a lot of time understanding data, pulling down data, aggregating data, and building features. The Features API is designed to deliver event based features in an easy to use format to make it very easy to try out events data in your models.
Contact Us for Features API Only Pricing
Time to value

Get up and running quickly

Don't spend months building your own features and integrating them into your models. We have done all the heavy lifting for you so you can focus on extracting the value immediately.

Beam (Relevancy Engine)

Quickly understand how surrounding events impact your specific business.

Jupyter Notebooks

We've created Jupyter notebooks for quick integration of event categories into your models.

API Explorer

Start testing our features within our simple but powerful API explorer.

Customer Stories

Demand intelligence gets results

Our customers include leading companies across retail, transport, accommodation and financial service sectors. They use our intelligent event data for labor optimization, demand forecasting, dynamic pricing, supply chain optimization and more.
Explore use casesMore customer stories
Erica Quintana
Erica Quintana
Marketplace Operations Manager, 
Instacart
With PredictHQ the guesswork is gone, and we can clearly see where disruptions in delivery may arise, helping us deliver orders on time while cutting our manual research time by 50%.
Read Instacart Story

Over $7 Trillion of Demand Processed

We know which events will impact your business. Fact.

PredictHQ is the only platform in the world that uncovers the impact that real-world events have on business, at scale. Shaping the next generation of real-world AI/ML models, and unlocking Billions in potential upside.

  • $7.2 Trillion
    of demand data correlated to
  • 127 Million
    events via
  • 1.19 Billion
    predictions, across
  • 0
    countries, territories, and special areas, unlocking
  • $150 Billion
    in potential ROI
Schedule Demo Try For Free

Get Started

Contact our data science experts to find out the best solutions for your business. We'll get back to you within 1 business day.

Talk to an Expert

Trusted by

Other Questions?