Metadata-Version: 2.1
Name: realdata
Version: 0.0.7
Summary: Get real univariate time series data easily for testing
Home-page: https://github.com/microprediction/realdata
Author: microprediction
Author-email: pcotton@intechinvestments.com
License: MIT
Description: # realdata
        
        Tired of using fake data for testing time series, or historical data that you gradually overfit to? This returns "medium" length (~2000) univariate time series from real world instrumented processes. The data is live, so this never gets stale. 
        
        ## Install
        
            pip install realdata
            
        ### Usage 
        
            from realdata import get_values
            values = get_values()
            
        ### Too bare bones for you? 
        
        See [other ways](https://www.microprediction.org/features.html) to grab the data directly, or see this short [tutorial](https://www.microprediction.com/python-3) on retrieving historical data using the microprediction package.  
        
        ### What's the data?
        
        All sorts of things: traffic, electricity, emoji usage, hospital wait times, altitude of UFOs, security wait times at airports, changes in crypto-currencies, ... you get the idea. 
        
        ![](https://i.imgur.com/LmrmLQF.png)
        
        Streams are live so the next time you get data, it will be different. 
        
        ### Can I add to the collection?
        
        Yes see the [tutorial](https://www.microprediction.com/python-4) on making a new stream of data. 
        
Platform: UNKNOWN
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.7
Description-Content-Type: text/markdown
