Step 2.3: Load time series raw data

Now that you are used to play with the stack and know the available series, let’s manipulate some data!

FETCH

Now that we have identified the right star, let’s get some data using FETCH! It is a function similar to find, with more arguments that gives the boundaries of a window to get raw datapoints. The UTC start date is 2009-05-02T00:56:10.000000Z and the end date is 2013-05-11T12:02:06.000000Z. Remember that the regular expression ~.* allow you to request all the series available in the platform. Limit it to only the sap.flux ones. In order to keep a restreint amount of data, update the labels sector to load only the data from Kepler-11, i.e. data with the label KEPLERID=6541920:

You will see a list containing a single time series, click on it to view it. As an exercise, you can re-use the find and create a regular expresssion to load a small amount of the stars data. Loading the complete set of data will probably crash your navigator.

To be continued

The full subset of data of Kepler-11 is now on top of a WarpScript stack. It includes around 65000 points. In the next steps, we will start to write a WarpScript data-flow to detect Kepler-11 exoplanet.