Output Formats¶
ROOT¶
Katydid has a variety of ROOT writers:
To write output from katydid processors with one of the writers, add a signal-slot connection from the processor to the root writer in the configuration file.
root-tree-writer¶
The root-tree-writer can be used to write processor output into root tree format. For example, discriminated-points, processed-tracks and multi-peak-events can all be written to the same root file using this writer.
basic-root-writer¶
The basic-root-writer can be used to write power spectra, gain variation, sparse spectrograms as well as output from event analysis.
HDF5¶
Since 2018, HDF5 is no longer supported. New data types can no longer be written in HDF5 format.
Katydid supports writing data to the HDF5 file format, which can be opened with various Python plotting tools. The data are structured according to their type, so that the HDF5 file will have the following (data dependent) structure:
- Spectra - stored in /spectra
- Frequency Spectra - stored in /spectra/frequency
~ Polar frequency spectra are called PolarFS_N where N is the number of the slice which the frequency spectrum was calculated over. The data are a 2MxN array, where N is equal to half the length of the original slice plus one, and M is equal to the number of components in the original time series data. In memory, to retrieve frequency bin i from component j, data[j][i] is the absolute value, and data[j+i][i] is the phase (argument).
Time Series - stored in /time_series
Candidates - stored in /candidates