"female" is a tool/package that assist in extracting features from different kinds of datasets, as for a further machine learning treatment (sklearn, tensorflow, etc). As our initial goal, female would provide users some libraries to read and handle astronomical data. However for a long term planning, this library is not restricted to the astronomical field and could also be applied to any kind of data.
With the advent of the era of Big Data, machine learning defines the future researching style for every fields, including astronomy. As for the modern synoptic survey, data gathered from the telescope would increase with the exponential growth in time, especially with the coming of ZTF and LSST. The time domain astronomy sets the tone for the future astronomy.
We plan to test female with as many kinds as possible data. In the framework of female, we cluster data into: 1) 1d factor relation (1FR); 2) 2d factor relation (2FR); 3) multiple factor relation (MFR).
1DF problem is for 1 dependent variable and 1 independent variable. In terms of astronomical data, it could be a time series data/light curves, which is the relation between flux and time, or a spectrum, that defines the relation between flux and frequency, and so on.
And the same, 2DF problem is defined as for 2 dependent variable and 1 independent variable. For instance, the images, which is composed of a CCD counts matrix, spreading into a two dimension plane.
For the MDF, it's quite complicated and not common used in the daily life, and we promise to find such data and implement our female as well.
We summarize in general our feature extraction methods, that is mainly two types, namely, the recovery and reduction.
what is recovery and reduction....
1DF reduction, curve feature extraction, like what FATS did.
1DF recovery, full light curve fitting and sampling with a specific interval of time, interesting!
2DF reduction, extract topological, geometric, or photometric features, like what sextractor/sewpy did.
2DF recovery, image stamps, like what ps paper do.
10pm, Dec, 7, 2018
pip install female
2.1 git clone https://github.com/saberyoung/female.git
2.2 python setup.py install
to be done
More examples of how to use female are shown in female gallery
to be published