Publications using astroNN

1. Deep learning of multi-element abundances from high-resolution spectroscopic data

   Henry W. Leung, Jo Bovy (2019)

   MNRAS, 483, 3255L

   Original astroNN paper
2. Dynamical heating across the Milky Way disc using APOGEE and Gaia

   J. Ted Mackereth, Jo Bovy, Henry W. Leung, et al. (2019)

   MNRAS, 489, 176M

   Use ApogeeBCNN to infer spectroscopic age
3. Simultaneous calibration of spectro-photometric distances and the Gaia DR2 parallax zero-point offset with deep learning

   Henry W. Leung, Jo Bovy (2019)

   MNRAS, 489, 2079L

   Use ApogeeDR14GaiaDR2BCNN to infer spectro-photometric distances
4. Solar image denoising with convolutional neural networks

   C. J. Díaz Baso, J. de la Cruz Rodríguez, S. Danilovic (2019)

   A&A, 629, 99D
5. A variational encoder-decoder approach to precise spectroscopic age estimation for large Galactic surveys

   Henry W. Leung, Jo Bovy, J. Ted Mackereth, Andrea Miglio (2023)

   arXiv:2302.05479

   Use ApokascEncoderDecoder to infer spectroscopic age trained on APOGEE and Kepler

Publication figure style

astroNN contains a function that helps me to standardize maplotlib figure style used in my publication.

astroNN.shared.matplotlib.pylab_style(paper=False)

Such function can be used by simply calling it before using matplotlib to plot any figure

import matplotlib.pylab as plt
from astroNN.shared import pylab_style

pylab_style(paper=True)

# matplotlib code goes here

If you do not have \(\LaTeX\) installed on your computer, you can set the paper option to False like pylab_style(paper=False)

Here are a figure that compare different styles using the following matplotlib code

plt.figure(figsize=(5, 5))
plt.plot([0, 1], [0, 1], label="Test")
plt.xlabel("x")
plt.ylabel("y")
plt.legend()