Joshuaalbert / DSA2000-Cal

DSA-2000 Calibration and Forward Modelling
https://www.deepsynoptic.org/overview
MIT License
1 stars 1 forks source link

Decide on how to handle flags when averaging #108

Closed Joshuaalbert closed 1 month ago

Joshuaalbert commented 1 month ago

There is an ambiguity when averaging flags

# any/all ambiguity with flags.
flags_cal = jnp.any(reshaped_vis_data.flags, axis=(0, 2), keepdims=True)[0]  # [num_row // num_time, 1, [, 4]]
Joshuaalbert commented 1 month ago

Choose the more conservative any.