Keming Zhang

icon location "AI in Science" Postdoc Fellow at UC San Diego. Recently graduated from UC Berkeley working at the intersection of AI and Astronomy

Results 6 comments of Keming Zhang

New UVIS model and update code (cont. from #24)

UVIS model use layer norm, ACS model use batch norm Revise code to support both

Jupyter Notebook showing end-to-end training example

I'll include more examples in readthedocs regarding dataset construction. Since users wouldn't have access to the dataset in the example the Jupyter notebook couldn't be rerun in which case detailed...

Allow usage of generator-like objects instead of numpy arrays for training

This is currently being developed and expected to be released soon.

Allow usage of generator-like objects instead of numpy arrays for training

Implemented and release as part of v.0.2.0 in DatasetSim. For HST style dataset I will implement that when the generic HST model is made

Substantial Python loop overhead for VBBL

You're welcome. I believe this also applies to 2L1S point source when the SG12 solver in VBBL is used. Since the finite source method is specified in time intervals in...

Substantial Python loop overhead for VBBL

@rpoleski Apologies for the late reply. I don't have a particular way of doing this, but perhaps you could use Valario's way of making python wrappers: https://github.com/valboz/VBBinaryLensing/tree/master/VBBinaryLensing/lib. He has a...