Niko
@nikosarcevic.com
human person (she)
| cosmologist at Duke University
| member of @LSSTDESC @Roman
| did physics @XENONexperiment
| loves plots, coats, and The Wire
| opinions mine
| cosmologist at Duke University
| member of @LSSTDESC @Roman
| did physics @XENONexperiment
| loves plots, coats, and The Wire
| opinions mine
Gotcha
I’ve tried Jax in my use cases and it was not a magical solution unfortunately
In any case we are py machine precision diagnosis errors the whole thing.
AND we have implemented forecasting so you can do your fisher fisher bias AND DALI ❤️
Lemme know if you are interested in a sneak peek
I’ve tried Jax in my use cases and it was not a magical solution unfortunately
In any case we are py machine precision diagnosis errors the whole thing.
AND we have implemented forecasting so you can do your fisher fisher bias AND DALI ❤️
Lemme know if you are interested in a sneak peek
November 17, 2025 at 2:24 PM
Gotcha
I’ve tried Jax in my use cases and it was not a magical solution unfortunately
In any case we are py machine precision diagnosis errors the whole thing.
AND we have implemented forecasting so you can do your fisher fisher bias AND DALI ❤️
Lemme know if you are interested in a sneak peek
I’ve tried Jax in my use cases and it was not a magical solution unfortunately
In any case we are py machine precision diagnosis errors the whole thing.
AND we have implemented forecasting so you can do your fisher fisher bias AND DALI ❤️
Lemme know if you are interested in a sneak peek
ALSO
after the package version 1 is out we plan to make a use case paper. So if you wanna use it and give us a plot and a paragraph or two comparing different methods on your use case — pls let us know
after the package version 1 is out we plan to make a use case paper. So if you wanna use it and give us a plot and a paragraph or two comparing different methods on your use case — pls let us know
November 17, 2025 at 1:04 PM
ALSO
after the package version 1 is out we plan to make a use case paper. So if you wanna use it and give us a plot and a paragraph or two comparing different methods on your use case — pls let us know
after the package version 1 is out we plan to make a use case paper. So if you wanna use it and give us a plot and a paragraph or two comparing different methods on your use case — pls let us know
Not the finite differences!
My understanding is that JAX also suffers on evil functions (it’s not a magical solutions)
We’ve been working hard this year and are almost done with a package that implements fancy stable derivs
(Machine precision level)
We need people to use it 🤣
My understanding is that JAX also suffers on evil functions (it’s not a magical solutions)
We’ve been working hard this year and are almost done with a package that implements fancy stable derivs
(Machine precision level)
We need people to use it 🤣
November 17, 2025 at 1:03 PM
Not the finite differences!
My understanding is that JAX also suffers on evil functions (it’s not a magical solutions)
We’ve been working hard this year and are almost done with a package that implements fancy stable derivs
(Machine precision level)
We need people to use it 🤣
My understanding is that JAX also suffers on evil functions (it’s not a magical solutions)
We’ve been working hard this year and are almost done with a package that implements fancy stable derivs
(Machine precision level)
We need people to use it 🤣
He is a very good lecturer!
November 10, 2025 at 12:54 PM
He is a very good lecturer!
Nothing like starting your week with my rants on what’s not working in science and how I’m trying to fix it 🤣🤣🤣
Joking aside — thank you for the kind words 😍🤩
It is always a pleasure to catch up with you
Joking aside — thank you for the kind words 😍🤩
It is always a pleasure to catch up with you
November 10, 2025 at 11:42 AM
Nothing like starting your week with my rants on what’s not working in science and how I’m trying to fix it 🤣🤣🤣
Joking aside — thank you for the kind words 😍🤩
It is always a pleasure to catch up with you
Joking aside — thank you for the kind words 😍🤩
It is always a pleasure to catch up with you
It’s very moulain rouge, is it? 🤣🤣
November 5, 2025 at 3:16 AM
It’s very moulain rouge, is it? 🤣🤣
If I did not have a position secured I would be absolutely applying!
November 4, 2025 at 2:45 PM
If I did not have a position secured I would be absolutely applying!
Haahahah — they are amazing!
November 2, 2025 at 9:57 PM
Haahahah — they are amazing!