Federico Vaggi

No apology necessary at all, and thank you for making this code all open. I was trying to understand the function signature of jax.jet a little bit better, and I...

Yes, oops. Just the doing (f(x+delta_x) - f(x)) / delta_x On Wed, Mar 5, 2014 at 7:32 PM, Steven G. Johnson [email protected]: > I assume you mean "finite differences", and...

Hi Christopher, I actually went ahead and developed a package to do this in pure python using sympy. It's very rough around the edges and definitely not fit for public...

Mauro: I might be mistaken but I think Sundials approximates the sensitivity equations/jacobian using a finite difference scheme? As far as I know, they don't have access to the symbolic...

If you use autodifferentiation directly, you don't really need to explicitly use sensitivity equations. The downside (and reason I didn't opt for that approach) is that it means that you...

http://orcid.org/0000-0001-8100-158X

Not very elegant, but this fixes it: ``` hist = ax.scatter_density(x, y) hist.set_extent(hist.get_extent()) ```

Hi @nikitos9000 thanks for looking into it. So, first of all, I did a bit more digging (this is with esm2_t36_3B_UR50D) ``` len_2 = 4 diffs = [] for len_2...

``` len_2 = 30 sequences = [ 'A' * 4, 'Y' * len_2 ] model_input = batch_converter([(None, seq) for seq in sequences])[2] model_input = model_input.to(device) # Here is the surprising...

I see a similar pattern with esm1b_t33_650M_UR50S: