Fix for freezing exponents #1
Conversation
| x0, | ||
| args=(mol, basis_str, exp_array, use_Jacobian), | ||
| method="SLSQP", | ||
| method="Nelder-Mead", |
There was a problem hiding this comment.
Why is SLSQP changed to Nelder-Mead? Please always remember to "pull" the latest version of a code before pushing any commits!
| print(x0) | ||
| bnds = ((1e-9, None) for _ in range(exp_array.shape[0])) | ||
| def minimize_energy(mol, basis_str, exp_array=None, use_Jacobian=False): | ||
| x0 = exp_array[np.where(exp_array[:, 1] == 0), 0] |
There was a problem hiding this comment.
Good, so it is meant to check that the exponent's freeze/free flag is ==0 before sending it to the optimizer. I will check to see if it works soon!
|
Thank you @deyanmihaylov! |
|
I would suggest a timeline of having the code adapted to do this is end of this week. I am already working on this (but I haven't got anything ready yet). |
|
@deyanmihaylov Is there anything specific that caused this code to take so much longer than the original estimated timeline? In Discord you asked me about the meaning of the X variable and I provided you the paper and the paragraph that introduces X. |
|
@ndattani I apologise for the delay, I had an emergency which kept me away from my work for the past few days, today I am back to work. I will discuss on Discord and follow up accordingly. |
2 participants
No description provided.