-
Hi, yuanchen,
I am using SumOfSquares to reproduce the results of a paper, the problem is when I finish the program locally, the solution may be varying each time when I run it, Is this a bug of Su…
-
## __Description__
AirPLS (Adaptive Iteratively Reweighted Penalized Least Squares) and ArPLS (Asymmetrically Reweighted Penalized Least Squares) are powerful algorithms for removing complex non-li…
-
There are quite a few cases where try/catch blocks are used to control the computational flow. For example in the tabular backends (try loading - fall back to generate tables) and in the HelmholtzEOSM…
-
> Moreover, von Neumann’s Minmax Theorem (1928) ... implies that
This does not sound right. That theorem should be the first to prove the existence of the value (along with the optimal strategies o…
-
When running biggish systems (500 states, 2000 samples/state), it takes between ~11 and 20 GB or so (on OS X), but only about 4 minutes. Once it exceeds the memory of the computer, then it slows down …
-
Hi,
I want to solve a 2D poisson equation Ax=b (to solve for pressure for some fluid dynamics simulation), but I realized that even setting up the matrix is strangly incredibly slow. I'm trying to se…
-
### Is your feature request related to a problem? Please describe.
The task of finding the MLE for a three-parameter (generalized) Pareto distribution boils down to finding _all_ roots of the (univ…
-
At the moment only one dependent variable `b` is allowed to solve `A @ x = b` for `x`.
In other algorithms, multiple dependent variables could be passed, e.g.:
https://numpy.org/devdocs/reference/…
-
I have a use case for calculating ground station accesses for satellites, however, the ground stations I need to use employ off-zenith pointings. It would be helpful if `ground_station_accesses` and `…
-
### Describe your issue.
I found this problem with handling events in `scipy.integrate.solve_ivp`.
I am solving a system of two coupled ODEs with a termination event triggered one of the evolved v…