-
Our "linear map" implementation of reverse mode yields bad code (#376) and also #58.
We assume "classic" reverse mode (CRM) would be better.
But CRM needs mutation (?)
So we need SUL:
- [x]…
-
Hi Patrick,
I'm trying to estimate different algorithm's tracking error. For that, I'm using two own databases; one with real faces (maybe you known [it](http://gi4e.unavarra.es/databases/hpdb/)) a…
-
Linearity seems ubiquitous enough that it would be useful to have special syntax for it.
```
A #-> B
-- as sugar for
(1 _ : A) -> B
```
In particular that allows you to easily make an existi…
-
Simeon and Keir's papers used a kernel that was a combination of a linear and a squared exponential (also known as radial basis function, or rbf):
2. Weight matrix (typically requires access to model object and varies with data …
-
Hi, I wonder why disabling PETSc in the in the demos/topology_optimisation example would lead to NaN problem.
The code to change here is
`fwd_pred = ad_wrapper(problem, linear=True, use_petsc=True…
-
There is now a working version of spatial neurons, with weave. Now it seems that the bottleneck in the Hodgkin-Huxley example is state updates. It reminds me that in Neuron, there is an intensive use …