-
I've come across an interesting segfault when compiling with optimisations inside a larger finite element program. The offending function:
```cpp
using sparse_matrix = Eigen::SparseMatrix;
using…
-
I want to add automatic parameter selection for λ
```julia
function cross_validate_spline(x, y, smoothing_parameter_grid, nfolds = length(X) -1)
function my_cumsum(x)
n = length(x)
…
-
The documentation says that the nearley parser includes the Joop Leo optimizations, but looking through the source for v2.15.1, I don't see anything that appears to implement the algorithm. I also cr…
-
When I tried `Linear` I noticed a performance/usability problem for my use case.
```csharp
public Image Linear(double[] a, double[] b, bool? uchar = null);
```
I only wanted to multiply/add ev…
-
This is a tracking issue for tests that are currently disabled, either with `// ignore-test` or by being placed in the `tests_old` directory. Please edit this issue if the list is changed and state th…
-
Rather self-explanatory, but taking the time to eliminate all runtime checks would be very beneficial to the proof itself.
-
Hi Samuel. I have run through the notebook modopt_example.ipynb, and inverse_problems_1.ipynb and sparsity_1.ipynb in the CosmoStat/Tutorials ada branch, playing around with parameters to better my u…
-
The `NumberWrapper` class opens up some interesting possibilities as to what one might be able to interpret the result of symbol operations as. Without looking under the hood, it would seem to me that…
-
Any thoughts/plans about YaRN support for the positional embeddings?
https://github.com/jquesnelle/yarn
I don't actually see them beat regular linear scaling w/ fine-tuning in the paper, but presu…
-
Hello Patrick,
Thanks a lot for all the working you are putting in here.
I have been working on a problem that I can solve with small scale neural network but need to use stochastic newton updat…