AtomicVar
AtomicVar
## Reason I'm a newbie and I encountered some issues while I'm following the `README`. The reason is the outdated API. ## What I see Some APIs are outdated, for...
**Describe the bug** This simple `matx::sum` test case failed to compile on the latest commit. ```cpp #include int main() { auto t1 = matx::make_tensor({32, 10}); auto t2 = matx::make_tensor({32}); matx::sum(t2,...
The module cannot work on Windows if the Windows is running in some languages, e.g. Chinese. The problem is that `netsh` outputs differently under different codepages. On Chinese Windows, the...
I tried to inspect the address of an variable `x` with: ```cpp int x = 42; &x ``` It showed: ``` @0x7ffc383a5a88 ``` But if I use `printf` to print...
Adding `override` makes it more explicit that we are implementing the interfaces of the `DataIter` class, which is good for learners who learn the design by reading source code.
### Describe your change: - Calculating the negative log likelihood (NLL) loss between true labels and predicted Gaussian distributions. - Reference: [GaussianNLLLoss](https://pytorch.org/docs/stable/generated/torch.nn.GaussianNLLLoss.html) * [x] Add an algorithm? * [ ]...
### Describe your change: [Sparse categorical cross entropy loss](https://www.tensorflow.org/api_docs/python/tf/keras/losses/SparseCategoricalCrossentropy) is used in cases where the true class labels are represented in a sparse matrix format (class indices, not one-hot encoded...
### Describe your change: - [Connectionist Temporal Classification (CTC)](https://en.wikipedia.org/wiki/Connectionist_temporal_classification) loss is used in speech recognition, handwriting recognition and other sequence problems. It's used to get around not knowing the alignment...
### Describe your change: Add the [Kullback-Leibler divergence](https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence) loss algorithm. It is a metric used to compare two data distributions. It has various practical use cases in data science, including...