arrayfire-ml
arrayfire-ml copied to clipboard
arrayfire
ArrayFire's Machine Learning Library.
The following functions and modules need to be implemented: - [ ] Basic Convolutions: No strides, No pooling - [ ] Strided Convolutions without pooling - [ ] Strided Convolutions...
Adds new indexing functions to address #43. Theses are needed to implement Losses ans SVM from #34 and #6 those changes are next in line. Operator overloading currently avoided in...
When trying to compile, [line 29 of the CMake file](https://github.com/arrayfire/arrayfire-ml/blob/master/CMakeLists.txt#L29) was giving me trouble. Changing the included item from `ArrayFire::af` to `af` solved the issue. I'm using CMake 3.10.2 See...
- [x] mean squared error - [x] mean absolute error - [x] binary cross entropy - [ ] negative log likelyhood
Specialization of `Network` class for: - [ ] logistic regression - [ ] softmax regression
RNN Models
Once we have an implementation of the Layer Class https://github.com/arrayfire/arrayfire_ml/issues/17 , the Optimizer class and the DataSet class we can go about creating RNN flavors. There are 3 models that...
### Base Classes - [x] [nn::Module](https://github.com/arrayfire/arrayfire-ml/pull/30) - [x] [autograd::Variable](https://github.com/arrayfire/arrayfire-ml/pull/30) - [x] [Solver](https://github.com/arrayfire/arrayfire-ml/issues/35) - [x] [nn::Loss](https://github.com/arrayfire/arrayfire-ml/issues/34) ### Autograd - [x] Broadcasting [(sum, tile)](https://github.com/arrayfire/arrayfire-ml/pull/30) - [x] [Math](https://github.com/arrayfire/arrayfire-ml/pull/30) (exp, pow, abs, sqrt, log)...
This still isn't quite perfect. The only thing left to do is to figure out how to make grad_func work with batched output.
