blaze_tensor
blaze_tensor copied to clipboard
3D Tensors for Blaze (https://bitbucket.org/blaze-lib/blaze)
This might be interesting to look at in order to generate more complicated tensor expressions: http://tensor-compiler.org/
In order to implement `reshape` and `flatten` a `ravel` expression is useful. An implementation of `bias_add` uses extracting the quatern value and then add it to the original array, so...
Here is an incomplete list of things that have to be implemented: - 3D Tensor operations - [x] tensor transposition/rotation - [ ] tensor/vector multiplication - [ ] tensor/matrix multiplication...
Using Blaze, the result of a matrix * vector is the same as their dot product: ```c++ blaze::DynamicVector a{1, 1, 0}; blaze::DynamicMatrix b{{1, 2, 3}, {-1, -2, -3}}; auto ans...
Blaze supports the softmax operation and its rowwise and columnwise overloads ([ref](https://bitbucket.org/blaze-lib/blaze/issues/218/provide-a-softmax-overload-with-support)). I like to request for the blaze_tensor to support this operation for tensors. As an example: ```C++ #include...