Qinka Computing Subprogram is collection of functions used to do computing.
The code is hosted prime at Visual Studio Team Services's repo, while the GitHub's repo is also available.
Host | Platform | |
---|---|---|
VSTS CI | Windows | |
VSTS CI | Linux | |
VSTS CI | macOS | |
Travis-CI | Linux |
The Qinka Computing Subprogram includes the following functions:
- Basic math operations (vector-based)
- add
- sub
- product
- division
- scalar
- abs
- sign
- exp
- expm1
- log
- log1p
- sqrt
- power
- sin
- cos
- tan
- asin
- acos
- atan
- cosh
- sinh
- tanh
- asinh
- acosh
- atanh
- ceil
- floor
- Max
- Min
- Matrix operations
- Matrix multiplication
- Reduction operations
- Cumulative sum
- Argmax
- Argmin
- Cumulative product
- The log of summed exponentials
logsumexp
- Means
- The p-norm
- Error functions
- erf:
$f(x) = \frac{2}{\sqrt{\pi}} \int_0^x e^{-t^2}dt $ - erfc:
$f(x) = 1 - \frac{2}{\sqrt{\pi}} \int_0^x e^{-t^2}dt $
- erf:
- Activation function
- Sigmoid
- ReLU
- Classifier
- SVM
- SoftMax
- Data generate
- Specific matrix (I, Z etc)
- Random matrix or vector
- Fill matrix or vector
- Neural Network
- Convolution neural network layer
- Fully connected layer
- Max-pooling layer
- Gradient Descent
- Gradient descent
- Ordinary Least Squares