Module treez::autograd [−][src]
Structs
| Context |
implementation of reverse automatic differentiation |
| Link | |
| OpAdd |
y = a + b; dy/da = 1; dy/db = 1; |
| OpCos |
y = cos(x); dy/dx = -sin(x) |
| OpExponential |
y = a^x; dy/dx = ln(a) * a^x |
| OpLog |
y = log_base(x); dy/dx = 1/(x*ln(base)) |
| OpMul |
y = a*b; dy/da = b; dy/db = a |
| OpRelu |
y = max(0,x); dy/dx = { 1 | x > 0, 0 | x < 0, 0 | x == 0 } |
| OpReluLeaky |
y = max(0.01x,x); dy/dx = { 1 | x > 0, 0.01x | x <= 0 } |
| OpSigmoid |
y = 1/(1+e^(-x)); dy/dx = y * ( 1 - y ) |
| OpSin |
y = sin(x); dy/dx = cos(x) |
| OpTan |
y = tan(x); dy/dx = 1/(cos(x))^2 |
| OpTanh |
y = sinh(x)/cosh(x) = (1-e^(-2x))/(1+e^(-2x)); dy/dx = 1 - tanh(x)^2 |
Enums
| OpType |
Traits
| Op |
forward Op and gradient interface |