-
Notifications
You must be signed in to change notification settings - Fork 5.9k
Closed
Description
The LaTeX equation of the operator HuberLoss isn't encapsulated by a pair of double-dollar $$:
Paddle/paddle/operators/huber_loss_op.cc
Lines 68 to 79 in c0d2ca5
| Huber loss is a loss function used in robust regression. We define X as the | |
| input value and Y as the target value. Huber loss can evaluate the fitness of | |
| X to Y. Different from MSE loss, Huber loss is more robust for outliers. The | |
| shape of X and Y are [batch_size, 1]. The equation is: | |
| L_{\delta}(y, f(x)) = | |
| \begin{cases} | |
| 0.5 * (y - f(x))^2, \quad |y - f(x)| \leq \delta \\ | |
| \delta * (|y - f(x)| - 0.5 * \delta), \quad otherwise | |
| \end{cases} | |
| )DOC"); |
Please add them, while taking the following code as an example:
Paddle/paddle/operators/lrn_op.cc
Lines 89 to 96 in d883547
| The original formula is: | |
| $$ | |
| Output(i, x, y) = Input(i, x, y) / \left( | |
| k + \alpha \sum\limits^{\min(C, c + n/2)}_{j = \max(0, c - n/2)} | |
| (Input(j, x, y))^2 | |
| \right)^{\beta} | |
| $$ |
Metadata
Metadata
Assignees
Labels
No labels