Raúl Gómez gombru
gombru
/ FocalLoss.py
Last active
November 13, 2022 12:43
Caffe Python layer Focal Loss implementation (From Focal Loss for Dense Object Detection, Facebook paper)
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import caffe | |
| import numpy as np | |
| import json | |
| class FocalLoss(caffe.Layer): | |
| """ | |
| Compute Focal Loss | |
| Inspired by https://arxiv.org/abs/1708.02002 | |
| Raul Gomez Bruballa | |
| https://gombru.github.io/ |
gombru
/ CustomSoftmaxLoss.py
Created
May 22, 2018 16:17
Caffe Python layer implementing Cross-Entropy with Softmax activation Loss to deal with multi-label classification, were labels can be input as real numbers
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import caffe | |
| import numpy as np | |
| class CustomSoftmaxLoss(caffe.Layer): | |
| """ | |
| Compute Cross Entropy loss with Softmax activations for multi-label classifications, accepting real numbers as labels | |
| Inspired by https://arxiv.org/abs/1805.00932 | |
| Raul Gomez Bruballa | |
| https://gombru.github.io/ | |
| """ |