Skip to content

Instantly share code, notes, and snippets.

View kaczmarj's full-sized avatar

Jakub Kaczmarzyk kaczmarj

102 followers · 30 following
View GitHub Profile
@wassname
wassname / jaccard_coef_loss.py
Last active January 30, 2024 15:45
jaccard_coef_loss for keras. This loss is usefull when you have unbalanced classes within a sample such as segmenting each pixel of an image. For example you are trying to predict if each pixel is cat, dog, or background. You may have 80% background, 10% dog, and 10% cat. Should a model that predicts 100% background be 80% right, or 30%? Categor…
from keras import backend as K
def jaccard_distance_loss(y_true, y_pred, smooth=100):
"""
Jaccard = (|X & Y|)/ (|X|+ |Y| - |X & Y|)
= sum(|A*B|)/(sum(|A|)+sum(|B|)-sum(|A*B|))
The jaccard distance loss is usefull for unbalanced datasets. This has been
shifted so it converges on 0 and is smoothed to avoid exploding or disapearing
gradient.
@mjdietzx
mjdietzx / residual_network.py
Last active March 26, 2024 06:33
Clean and simple Keras implementation of residual networks (ResNeXt and ResNet) accompanying accompanying Deep Residual Learning: https://blog.waya.ai/deep-residual-learning-9610bb62c355.
"""
Clean and simple Keras implementation of network architectures described in:
- (ResNet-50) [Deep Residual Learning for Image Recognition](https://arxiv.org/pdf/1512.03385.pdf).
- (ResNeXt-50 32x4d) [Aggregated Residual Transformations for Deep Neural Networks](https://arxiv.org/pdf/1611.05431.pdf).
Python 3.
"""
from keras import layers
from keras import models
@wassname
wassname / dice_loss_for_keras.py
Created September 26, 2016 08:32
dice_loss_for_keras
"""
Here is a dice loss for keras which is smoothed to approximate a linear (L1) loss.
It ranges from 1 to 0 (no error), and returns results similar to binary crossentropy
"""
# define custom loss and metric functions
from keras import backend as K
def dice_coef(y_true, y_pred, smooth=1):