Skip to content

Instantly share code, notes, and snippets.

View pinkeshbadjatiya's full-sized avatar

Pinkesh Badjatiya pinkeshbadjatiya

171 followers · 18 following
View GitHub Profile
@cbaziotis
cbaziotis / class_weights.py
Created January 23, 2017 11:55
def get_class_weights(y, smooth_factor=0):
"""
Returns the weights for each class based on the frequencies of the samples
:param smooth_factor: factor that smooths extremely uneven weights
:param y: list of true labels (the labels must be hashable)
:return: dictionary with the weight for each class
"""
counter = Counter(y)
if smooth_factor > 0:
@cbaziotis
cbaziotis / AttentionWithContext.py
Last active April 25, 2022 14:37
Keras Layer that implements an Attention mechanism, with a context/query vector, for temporal data. Supports Masking. Follows the work of Yang et al. [https://www.cs.cmu.edu/~diyiy/docs/naacl16.pdf] "Hierarchical Attention Networks for Document Classification"
def dot_product(x, kernel):
"""
Wrapper for dot product operation, in order to be compatible with both
Theano and Tensorflow
Args:
x (): input
kernel (): weights
Returns:
"""
if K.backend() == 'tensorflow':
@cbaziotis
cbaziotis / Attention.py
Last active October 22, 2024 08:31
Keras Layer that implements an Attention mechanism for temporal data. Supports Masking. Follows the work of Raffel et al. [https://arxiv.org/abs/1512.08756]
from keras import backend as K, initializers, regularizers, constraints
from keras.engine.topology import Layer
def dot_product(x, kernel):
"""
Wrapper for dot product operation, in order to be compatible with both
Theano and Tensorflow
Args:
@wassname
wassname / keras_attention_wrapper.py
Created November 1, 2016 08:06
A keras attention layer that wraps RNN layers.
"""
A keras attention layer that wraps RNN layers.
Based on tensorflows [attention_decoder](https://github.com/tensorflow/tensorflow/blob/c8a45a8e236776bed1d14fd71f3b6755bd63cc58/tensorflow/python/ops/seq2seq.py#L506)
and [Grammar as a Foreign Language](https://arxiv.org/abs/1412.7449).
date: 20161101
author: wassname
url: https://gist.github.com/wassname/5292f95000e409e239b9dc973295327a
"""