Skip to content

This is a TensorFlow implementation of the following paper: DropBlock: A regularization method for convolutional networks

License

Notifications You must be signed in to change notification settings

mbonyani/DropBlock-tf

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

DropBlock-tf

This is a TensorFlow implementation of the following paper: DropBlock: A regularization method for convolutional networks

arXiv. https://arxiv.org/abs/1810.12890

import numpy as np
import tensorflow as tf
from utils.dropblock import DropBlock2D

# only support `channels_last` data format
net = tf.keras.layers.Conv2D(256, (3, 3), padding='same',kernel_initializer=initializer)(net)
net = tf.keras.layers.BatchNormalization()(net)
net = tf.keras.layers.LeakyReLU(alpha=0.1)(net)
net = tf.keras.layers.MaxPooling2D((2, 2), padding='same')(net)
net = DropBlock2D(keep_prob=0.5, block_size=3)(net,training=True)

About

This is a TensorFlow implementation of the following paper: DropBlock: A regularization method for convolutional networks

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages