深度学习2.0-41.GRU原理及实战

文章目录

1.GRU的原理
2.GRU实战
- 1.layer
- 2.cell

1.GRU的原理

2.GRU实战

1.layer

import osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
import tensorflow as tf
import numpy as np
from tensorflow import keras
from tensorflow.keras import layerstf.random.set_seed(22)
np.random.seed(22)
assert tf.__version__.startswith('2.')batchsz = 128# the most frequest words
total_words = 10000
max_review_len = 80
embedding_len = 100
(x_train, y_train), (x_test, y_test) = keras.datasets.imdb.load_data(num_words=total_words)
# x_train:[b, 80]
# x_test: [b, 80]
x_train = keras.preprocessing.sequence.pad_sequences(x_train, maxlen=max_review_len)
x_test = keras.preprocessing.sequence.pad_sequences(x_test, maxlen=max_review_len)db_train = tf.data.Dataset.from_tensor_slices((x_train, y_train))
db_train = db_train.shuffle(1000).batch(batchsz, drop_remainder=True)
db_test = tf.data.Dataset.from_tensor_slices((x_test, y_test))
db_test = db_test.batch(batchsz, drop_remainder=True)
print('x_train shape:', x_train.shape, tf.reduce_max(y_train), tf.reduce_min(y_train))
print('x_test shape:', x_test.shape)class MyRNN(keras.Model):def __init__(self, units):super(MyRNN, self).__init__()# transform text to embedding representation# [b, 80] => [b, 80, 100]self.embedding = layers.Embedding(total_words, embedding_len,input_length=max_review_len)# [b, 80, 100] , h_dim: 64self.rnn = keras.Sequential([# layers.SimpleRNN(units, dropout=0.5, return_sequences=True, unroll=True),# layers.SimpleRNN(units, dropout=0.5, unroll=True)# unroll: Boolean (default False). If True, the network will be unrolled,# else a symbolic loop will be used.# Unrolling can speed-up a RNN, although it tends to be more memory-intensive.# Unrolling is only suitable for short sequences.layers.GRU(units, dropout=0.5, return_sequences=True, unroll=False),layers.GRU(units, dropout=0.5, unroll=False)])# fc, [b, 80, 100] => [b, 64] => [b, 1]self.outlayer = layers.Dense(1)def call(self, inputs, training=None):"""net(x) net(x, training=True) :train modenet(x, training=False): test:param inputs: [b, 80]:param training::return:"""# [b, 80]x = inputs# embedding: [b, 80] => [b, 80, 100]x = self.embedding(x)# rnn cell compute# x: [b, 80, 100] => [b, 64]x = self.rnn(x)# out: [b, 64] => [b, 1]x = self.outlayer(x)# p(y is pos|x)prob = tf.sigmoid(x)return probdef main():units = 64epochs = 4import timet0 = time.time()model = MyRNN(units)model.compile(optimizer=keras.optimizers.Adam(0.001),loss=tf.losses.BinaryCrossentropy(),metrics=['accuracy'],experimental_run_tf_function=False)model.fit(db_train, epochs=epochs, validation_data=db_test)model.evaluate(db_test)t1 = time.time()# Unroll=True# LSTM: 69.3 secnods, 83%# GRU: 100 seconds, 83.4%# Unroll=False# LSTM：23.71， 81.24# GRU 23.05， 83.11# 观测结果与官方文档不一致print('total time cost:', t1 - t0)if __name__ == '__main__':main()

2.cell

import osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
import tensorflow as tf
import numpy as np
from tensorflow import keras
from tensorflow.keras import layerstf.random.set_seed(22)
np.random.seed(22)
assert tf.__version__.startswith('2.')batchsz = 128# the most frequest words
total_words = 10000
max_review_len = 80
embedding_len = 100
(x_train, y_train), (x_test, y_test) = keras.datasets.imdb.load_data(num_words=total_words)
# x_train:[b, 80]
# x_test: [b, 80]
x_train = keras.preprocessing.sequence.pad_sequences(x_train, maxlen=max_review_len)
x_test = keras.preprocessing.sequence.pad_sequences(x_test, maxlen=max_review_len)db_train = tf.data.Dataset.from_tensor_slices((x_train, y_train))
db_train = db_train.shuffle(1000).batch(batchsz, drop_remainder=True)
db_test = tf.data.Dataset.from_tensor_slices((x_test, y_test))
db_test = db_test.batch(batchsz, drop_remainder=True)
print('x_train shape:', x_train.shape, tf.reduce_max(y_train), tf.reduce_min(y_train))
print('x_test shape:', x_test.shape)class MyRNN(keras.Model):def __init__(self, units):super(MyRNN, self).__init__()# [b, 64]self.state0 = [tf.zeros([batchsz, units])]self.state1 = [tf.zeros([batchsz, units])]# transform text to embedding representation# [b, 80] => [b, 80, 100]self.embedding = layers.Embedding(total_words, embedding_len,input_length=max_review_len)# [b, 80, 100] , h_dim: 64# RNN: cell1 ,cell2, cell3# SimpleRNN# self.rnn_cell0 = layers.SimpleRNNCell(units, dropout=0.5)# self.rnn_cell1 = layers.SimpleRNNCell(units, dropout=0.5)self.rnn_cell0 = layers.GRUCell(units, dropout=0.5)self.rnn_cell1 = layers.GRUCell(units, dropout=0.5)# fc, [b, 80, 100] => [b, 64] => [b, 1]self.outlayer = layers.Dense(1)def call(self, inputs, training=None):"""net(x) net(x, training=True) :train modenet(x, training=False): test:param inputs: [b, 80]:param training::return:"""# [b, 80]x = inputs# embedding: [b, 80] => [b, 80, 100]x = self.embedding(x)# rnn cell compute# [b, 80, 100] => [b, 64]state0 = self.state0state1 = self.state1for word in tf.unstack(x, axis=1):  # word: [b, 100]# h1 = x*wxh+h0*whh# out0: [b, 64]out0, state0 = self.rnn_cell0(word, state0, training)# out1: [b, 64]out1, state1 = self.rnn_cell1(out0, state1, training)# out: [b, 64] => [b, 1]x = self.outlayer(out1)# p(y is pos|x)prob = tf.sigmoid(x)return probdef main():units = 64epochs = 4import timet0 = time.time()model = MyRNN(units)model.compile(optimizer=keras.optimizers.Adam(0.001),loss=tf.losses.BinaryCrossentropy(),metrics=['accuracy'],experimental_run_tf_function=False)model.fit(db_train, epochs=epochs, validation_data=db_test)model.evaluate(db_test)t1 = time.time()# LSTM: 64.3 seconds, 83.4%# GRU:  96.7s, 83.4%print('total time cost:', t1 - t0)if __name__ == '__main__':main()

深度学习2.0-41.GRU原理及实战相关推荐

深度学习 LSTM长短期记忆网络原理与Pytorch手写数字识别
深度学习 LSTM长短期记忆网络原理与Pytorch手写数字识别一.前言二.网络结构三.可解释性四.记忆主线五.遗忘门六.输入门七.输出门八.手写数字识别实战 8.1 引入依赖库 8. ...
深度学习时间序列预测：GRU算法构建单变量时间序列预测模型+代码实战
深度学习时间序列预测:GRU算法构建单变量时间序列预测模型+代码实战 GRU(Gate Recurrent Unit)是循环神经网络(Recurrent Neural Network, RNN)的一种 ...
深度学习时间序列预测：GRU算法构建多变量时间序列预测模型+代码实战
深度学习时间序列预测:GRU算法构建多变量时间序列预测模型+代码实战注意参考:深度学习多变量时间序列预测:GRU算法构建单变量时间序列预测模型+代码实战 GRU(Gate Recurrent Uni ...
深度学习笔记（41）候选区域
深度学习笔记(41) 候选区域 1. 无对象区域 2. 候选区域 3. Faster R-CNN 1. 无对象区域记得滑动窗法吧,使用训练过的分类器在这些窗口中全部运行一遍,然后运行一个检测器,看 ...
halcon 深度学习标注_HALCON深度学习工具0.4 早鸟版发布了
原标题:HALCON深度学习工具0.4 早鸟版发布了 HALOCN深度学习工具在整个深度学习过程中扮演着重要的作用,而且在将来将扮演更重要的辅助作用,大大加快深度学习的开发流程,目前发布版本工具的主要 ...
halcon显示坐标_HALCON深度学习工具0.4 早鸟版发布了
HALOCN深度学习工具在整个深度学习过程中扮演着重要的作用,而且在将来将扮演更重要的辅助作用,大大加快深度学习的开发流程,目前发布版本工具的主要作用是图像数据处理和目标检测和分类中的标注. 标注训练 ...
python 主语_前深度学习时代--FFM模型的原理与Python实现
基于上一篇分析中协同过滤.逻辑回归及FM的比较,可以得出这样一个结论: 主流模型迭代的关键在于增强模型表达能力,而增强方式的主要脉络为: 引入其它可用特征信息(CF->LR). 将现有特征进行组 ...
系统学习深度学习（四） --CNN原理，推导及实现源码分析
之前看机器学习中,多层感知器部分,提到可以在设计多层感知器时,对NN的结构设计优化,例如结构化设计和权重共享,当时还没了解深度学习,现在看到CNN,原来CNN就是这方面的一个代表.CNN由纽约大学的Y ...
深度学习-目标检测YOLOV3 YOLOv4原理详解
2.YoloV3核心基础内容 2.1 网络结构可视化 Yolov3是目标检测Yolo系列非常非常经典的算法,不过很多同学拿到Yolov3或者Yolov4的cfg文件时,并不知道如何直观的可视化查看网络 ...
亚马逊Alexa的深度学习与语音识别的核心技术原理
相关推荐随着语音巨头抢占语音入口,蓝牙设备智能化已成为下一波竞争的关键,而身处其中的无线蓝牙耳机正披着"智能... 发表于 2018-04-04 10:34 • 37次阅读训练数据来自于 ...

深度学习2.0-41.GRU原理及实战

文章目录

1.GRU的原理

2.GRU实战

1.layer

2.cell

深度学习2.0-41.GRU原理及实战相关推荐

最新文章

热门文章