ML/NN/RNN/单向rnn、双向rnn_embedding.py

# -*- coding: utf-8 -*-
"""单向RNN、双向RNN-embedding.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/18T6WUWX_fdG23ufTD6CkelZnQVYnaU3w
"""

import matplotlib as mpl
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import tensorflow as tf
from tensorflow import keras
import sklearn 
import os
import sys
import time

print(tf.__version__)
print(sys.version_info)
for module in mpl,np,pd,sklearn,tf,keras:
  print(module.__name__,module.__version__)

imdb=keras.datasets.imdb
vocab_size=10000
index_from=3
(train_data,train_labels),(test_data,test_labels)=imdb.load_data(num_words=vocab_size,index_from=index_from)

word_index=imdb.get_word_index()
print(len(word_index))

word_index={k:(v+3) for k,v in word_index.items()}

word_index['<PAD>']=0
word_index['<START>']=1
word_index['<UNK>']=2
word_index['<END>']=3

reverse_word_index=dict([
    (value,key) for key,value in word_index.items()
])

def decode_review(text_ids):
    return ' '.join([reverse_word_index.get(word_id,'<UNK>') for word_id in text_ids])

decode_review(train_data[0])

max_length=500

train_data=keras.preprocessing.sequence.pad_sequences(
    train_data,value=word_index['<PAD>'],
    padding='post',maxlen=max_length
)

test_data=keras.preprocessing.sequence.pad_sequences(
    test_data,value=word_index['<PAD>'],
    padding='post',maxlen=max_length
)

print(train_data[0])

embedding_dim=16
batch_size=512
# 把DNN的全局平均换成单向RNN，时间变长，不断修改效果变好
# return_sequences:Boolean. Whether to return the last output in the output sequence, or the full sequence 文本生成、机器翻译是要返回所有序列的True,只要最后一个序列False
single_rnn_model=keras.models.Sequential([
          keras.layers.Embedding(vocab_size,embedding_dim,input_length=max_length),
          keras.layers.SimpleRNN(units=64,return_sequences=False),
          # w=64,b=64
          keras.layers.Dense(64,activation='relu'),
          keras.layers.Dense(1,activation='sigmoid'),
])

single_rnn_model.summary()
single_rnn_model.compile(optimizer='adam',loss='binary_crossentropy',metrics=['accuracy'])

# 全连接层参数是4160 wx+b:x是一维的64

single_rnn_model.variables

history_single_rnn=single_rnn_model.fit(
    train_data,train_labels,
    epochs=30,
    batch_size=batch_size,
    validation_split=0.2
)

def plot_learning_curves(history,label,epochs,min_value,max_value):
  data={}
  data[label]=history.history[label]
  data['val_'+label]=history.history['val_'+label]
  pd.DataFrame(data).plot(figsize=(8,5))
  plt.grid(True)
  plt.axis([0,epochs,min_value,max_value])
  plt.show()


# 训练集、验证集上的准确率
plot_learning_curves(history_single_rnn,'accuracy',30,0,1)
# 训练集、验证集上的损失
plot_learning_curves(history_single_rnn,'loss',30,0,1)

single_rnn_model.evaluate(
    test_data,test_labels,
    batch_size=batch_size,
    verbose=0
)

"""损失接近70%，准确率是50%—单向RNN没啥用"""

# !nvidia-smi

# 改成双向RNN

embedding_dim=16
batch_size=512

model=keras.models.Sequential([
             keras.layers.Embedding(vocab_size,embedding_dim,input_length=max_length),
            #  增加数据，2层双向RNN
             keras.layers.Bidirectional(keras.layers.SimpleRNN(units=64,return_sequences=True)),
             keras.layers.Bidirectional(keras.layers.SimpleRNN(units=64,return_sequences=False)),
             keras.layers.Dense(64,activation='relu'),
             keras.layers.Dense(1,activation='sigmoid'),
             ])

model.summary()
model.compile(optimizer='adam',
              loss='binary_crossentropy',
              metrics=['accuracy'])

history=model.fit(train_data,train_labels,epochs=30,batch_size=batch_size,validation_split=0.2)

"""在训练集上准确率能达到100%，就足够说明模型强大了"""

plot_learning_curves(history,'accuracy',30,0,1)
plot_learning_curves(history,'loss',30,0,4)

"""过拟合了，可能是模型太复杂，改为单层的RNN"""

# 改成双向RNN

embedding_dim=16
batch_size=512

model=keras.models.Sequential([
             keras.layers.Embedding(vocab_size,embedding_dim,input_length=max_length),
            #  增加数据，2层双向RNN
             keras.layers.Bidirectional(keras.layers.SimpleRNN(units=64,return_sequences=True)),
            #  keras.layers.Bidirectional(keras.layers.SimpleRNN(units=64,return_sequences=False)),
             keras.layers.Dense(64,activation='relu'),
             keras.layers.Dense(1,activation='sigmoid'),
             ])

model.summary()
model.compile(optimizer='adam',
              loss='binary_crossentropy',
              metrics=['accuracy'])

history=model.fit(train_data,train_labels,epochs=30,batch_size=batch_size,validation_split=0.2)

plot_learning_curves(history,'accuracy',30,0,1)
plot_learning_curves(history,'loss',30,0,4)

model.evaluate(test_data,test_labels,batch_size=batch_size,verbose=0)

"""与单向RNN相比loss减少，accuracy上升，效果变好；但是仍然是过拟合的，可以看作模型强大"""