使用CNN进行文本分类
卷积神经网络
英文邮件分类
语料
simplistic , silly and tedious .
it's so laddish and juvenile , only teenage boys could possibly find it funny .
exploitative and largely devoid of the depth or sophistication that would make watching such a graphic treatment of the crimes bearable .
[garbus] discards the potential for pathological study , exhuming instead , the skewed melodrama of the circumstantial situation .
a visually flashy but narratively opaque and emotionally vapid exercise in style and mystification .
the story is also as unoriginal as they come , already having been recycled more times than i'd care to count .
about the only thing to give the movie points for is bravado -- to take an entirely stale concept and push it through the audience's meat grinder one more time .
not so much farcical as sour .
unfortunately the story and the actors are served with a hack script .
all the more disquieting for its relatively gore-free allusions to the serial murders , but it falls down in its attempts to humanize its subject .
a sentimental mess that never rings true .
while the performances are often engaging , this loose collection of largely improvised numbers would probably have worked better as a one-hour tv documentary .
interesting , but not compelling .
on a cutting room floor somewhere lies . . . footage that might have made no such thing a trenchant , ironic cultural satire instead of a frustrating misfire .
while the ensemble player who gained notice in guy ritchie's lock , stock and two smoking barrels and snatch has the bod , he's unlikely to become a household name on the basis of his first starring vehicle .
there is a difference between movies with the courage to go over the top and movies that don't care about being stupid
nothing here seems as funny as it did in analyze this , not even joe viterelli as de niro's right-hand goombah .
such master screenwriting comes courtesy of john pogue , the yale grad who previously gave us " the skulls " and last year's " rollerball . " enough said , except : film overboard !
here , common sense flies out the window , along with the hail of bullets , none of which ever seem to hit sascha .
this 100-minute movie only has about 25 minutes of decent material .
the execution is so pedestrian that the most positive comment we can make is that rob schneider actually turns in a pretty convincing performance as a prissy teenage girl .
on its own , it's not very interesting . as a remake , it's a pale imitation .
it shows that some studios firmly believe that people have lost the ability to think and will forgive any shoddy product as long as there's a little girl-on-girl action .
a farce of a parody of a comedy of a premise , it isn't a comparison to reality so much as it is a commentary about our knowledge of films .
as exciting as all this exoticism might sound to the typical pax viewer , the rest of us will be lulled into a coma .
the party scenes deliver some tawdry kicks . the rest of the film . . . is dudsville .
our culture is headed down the toilet with the ferocity of a frozen burrito after an all-night tequila bender — and i know this because i've seen 'jackass : the movie . '
the criticism never rises above easy , cynical potshots at morally bankrupt characters . . .
the movie's something-borrowed construction feels less the product of loving , well integrated homage and more like a mere excuse for the wan , thinly sketched story . killing time , that's all that's going on here .
stupid , infantile , redundant , sloppy , over-the-top , and amateurish . yep , it's " waking up in reno . " go back to sleep .
somewhere in the middle , the film compels , as demme experiments he harvests a few movie moment gems , but the field of roughage dominates .
the action clichés just pile up .
、、、、、
data_helpers.py
import numpy as np
import re
import itertools
from collections import Counter
def clean_str(string):
"""
Tokenization/string cleaning for all datasets except for SST.
Original taken from https://github.com/yoonkim/CNN_sentence/blob/master/process_data.py
"""
string = re.sub(r"[^A-Za-z0-9(),!?\'\`]", " ", string)
string = re.sub(r"\'s", " \'s", string)
string = re.sub(r"\'ve", " \'ve", string)
string = re.sub(r"n\'t", " n\'t", string)
string = re.sub(r"\'re", " \'re", string)
string = re.sub(r"\'d", " \'d", string)
string = re.sub(r"\'ll", " \'ll", string)
string = re.sub(r",", " , ", string)
string = re.sub(r"!", " ! ", string)
string = re.sub(r"\(", " \( ", string)
string = re.sub(r"\)", " \) ", string)
string = re.sub(r"\?", " \? ", string)
string = re.sub(r"\s{2,}", " ", string)
return string.strip().lower()
def load_data_and_labels(positive_data_file, negative_data_file):
"""
Loads MR polarity data from files, splits the data into words and generates labels.
Returns split sentences and labels.
"""
# Load data from files
positive = open(positive_data_file, "rb").read().decode('utf-8')
negative = open(negative_data_file, "rb").read().decode('utf-8')
positive_examples = positive.split('\n')[:-1]
negative_examples = negative.split('\n')[:-1]
positive_examples = [s.strip() for s in positive_examples]
negative_examples = [s.strip() for s in negative_examples]
#positive_examples = list(open(positive_data_file, "rb").read().decode('utf-8'))
#positive_examples = [s.strip() for s in positive_examples]
#negative_examples = list(open(negative_data_file, "rb").read().decode('utf-8'))
#negative_examples = [s.strip() for s in negative_examples]
# Split by words
x_text = positive_examples + negative_examples
x_text = [clean_str(sent) for sent in x_text]
# Generate labels
positive_labels = [[0, 1] for _ in positive_examples]
negative_labels = [[1, 0] for _ in negative_examples]
y = np.concatenate([positive_labels, negative_labels], 0)
return [x_text, y]
def batch_iter(data, batch_size, num_epochs, shuffle=True):
"""
Generates a batch iterator for a dataset.
"""
data = np.array(data)
data_size = len(data)
num_batches_per_epoch = int((len(data)-1)/batch_size) + 1
for epoch in range(num_epochs):
# Shuffle the data at each epoch
if shuffle:
shuffle_indices = np.random.permutation(np.arange(data_size))
shuffled_data = data[shuffle_indices]
else:
shuffled_data = data
for batch_num in range(num_batches_per_epoch):
start_index = batch_num * batch_size
end_index = min((batch_num + 1) * batch_size, data_size)
yield shuffled_data[start_index:end_index]
text_cnn.py
import tensorflow as tf
import numpy as np
class TextCNN(object):
"""
A CNN for text classification.
Uses an embedding layer, followed by a convolutional, max-pooling and softmax layer.
"""
def __init__(
self, sequence_length, num_classes, vocab_size,
embedding_size, filter_sizes, num_filters, l2_reg_lambda=0.0):
# Placeholders for input, output and dropout
self.input_x = tf.placeholder(tf.int32, [None, sequence_length], name="input_x")
self.input_y = tf.placeholder(tf.float32, [None, num_classes], name="input_y")
self.dropout_keep_prob = tf.placeholder(tf.float32, name="dropout_keep_prob")
# Keeping track of l2 regularization loss (optional)
l2_loss = tf.constant(0.0)
# Embedding layer
with tf.device('/cpu:0'), tf.name_scope("embedding"):
self.W = tf.Variable(
tf.random_uniform([vocab_size, embedding_size], -1.0, 1.0),
name="W")
self.embedded_chars = tf.nn.embedding_lookup(self.W, self.input_x)
self.embedded_chars_expanded = tf.expand_dims(self.embedded_chars, -1)
# Create a convolution + maxpool layer for each filter size
pooled_outputs = []
for i, filter_size in enumerate(filter_sizes):
with tf.name_scope("conv-maxpool-%s" % filter_size):
# Convolution Layer
filter_shape = [filter_size, embedding_size, 1, num_filters]
W = tf.Variable(tf.truncated_normal(filter_shape, stddev=0.1), name="W")
b = tf.Variable(tf.constant(0.1, shape=[num_filters]), name="b")
conv = tf.nn.conv2d(
self.embedded_chars_expanded,
W,
strides=[1, 1, 1, 1],
padding="VALID",
name="conv")
# Apply nonlinearity
h = tf.nn.relu(tf.nn.bias_add(conv, b), name="relu")
# Maxpooling over the outputs
pooled = tf.nn.max_pool(
h,
ksize=[1, sequence_length - filter_size + 1, 1, 1],
strides=[1, 1, 1, 1],
padding='VALID',
name="pool")
pooled_outputs.append(pooled)
# Combine all the pooled features
num_filters_total = num_filters * len(filter_sizes)
#self.h_pool = tf.concat(pooled_outputs, 3)
self.h_pool = tf.concat(3, pooled_outputs)
self.h_pool_flat = tf.reshape(self.h_pool, [-1, num_filters_total])
# Add dropout
with tf.name_scope("dropout"):
self.h_drop = tf.nn.dropout(self.h_pool_flat, self.dropout_keep_prob)
# Final (unnormalized) scores and predictions
with tf.name_scope("output"):
W = tf.get_variable(
"W",
shape=[num_filters_total, num_classes],
initializer=tf.contrib.layers.xavier_initializer())
b = tf.Variable(tf.constant(0.1, shape=[num_classes]), name="b")
l2_loss += tf.nn.l2_loss(W)
l2_loss += tf.nn.l2_loss(b)
self.scores = tf.nn.xw_plus_b(self.h_drop, W, b, name="scores")
self.predictions = tf.argmax(self.scores, 1, name="predictions")
# CalculateMean cross-entropy loss
with tf.name_scope("loss"):
losses = tf.nn.softmax_cross_entropy_with_logits(logits=self.scores, labels=self.input_y)
self.loss = tf.reduce_mean(losses) + l2_reg_lambda * l2_loss
# Accuracy
with tf.name_scope("accuracy"):
correct_predictions = tf.equal(self.predictions, tf.argmax(self.input_y, 1))
self.accuracy = tf.reduce_mean(tf.cast(correct_predictions, "float"), name="accuracy")
train.py
#! /usr/bin/env python
import tensorflow as tf
import numpy as np
import os
import time
import datetime
import data_helpers
from text_cnn import TextCNN
from tensorflow.contrib import learn
# Parameters
# ==================================================
# Data loading params
tf.flags.DEFINE_float("dev_sample_percentage", .1, "Percentage of the training data to use for validation")
tf.flags.DEFINE_string("positive_data_file", "./data/rt-polaritydata/rt-polarity.pos", "Data source for the positive data.")
tf.flags.DEFINE_string("negative_data_file", "./data/rt-polaritydata/rt-polarity.neg", "Data source for the negative data.")
# Model Hyperparameters
tf.flags.DEFINE_integer("embedding_dim", 128, "Dimensionality of character embedding (default: 128)")
tf.flags.DEFINE_string("filter_sizes", "3,4,5", "Comma-separated filter sizes (default: '3,4,5')")
tf.flags.DEFINE_integer("num_filters", 128, "Number of filters per filter size (default: 128)")
tf.flags.DEFINE_float("dropout_keep_prob", 0.5, "Dropout keep probability (default: 0.5)")
tf.flags.DEFINE_float("l2_reg_lambda", 0.0, "L2 regularization lambda (default: 0.0)")
# Training parameters
tf.flags.DEFINE_integer("batch_size", 64, "Batch Size (default: 64)")
tf.flags.DEFINE_integer("num_epochs", 200, "Number of training epochs (default: 200)")
tf.flags.DEFINE_integer("evaluate_every", 100, "Evaluate model on dev set after this many steps (default: 100)")
tf.flags.DEFINE_integer("checkpoint_every", 100, "Save model after this many steps (default: 100)")
tf.flags.DEFINE_integer("num_checkpoints", 5, "Number of checkpoints to store (default: 5)")
# Misc Parameters
tf.flags.DEFINE_boolean("allow_soft_placement", True, "Allow device soft device placement")
tf.flags.DEFINE_boolean("log_device_placement", False, "Log placement of ops on devices")
FLAGS = tf.flags.FLAGS
FLAGS._parse_flags()
print("\nParameters:")
for attr, value in sorted(FLAGS.__flags.items()):
print("{}={}".format(attr.upper(), value))
print("")
# Data Preparation
# ==================================================
# Load data
print("Loading data...")
x_text, y = data_helpers.load_data_and_labels(FLAGS.positive_data_file, FLAGS.negative_data_file)
# Build vocabulary
max_document_length = max([len(x.split(" ")) for x in x_text])
vocab_processor = learn.preprocessing.VocabularyProcessor(max_document_length)
x = np.array(list(vocab_processor.fit_transform(x_text)))
# Randomly shuffle data
np.random.seed(10)
shuffle_indices = np.random.permutation(np.arange(len(y)))
x_shuffled = x[shuffle_indices]
y_shuffled = y[shuffle_indices]
# Split train/test set
# TODO: This is very crude, should use cross-validation
dev_sample_index = -1 * int(FLAGS.dev_sample_percentage * float(len(y)))
x_train, x_dev = x_shuffled[:dev_sample_index], x_shuffled[dev_sample_index:]
y_train, y_dev = y_shuffled[:dev_sample_index], y_shuffled[dev_sample_index:]
print("Vocabulary Size: {:d}".format(len(vocab_processor.vocabulary_)))
print("Train/Dev split: {:d}/{:d}".format(len(y_train), len(y_dev)))
# Training
# ==================================================
with tf.Graph().as_default():
session_conf = tf.ConfigProto(
allow_soft_placement=FLAGS.allow_soft_placement,
log_device_placement=FLAGS.log_device_placement)
sess = tf.Session(config=session_conf)
with sess.as_default():
cnn = TextCNN(
sequence_length=x_train.shape[1],
num_classes=y_train.shape[1],
vocab_size=len(vocab_processor.vocabulary_),
embedding_size=FLAGS.embedding_dim,
filter_sizes=list(map(int, FLAGS.filter_sizes.split(","))),
num_filters=FLAGS.num_filters,
l2_reg_lambda=FLAGS.l2_reg_lambda)
# Define Training procedure
global_step = tf.Variable(0, name="global_step", trainable=False)
optimizer = tf.train.AdamOptimizer(1e-3)
grads_and_vars = optimizer.compute_gradients(cnn.loss)
train_op = optimizer.apply_gradients(grads_and_vars, global_step=global_step)
# Keep track of gradient values and sparsity (optional)
grad_summaries = []
for g, v in grads_and_vars:
if g is not None:
grad_hist_summary = tf.summary.histogram("{}/grad/hist".format(v.name), g)
sparsity_summary = tf.summary.scalar("{}/grad/sparsity".format(v.name), tf.nn.zero_fraction(g))
grad_summaries.append(grad_hist_summary)
grad_summaries.append(sparsity_summary)
grad_summaries_merged = tf.summary.merge(grad_summaries)
# Output directory for models and summaries
timestamp = str(int(time.time()))
out_dir = os.path.abspath(os.path.join(os.path.curdir, "runs", timestamp))
print("Writing to {}\n".format(out_dir))
# Summaries for loss and accuracy
loss_summary = tf.summary.scalar("loss", cnn.loss)
acc_summary = tf.summary.scalar("accuracy", cnn.accuracy)
# Train Summaries
train_summary_op = tf.summary.merge([loss_summary, acc_summary, grad_summaries_merged])
train_summary_dir = os.path.join(out_dir, "summaries", "train")
train_summary_writer = tf.summary.FileWriter(train_summary_dir, sess.graph)
# Dev summaries
dev_summary_op = tf.summary.merge([loss_summary, acc_summary])
dev_summary_dir = os.path.join(out_dir, "summaries", "dev")
dev_summary_writer = tf.summary.FileWriter(dev_summary_dir, sess.graph)
# Checkpoint directory. Tensorflow assumes this directory already exists so we need to create it
checkpoint_dir = os.path.abspath(os.path.join(out_dir, "checkpoints"))
checkpoint_prefix = os.path.join(checkpoint_dir, "model")
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
saver = tf.train.Saver(tf.global_variables(), max_to_keep=FLAGS.num_checkpoints)
# Write vocabulary
vocab_processor.save(os.path.join(out_dir, "vocab"))
# Initialize all variables
sess.run(tf.global_variables_initializer())
def train_step(x_batch, y_batch):
"""
A single training step
"""
feed_dict = {
cnn.input_x: x_batch,
cnn.input_y: y_batch,
cnn.dropout_keep_prob: FLAGS.dropout_keep_prob
}
_, step, summaries, loss, accuracy = sess.run(
[train_op, global_step, train_summary_op, cnn.loss, cnn.accuracy],
feed_dict)
time_str = datetime.datetime.now().isoformat()
print("{}: step {}, loss {:g}, acc {:g}".format(time_str, step, loss, accuracy))
train_summary_writer.add_summary(summaries, step)
def dev_step(x_batch, y_batch, writer=None):
"""
Evaluates model on a dev set
"""
feed_dict = {
cnn.input_x: x_batch,
cnn.input_y: y_batch,
cnn.dropout_keep_prob: 1.0
}
step, summaries, loss, accuracy = sess.run(
[global_step, dev_summary_op, cnn.loss, cnn.accuracy],
feed_dict)
time_str = datetime.datetime.now().isoformat()
print("{}: step {}, loss {:g}, acc {:g}".format(time_str, step, loss, accuracy))
if writer:
writer.add_summary(summaries, step)
# Generate batches
batches = data_helpers.batch_iter(
list(zip(x_train, y_train)), FLAGS.batch_size, FLAGS.num_epochs)
# Training loop. For each batch...
for batch in batches:
x_batch, y_batch = zip(*batch)
train_step(x_batch, y_batch)
current_step = tf.train.global_step(sess, global_step)
if current_step % FLAGS.evaluate_every == 0:
print("\nEvaluation:")
dev_step(x_dev, y_dev, writer=dev_summary_writer)
print("")
if current_step % FLAGS.checkpoint_every == 0:
path = saver.save(sess, './', global_step=current_step)
print("Saved model checkpoint to {}\n".format(path))
eval.py
#! /usr/bin/env python
import tensorflow as tf
import numpy as np
import os
import time
import datetime
import data_helpers
from text_cnn import TextCNN
from tensorflow.contrib import learn
import csv
# Parameters
# ==================================================
# Data Parameters
tf.flags.DEFINE_string("positive_data_file", "./data/rt-polaritydata/rt-polarity.pos", "Data source for the positive data.")
tf.flags.DEFINE_string("negative_data_file", "./data/rt-polaritydata/rt-polarity.neg", "Data source for the positive data.")
# Eval Parameters
tf.flags.DEFINE_integer("batch_size", 64, "Batch Size (default: 64)")
tf.flags.DEFINE_string("checkpoint_dir", "./", "Checkpoint directory from training run")
tf.flags.DEFINE_boolean("eval_train", False, "Evaluate on all training data")
# Misc Parameters
tf.flags.DEFINE_boolean("allow_soft_placement", True, "Allow device soft device placement")
tf.flags.DEFINE_boolean("log_device_placement", False, "Log placement of ops on devices")
FLAGS = tf.flags.FLAGS
FLAGS._parse_flags()
print("\nParameters:")
for attr, value in sorted(FLAGS.__flags.items()):
print("{}={}".format(attr.upper(), value))
print("")
# CHANGE THIS: Load data. Load your own data here
if FLAGS.eval_train:
x_raw, y_test = data_helpers.load_data_and_labels(FLAGS.positive_data_file, FLAGS.negative_data_file)
y_test = np.argmax(y_test, axis=1)
else:
x_raw = ["a masterpiece four years in the making", "everything is off."]
y_test = [1, 0]
# Map data into vocabulary
vocab_path = os.path.join(FLAGS.checkpoint_dir, "..", "vocab")
vocab_processor = learn.preprocessing.VocabularyProcessor.restore(vocab_path)
x_test = np.array(list(vocab_processor.transform(x_raw)))
print("\nEvaluating...\n")
# Evaluation
# ==================================================
checkpoint_file = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
graph = tf.Graph()
with graph.as_default():
session_conf = tf.ConfigProto(
allow_soft_placement=FLAGS.allow_soft_placement,
log_device_placement=FLAGS.log_device_placement)
sess = tf.Session(config=session_conf)
with sess.as_default():
# Load the saved meta graph and restore variables
saver = tf.train.import_meta_graph("{}.meta".format(checkpoint_file))
saver.restore(sess, checkpoint_file)
# Get the placeholders from the graph by name
input_x = graph.get_operation_by_name("input_x").outputs[0]
# input_y = graph.get_operation_by_name("input_y").outputs[0]
dropout_keep_prob = graph.get_operation_by_name("dropout_keep_prob").outputs[0]
# Tensors we want to evaluate
predictions = graph.get_operation_by_name("output/predictions").outputs[0]
# Generate batches for one epoch
batches = data_helpers.batch_iter(list(x_test), FLAGS.batch_size, 1, shuffle=False)
# Collect the predictions here
all_predictions = []
for x_test_batch in batches:
batch_predictions = sess.run(predictions, {input_x: x_test_batch, dropout_keep_prob: 1.0})
all_predictions = np.concatenate([all_predictions, batch_predictions])
# Print accuracy if y_test is defined
if y_test is not None:
correct_predictions = float(sum(all_predictions == y_test))
print("Total number of test examples: {}".format(len(y_test)))
print("Accuracy: {:g}".format(correct_predictions/float(len(y_test))))
# Save the evaluation to a csv
predictions_human_readable = np.column_stack((np.array(x_raw), all_predictions))
out_path = os.path.join(FLAGS.checkpoint_dir, "..", "prediction.csv")
print("Saving evaluation to {0}".format(out_path))
with open(out_path, 'w') as f:
csv.writer(f).writerows(predictions_human_readable)
中文邮件分类
语料
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人：李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人：李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
网址：www.8wx8.com,电话：0551-3961292，QQ：415455699 2、网站制作（复制，你所见的网站，就是你的网上家园） 为您提供最可靠的域名,主机服务，网站建设,邮件群发，百度首页推广，发广告服务等…… 网址：www.hf263.net,电话：0551-3961292，QQ：415455699 ########价格全国最低#################### 客户就是上帝，我希望所有我接待的客户都成为我的朋友！
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人：李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人： 李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
贵公司经理 你好！ 本公司为广州市荣兴贸易有限公司，本着双方公司利益的出发点可向外代开 普通国（地）税发票，商品销售发票，运输发票，餐饮发票，广告发票，服务行业， 等等．．．．．．普通发票只收２％个税点． 如贵公司在这方面上有需要可来电咨询，欢迎诚意的合作． 联系人：唐娜 联系电话：020-33520954 联系手机：13533999436
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人：李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人：李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
您好， 邮件无法发送到您指定的地址：andrew@vip.sina.com 该邮箱内存储信件过多已超出所限制的使用空间，无法再接收新邮件。 请您通知该邮箱用户及时清理自己邮箱内的信件，以避免因接收不到新邮件而带来的不必要的损失。谢谢您的配合。 --- 附件中的内容是原信件的一份拷贝 The message contains Unicode characters and has been sent as a binary attachment.
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人： 李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
妇科囊肿不开刀,盆腔积液几瓶消. 新乡市花园妇幼门诊多年来研制的中 药对治疗妇科囊肿,盆腔炎,附件炎,输 卵管炎,输卵管阻塞不孕有特效. 本门诊开展多种妇科手术,技高,诚信, 优质,安全,无痛人流术更是技高一筹, {就象睡了几分钟}"妇炎二号"名声在外, 不少人千里买药,真是"谁用谁知道"..... 祥情请点击www.hyfymz.com 中文网址:花园妇幼门诊
妇科囊肿不开刀,盆腔积液几瓶消. 新乡市花园妇幼门诊多年来研制的中 药对治疗妇科囊肿,盆腔炎,附件炎,输 卵管炎,输卵管阻塞不孕有特效. 本门诊开展多种妇科手术,技高,诚信, 优质,安全,无痛人流术更是技高一筹, {就象睡了几分钟}"妇炎二号"名声在外, 不少人千里买药,真是"谁用谁知道"..... 祥情请点击www.hyfymz.com 中文网址:花园妇幼门诊
妇科囊肿不开刀,盆腔积液几瓶消. 新乡市花园妇幼门诊多年来研制的中 药对治疗妇科囊肿,盆腔炎,附件炎,输 卵管炎,输卵管阻塞不孕有特效. 本门诊开展多种妇科手术,技高,诚信, 优质,安全,无痛人流术更是技高一筹, {就象睡了几分钟}"妇炎二号"名声在外, 不少人千里买药,真是"谁用谁知道"..... 祥情请点击www.hyfymz.com 中文网址:花园妇幼门诊
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人：李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
妇科囊肿不开刀,盆腔积液几瓶消. 新乡市花园妇幼门诊多年来研制的中 药对治疗妇科囊肿,盆腔炎,附件炎,输 卵管炎,输卵管阻塞不孕有特效. 本门诊开展多种妇科手术,技高,诚信, 优质,安全,无痛人流术更是技高一筹, {就象睡了几分钟}"妇炎二号"名声在外, 不少人千里买药,真是"谁用谁知道"..... 祥情请点击www.hyfymz.com 中文网址:花园妇幼门诊
妇科囊肿不开刀,盆腔积液几瓶消. 新乡市花园妇幼门诊多年来研制的中 药对治疗妇科囊肿,盆腔炎,附件炎,输 卵管炎,输卵管阻塞不孕有特效. 本门诊开展多种妇科手术,技高,诚信, 优质,安全,无痛人流术更是技高一筹, {就象睡了几分钟}"妇炎二号"名声在外, 不少人千里买药,真是"谁用谁知道"..... 祥情请点击www.hyfymz.com 中文网址:花园妇幼门诊
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人： 李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
贵公司财务(经理); 本公司是<<深圳市海天实业有限公司>>,成立于深圳多年,有良好的贸易信誉. 长期为各大公司代开电脑发票和各种地税发票;(广告,运输.建筑安装.服务行业). 价格优惠.并以和各大公司长期合作为目的,承诺所开发票可到税务抵扣验证. 欢迎有意着来电咨询. 联系电话;0755-81535924 013926596824 联系人;张永健 深圳市海天实业有限公司
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人： 李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
贵公司财务(经理); 本公司是<<深圳市海天实业有限公司>>,成立于深圳多年,有良好的贸易信誉. 长期为各大公司代开电脑发票和各种地税发票;(广告,运输.建筑安装.服务行业). 价格优惠.并以和各大公司长期合作为目的,承诺所开发票可到税务抵扣验证. 欢迎有意着来电咨询. 联系电话;0755-81535924 013926596824 联系人;张永健 深圳市海天实业有限公司
贵公司经理 你好！ 本公司为广州市荣兴贸易有限公司，本着双方公司利益的出发点可向外代开 普通国（地）税发票，商品销售发票，运输发票，餐饮发票，广告发票，服务行业， 等等．．．．．．普通发票只收２％个税点． 如贵公司在这方面上有需要可来电咨询，欢迎诚意的合作． 联系人：唐娜 联系电话：020-33520954 联系手机：13533999436
贵公司经理 你好！ 本公司为广州市荣兴贸易有限公司，本着双方公司利益的出发点可向外代开 普通国（地）税发票，商品销售发票，运输发票，餐饮发票，广告发票，服务行业， 等等．．．．．．普通发票只收２％个税点． 如贵公司在这方面上有需要可来电咨询，欢迎诚意的合作． 联系人：唐娜 联系电话：020-33520954 联系手机：13533999436
国家企业培训师资格认证考试通知 今年下半年第二期企业培训师资格认证培训班定于10月22日开班，想参加本届国 家统一考试的学员，请在10月份之前准备好相关资料报名。由于人数众多，报名 从速，额满为止，敬请谅解！详情请见：http://www.21renzheng.com/servi ce/hrd.htm 咨询电话：（020）61311042 杨老师 24小时服务热线：13316081650 张老师 [BUTTON]
TOTO卫洗丽 5.3 折 / 美标 5.4 折 数百种建材产品特价销售 更多特色服务： ・免费提供装潢预算 ・装潢专家在线答疑 ・数千张装潢照片可供查询 ・建材淘宝乐，超特价满载 ・数百种建材产品组织团购 欲知更多团购产品的价格，请登陆 www.funasia.cn ！网站还会广大网友提供在线答疑、免费制作装潢预算等帮助！千万别错过这个 机会！
本公司有部分普通发票（商品销售发票）增值税发票及海关代征增值税专用缴款书及其它服务行业发票, 公路、内河运输发票。可以以低税率为贵公司代开，本公司具有内、外贸生意实力，保证我司开具的票据的真实性。 希望可以合作!共同发展!敬侯您的来电洽谈、咨询！ 联系人： 李先生 联系电话：13632588281 如有打扰望谅解，祝商琪。
致贵公司财务： 我司是广州平云贸易有限公司，在广东有一定的经济能力和相当的社会地位， 近来由于业务括展，经研究决定对外代开结余的商品-服务-广告-建筑等一系列发票。 所收税点依所开之数量取决。 本公司宗旨是所开出发票皆可验证后付款，如有需欢迎来电咨询。 联系人：罗先生 联系电话：13828451326 广州平云贸易有限公司
国家企业培训师资格认证考试通知 今年下半年第二期企业培训师资格认证培训班定于10月22日开班，想参加本届国 家统一考试的学员，请在10月份之前准备好相关资料报名。由于人数众多，报名 从速，额满为止，敬请谅解！详情请见：http://www.21renzheng.com/servi ce/hrd.htm 咨询电话：（020）61311042 杨老师 24小时服务热线：13316081650 张老师 [BUTTON]
国家企业培训师资格认证考试通知 今年下半年第二期企业培训师资格认证培训班定于10月22日开班，想参加本届国 家统一考试的学员，请在10月份之前准备好相关资料报名。由于人数众多，报名 从速，额满为止，敬请谅解！详情请见：http://www.21renzheng.com/servi ce/hrd.htm 咨询电话：（020）61311042 杨老师 24小时服务热线：13316081650 张老师 [BUTTON]
TOTO卫洗丽 5.3 折 / 美标 5.4 折 数百种建材产品特价销售 更多特色服务： ・免费提供装潢预算 ・装潢专家在线答疑 ・数千张装潢照片可供查询 ・建材淘宝乐，超特价满载 ・数百种建材产品组织团购 欲知更多团购产品的价格，请登陆 www.funasia.cn ！网站还会广大网友提供在线答疑、免费制作装潢预算等帮助！千万别错过这个 机会！
转让超薄蓝光DVD播放机一台： 可播放DVD，DVCD，SVCD，VCD，CD，HDCD，CD-R，CD-RW，JPEG， MR。OKO，MP3，DVD+RW/R，DVD-RW/R，等光盘 各种接口齐全，S端子，PPY色差，视频输入，AV，铜轴，光钎， 重低音，中置，话筒，左右环绕等，电压110V-250V自适应 随机附送：DVD-8000MB光盘 产品已通过CCC认证，证书编号：2005010806076646执行标准:ISO9002 有意者邮件联系
国家企业培训师资格认证考试通知 今年下半年第二期企业培训师资格认证培训班定于10月22日开班，想参加本届国 家统一考试的学员，请在10月份之前准备好相关资料报名。由于人数众多，报名 从速，额满为止，敬请谅解！详情请见：http://www.21renzheng.com/servi ce/hrd.htm 咨询电话：（020）61311042 杨老师 24小时服务热线：13316081650 张老师 [BUTTON]
新时代健康产业作为中国新时代集团的支柱产业，以“国珍”牌松花粉成功上市为标志，迄今已研制出了以“国珍” 为主品牌的系列健康食品和生活用品，受益者逾万。被商务部评为 AAA级（最高级别）诚心企业。 目前以进入蓄势待发阶段，一个创造财富的黄金时期即将来临。 详情登陆http://www.sab365.com/ Emil:sjygz@sab365.com 电话：010－65531187
真的太好了，这个软件可以合成文字，随便你输入什么内容，他都能够合成，有男声和甜美的女声，朗读速 度可以随便调节。还可以保存合成好的语音文件，以后你看新闻时就不用担心眼睛累了，可以让他来读，你泡 杯茶坐在沙发上听，彻底解放您的眼睛，犒劳您的耳朵。软件的下载地址是http://www.spqj.com/KDVoice/in dex.asp，软件完全免费
网址：www.8wx8.com,电话：0551-3961292，QQ：415455699 2、网站制作（复制，你所见的网站，就是你的网上家园）
、、、、、、、
data_helpers.py
# encoding: UTF-8
import numpy as np
import re
import itertools
from collections import Counter
import os
import word2vec_helpers
import time
import pickle
def load_data_and_labels(input_text_file, input_label_file, num_labels):
x_text = read_and_clean_zh_file(input_text_file)
y = None if not os.path.exists(input_label_file) else map(int, list(open(input_label_file, "r").readlines()))
return (x_text, y)
def load_positive_negative_data_files(positive_data_file, negative_data_file):
"""
Loads MR polarity data from files, splits the data into words and generates labels.
Returns split sentences and labels.
"""
# Load data from files
positive_examples = read_and_clean_zh_file(positive_data_file)
negative_examples = read_and_clean_zh_file(negative_data_file)
# Combine data
x_text = positive_examples + negative_examples
# Generate labels
positive_labels = [[0, 1] for _ in positive_examples]
negative_labels = [[1, 0] for _ in negative_examples]
y = np.concatenate([positive_labels, negative_labels], 0)
return [x_text, y]
def padding_sentences(input_sentences, padding_token, padding_sentence_length = None):
sentences = [sentence.split(' ') for sentence in input_sentences]
max_sentence_length = padding_sentence_length if padding_sentence_length is not None else max([len(sentence) for sentence in sentences])
for sentence in sentences:
if len(sentence) > max_sentence_length:
sentence = sentence[:max_sentence_length]
else:
sentence.extend([padding_token] * (max_sentence_length - len(sentence)))
return (sentences, max_sentence_length)
def batch_iter(data, batch_size, num_epochs, shuffle=True):
'''
Generate a batch iterator for a dataset
'''
data = np.array(data)
data_size = len(data)
num_batches_per_epoch = int((data_size - 1) / batch_size) + 1
for epoch in range(num_epochs):
if shuffle:
# Shuffle the data at each epoch
shuffle_indices = np.random.permutation(np.arange(data_size))
shuffled_data = data[shuffle_indices]
else:
shuffled_data = data
for batch_num in range(num_batches_per_epoch):
start_idx = batch_num * batch_size
end_idx = min((batch_num + 1) * batch_size, data_size)
yield shuffled_data[start_idx : end_idx]
def test():
# Test clean_str
print("Test")
#print(clean_str("This's a huge dog! Who're going to the top."))
# Test load_positive_negative_data_files
#x_text,y = load_positive_negative_data_files("./tiny_data/rt-polarity.pos", "./tiny_data/rt-polarity.neg")
#print(x_text)
#print(y)
# Test batch_iter
#batches = batch_iter(x_text, 2, 4)
#for batch in batches:
# print(batch)
def mkdir_if_not_exist(dirpath):
if not os.path.exists(dirpath):
os.mkdir(dirpath)
def seperate_line(line):
return ''.join([word + ' ' for word in line])
def read_and_clean_zh_file(input_file, output_cleaned_file = None):
lines = list(open(input_file, "rb").readlines())
lines = [clean_str(seperate_line(line.decode('utf-8'))) for line in lines]
if output_cleaned_file is not None:
with open(output_cleaned_file, 'w') as f:
for line in lines:
f.write((line + '\n').encode('utf-8'))
return lines
def clean_str(string):
string = re.sub(r"[^\u4e00-\u9fff]", " ", string)
#string = re.sub(r"[^A-Za-z0-9(),!?\'\`]", " ", string)
#string = re.sub(r"\'s", " \'s", string)
#string = re.sub(r"\'ve", " \'ve", string)
#string = re.sub(r"n\'t", " n\'t", string)
#string = re.sub(r"\'re", " \'re", string)
#string = re.sub(r"\'d", " \'d", string)
#string = re.sub(r"\'ll", " \'ll", string)
#string = re.sub(r",", " , ", string)
#string = re.sub(r"!", " ! ", string)
#string = re.sub(r"\(", " \( ", string)
#string = re.sub(r"\)", " \) ", string)
#string = re.sub(r"\?", " \? ", string)
string = re.sub(r"\s{2,}", " ", string)
#return string.strip().lower()
return string.strip()
def saveDict(input_dict, output_file):
with open(output_file, 'wb') as f:
pickle.dump(input_dict, f)
def loadDict(dict_file):
output_dict = None
with open(dict_file, 'rb') as f:
output_dict = pickle.load(f)
return output_dict
text_cnn.py
import tensorflow as tf
import numpy as np
class TextCNN(object):
'''
A CNN for text classification
Uses and embedding layer, followed by a convolutional, max-pooling and softmax layer.
'''
def __init__(
self, sequence_length, num_classes,
embedding_size, filter_sizes, num_filters, l2_reg_lambda=0.0):
# Placeholders for input, output, dropout
self.input_x = tf.placeholder(tf.float32, [None, sequence_length, embedding_size], name = "input_x")
self.input_y = tf.placeholder(tf.float32, [None, num_classes], name = "input_y")
self.dropout_keep_prob = tf.placeholder(tf.float32, name = "dropout_keep_prob")
# Keeping track of l2 regularization loss (optional)
l2_loss = tf.constant(0.0)
# Embedding layer
# self.embedded_chars = [None(batch_size), sequence_size, embedding_size]
# self.embedded_chars = [None(batch_size), sequence_size, embedding_size, 1(num_channels)]
self.embedded_chars = self.input_x
self.embedded_chars_expended = tf.expand_dims(self.embedded_chars, -1)
# Create a convolution + maxpool layer for each filter size
pooled_outputs = []
for i, filter_size in enumerate(filter_sizes):
with tf.name_scope("conv-maxpool-%s" % filter_size):
# Convolution layer
filter_shape = [filter_size, embedding_size, 1, num_filters]
W = tf.Variable(tf.truncated_normal(filter_shape, stddev=0.1), name="W")
b = tf.Variable(tf.constant(0.1, shape=[num_filters]), name="b")
conv = tf.nn.conv2d(self.embedded_chars_expended,
W,
strides=[1,1,1,1],
padding="VALID",
name="conv")
# Apply nonlinearity
h = tf.nn.relu(tf.nn.bias_add(conv, b), name = "relu")
# Maxpooling over the outputs
pooled = tf.nn.max_pool(
h,
ksize=[1, sequence_length - filter_size + 1, 1, 1],
strides=[1,1,1,1],
padding="VALID",
name="pool")
pooled_outputs.append(pooled)
# Combine all the pooled features
num_filters_total = num_filters * len(filter_sizes)
self.h_pool = tf.concat(3, pooled_outputs)
self.h_pool_flat = tf.reshape(self.h_pool, [-1, num_filters_total])
# Add dropout
with tf.name_scope("dropout"):
self.h_drop = tf.nn.dropout(self.h_pool_flat, self.dropout_keep_prob)
# Final (unnomalized) scores and predictions
with tf.name_scope("output"):
W = tf.get_variable(
"W",
shape = [num_filters_total, num_classes],
initializer = tf.contrib.layers.xavier_initializer())
b = tf.Variable(tf.constant(0.1, shape=[num_classes], name = "b"))
l2_loss += tf.nn.l2_loss(W)
l2_loss += tf.nn.l2_loss(b)
self.scores = tf.nn.xw_plus_b(self.h_drop, W, b, name = "scores")
self.predictions = tf.argmax(self.scores, 1, name = "predictions")
# Calculate Mean cross-entropy loss
with tf.name_scope("loss"):
losses = tf.nn.softmax_cross_entropy_with_logits(logits = self.scores, labels = self.input_y)
self.loss = tf.reduce_mean(losses) + l2_reg_lambda * l2_loss
# Accuracy
with tf.name_scope("accuracy"):
correct_predictions = tf.equal(self.predictions, tf.argmax(self.input_y, 1))
self.accuracy = tf.reduce_mean(tf.cast(correct_predictions, "float"), name = "accuracy")
word2vec_helpers.py
# -*- coding: utf-8 -*-
'''
python word2vec_helpers.py input_file output_model_file output_vector_file
'''
# import modules & set up logging
import os
import sys
import logging
import multiprocessing
import time
import json
from gensim.models import Word2Vec
from gensim.models.word2vec import LineSentence
def output_vocab(vocab):
for k, v in vocab.items():
print(k)
def embedding_sentences(sentences, embedding_size = 128, window = 5, min_count = 5, file_to_load = None, file_to_save = None):
if file_to_load is not None:
w2vModel = Word2Vec.load(file_to_load)
else:
w2vModel = Word2Vec(sentences, size = embedding_size, window = window, min_count = min_count, workers = multiprocessing.cpu_count())
if file_to_save is not None:
w2vModel.save(file_to_save)
all_vectors = []
embeddingDim = w2vModel.vector_size
embeddingUnknown = [0 for i in range(embeddingDim)]
for sentence in sentences:
this_vector = []
for word in sentence:
if word in w2vModel.wv.vocab:
this_vector.append(w2vModel[word])
else:
this_vector.append(embeddingUnknown)
all_vectors.append(this_vector)
return all_vectors
def generate_word2vec_files(input_file, output_model_file, output_vector_file, size = 128, window = 5, min_count = 5):
start_time = time.time()
# trim unneeded model memory = use(much) less RAM
# model.init_sims(replace=True)
model = Word2Vec(LineSentence(input_file), size = size, window = window, min_count = min_count, workers = multiprocessing.cpu_count())
model.save(output_model_file)
model.wv.save_word2vec_format(output_vector_file, binary=False)
end_time = time.time()
print("used time : %d s" % (end_time - start_time))
def run_main():
program = os.path.basename(sys.argv[0])
logger = logging.getLogger(program)
logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)
logger.info("running %s" % ' '.join(sys.argv))
# check and process input arguments
if len(sys.argv) < 4:
print (globals()['__doc__'] % locals())
sys.exit(1)
input_file, output_model_file, output_vector_file = sys.argv[1:4]
generate_word2vec_files(input_file, output_model_file, output_vector_file)
def test():
vectors = embedding_sentences([['first', 'sentence'], ['second', 'sentence']], embedding_size = 4, min_count = 1)
print(vectors)
train.py
#! /usr/bin/env python
# encoding: utf-8
import tensorflow as tf
import numpy as np
import os
import time
import datetime
import data_helpers
import word2vec_helpers
from text_cnn import TextCNN
# Parameters
# =======================================================
# Data loading parameters
tf.flags.DEFINE_float("dev_sample_percentage", .1, "Percentage of the training data to use for validation")
#tf.flags.DEFINE_string("positive_data_file", "./data/rt-polaritydata/rt-polarity.pos", "Data source for the positive data.")
#tf.flags.DEFINE_string("negative_data_file", "./data/rt-polaritydata/rt-polarity.neg", "Data source for the negative data.")
tf.flags.DEFINE_string("positive_data_file", "./data/ham_5000.utf8", "Data source for the positive data.")
tf.flags.DEFINE_string("negative_data_file", "./data/spam_5000.utf8", "Data source for the negative data.")
tf.flags.DEFINE_integer("num_labels", 2, "Number of labels for data. (default: 2)")
# Model hyperparameters
tf.flags.DEFINE_integer("embedding_dim", 128, "Dimensionality of character embedding (default: 128)")
tf.flags.DEFINE_string("filter_sizes", "3,4,5", "Comma-spearated filter sizes (default: '3,4,5')")
tf.flags.DEFINE_integer("num_filters", 128, "Number of filters per filter size (default: 128)")
tf.flags.DEFINE_float("dropout_keep_prob", 0.5, "Dropout keep probability (default: 0.5)")
tf.flags.DEFINE_float("l2_reg_lambda", 0.0, "L2 regularization lambda (default: 0.0)")
# Training paramters
tf.flags.DEFINE_integer("batch_size", 64, "Batch Size (default: 64)")
tf.flags.DEFINE_integer("num_epochs", 200, "Number of training epochs (default: 200)")
tf.flags.DEFINE_integer("evaluate_every", 100, "Evalue model on dev set after this many steps (default: 100)")
tf.flags.DEFINE_integer("checkpoint_every", 100, "Save model after this many steps (defult: 100)")
tf.flags.DEFINE_integer("num_checkpoints", 5, "Number of checkpoints to store (default: 5)")
# Misc parameters
tf.flags.DEFINE_boolean("allow_soft_placement", True, "Allow device soft device placement")
tf.flags.DEFINE_boolean("log_device_placement", False, "Log placement of ops on devices")
# Parse parameters from commands
FLAGS = tf.flags.FLAGS
FLAGS._parse_flags()
print("\nParameters:")
for attr, value in sorted(FLAGS.__flags.items()):
print("{}={}".format(attr.upper(), value))
print("")
# Prepare output directory for models and summaries
# =======================================================
timestamp = str(int(time.time()))
out_dir = os.path.abspath(os.path.join(os.path.curdir, "runs", timestamp))
print("Writing to {}\n".format(out_dir))
if not os.path.exists(out_dir):
os.makedirs(out_dir)
# Data preprocess
# =======================================================
# Load data
print("Loading data...")
x_text, y = data_helpers.load_positive_negative_data_files(FLAGS.positive_data_file, FLAGS.negative_data_file)
# Get embedding vector
sentences, max_document_length = data_helpers.padding_sentences(x_text, '<PADDING>')
x = np.array(word2vec_helpers.embedding_sentences(sentences, embedding_size = FLAGS.embedding_dim, file_to_save = os.path.join(out_dir, 'trained_word2vec.model')))
print("x.shape = {}".format(x.shape))
print("y.shape = {}".format(y.shape))
# Save params
training_params_file = os.path.join(out_dir, 'training_params.pickle')
params = {'num_labels' : FLAGS.num_labels, 'max_document_length' : max_document_length}
data_helpers.saveDict(params, training_params_file)
# Shuffle data randomly
np.random.seed(10)
shuffle_indices = np.random.permutation(np.arange(len(y)))
x_shuffled = x[shuffle_indices]
y_shuffled = y[shuffle_indices]
# Split train/test set
# TODO: This is very crude, should use cross-validation
dev_sample_index = -1 * int(FLAGS.dev_sample_percentage * float(len(y)))
x_train, x_dev = x_shuffled[:dev_sample_index], x_shuffled[dev_sample_index:]
y_train, y_dev = y_shuffled[:dev_sample_index], y_shuffled[dev_sample_index:]
print("Train/Dev split: {:d}/{:d}".format(len(y_train), len(y_dev)))
# Training
# =======================================================
with tf.Graph().as_default():
session_conf = tf.ConfigProto(
allow_soft_placement = FLAGS.allow_soft_placement,
log_device_placement = FLAGS.log_device_placement)
sess = tf.Session(config = session_conf)
with sess.as_default():
cnn = TextCNN(
sequence_length = x_train.shape[1],
num_classes = y_train.shape[1],
embedding_size = FLAGS.embedding_dim,
filter_sizes = list(map(int, FLAGS.filter_sizes.split(","))),
num_filters = FLAGS.num_filters,
l2_reg_lambda = FLAGS.l2_reg_lambda)
# Define Training procedure
global_step = tf.Variable(0, name="global_step", trainable=False)
optimizer = tf.train.AdamOptimizer(1e-3)
grads_and_vars = optimizer.compute_gradients(cnn.loss)
train_op = optimizer.apply_gradients(grads_and_vars, global_step=global_step)
# Keep track of gradient values and sparsity (optional)
grad_summaries = []
for g, v in grads_and_vars:
if g is not None:
grad_hist_summary = tf.summary.histogram("{}/grad/hist".format(v.name), g)
sparsity_summary = tf.summary.scalar("{}/grad/sparsity".format(v.name), tf.nn.zero_fraction(g))
grad_summaries.append(grad_hist_summary)
grad_summaries.append(sparsity_summary)
grad_summaries_merged = tf.summary.merge(grad_summaries)
# Output directory for models and summaries
print("Writing to {}\n".format(out_dir))
# Summaries for loss and accuracy
loss_summary = tf.summary.scalar("loss", cnn.loss)
acc_summary = tf.summary.scalar("accuracy", cnn.accuracy)
# Train Summaries
train_summary_op = tf.summary.merge([loss_summary, acc_summary, grad_summaries_merged])
train_summary_dir = os.path.join(out_dir, "summaries", "train")
train_summary_writer = tf.summary.FileWriter(train_summary_dir, sess.graph)
# Dev summaries
dev_summary_op = tf.summary.merge([loss_summary, acc_summary])
dev_summary_dir = os.path.join(out_dir, "summaries", "dev")
dev_summary_writer = tf.summary.FileWriter(dev_summary_dir, sess.graph)
# Checkpoint directory. Tensorflow assumes this directory already exists so we need to create it
checkpoint_dir = os.path.abspath(os.path.join(out_dir, "checkpoints"))
checkpoint_prefix = os.path.join(checkpoint_dir, "model")
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
saver = tf.train.Saver(tf.global_variables(), max_to_keep=FLAGS.num_checkpoints)
# Initialize all variables
sess.run(tf.global_variables_initializer())
def train_step(x_batch, y_batch):
"""
A single training step
"""
feed_dict = {
cnn.input_x: x_batch,
cnn.input_y: y_batch,
cnn.dropout_keep_prob: FLAGS.dropout_keep_prob
}
_, step, summaries, loss, accuracy = sess.run(
[train_op, global_step, train_summary_op, cnn.loss, cnn.accuracy],
feed_dict)
time_str = datetime.datetime.now().isoformat()
print("{}: step {}, loss {:g}, acc {:g}".format(time_str, step, loss, accuracy))
train_summary_writer.add_summary(summaries, step)
def dev_step(x_batch, y_batch, writer=None):
"""
Evaluates model on a dev set
"""
feed_dict = {
cnn.input_x: x_batch,
cnn.input_y: y_batch,
cnn.dropout_keep_prob: 1.0
}
step, summaries, loss, accuracy = sess.run(
[global_step, dev_summary_op, cnn.loss, cnn.accuracy],
feed_dict)
time_str = datetime.datetime.now().isoformat()
print("{}: step {}, loss {:g}, acc {:g}".format(time_str, step, loss, accuracy))
if writer:
writer.add_summary(summaries, step)
# Generate batches
batches = data_helpers.batch_iter(
list(zip(x_train, y_train)), FLAGS.batch_size, FLAGS.num_epochs)
# Training loop. For each batch...
for batch in batches:
x_batch, y_batch = zip(*batch)
train_step(x_batch, y_batch)
current_step = tf.train.global_step(sess, global_step)
if current_step % FLAGS.evaluate_every == 0:
print("\nEvaluation:")
dev_step(x_dev, y_dev, writer=dev_summary_writer)
print("")
if current_step % FLAGS.checkpoint_every == 0:
path = saver.save(sess, checkpoint_prefix, global_step=current_step)
print("Saved model checkpoint to {}\n".format(path))
eval.py
#! /usr/bin/env python
import tensorflow as tf
import numpy as np
import os
import time
import datetime
import data_helpers
import word2vec_helpers
from text_cnn import TextCNN
import csv
# Parameters
# ==================================================
# Data Parameters
tf.flags.DEFINE_string("input_text_file", "./data/spam_100.utf8", "Test text data source to evaluate.")
tf.flags.DEFINE_string("input_label_file", "", "Label file for test text data source.")
# Eval Parameters
tf.flags.DEFINE_integer("batch_size", 64, "Batch Size (default: 64)")
tf.flags.DEFINE_string("checkpoint_dir", "", "Checkpoint directory from training run")
tf.flags.DEFINE_boolean("eval_train", True, "Evaluate on all training data")
# Misc Parameters
tf.flags.DEFINE_boolean("allow_soft_placement", True, "Allow device soft device placement")
tf.flags.DEFINE_boolean("log_device_placement", False, "Log placement of ops on devices")
FLAGS = tf.flags.FLAGS
FLAGS._parse_flags()
print("\nParameters:")
for attr, value in sorted(FLAGS.__flags.items()):
print("{}={}".format(attr.upper(), value))
print("")
# validate
# ==================================================
# validate checkout point file
checkpoint_file = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
if checkpoint_file is None:
print("Cannot find a valid checkpoint file!")
exit(0)
print("Using checkpoint file : {}".format(checkpoint_file))
# validate word2vec model file
trained_word2vec_model_file = os.path.join(FLAGS.checkpoint_dir, "..", "trained_word2vec.model")
if not os.path.exists(trained_word2vec_model_file):
print("Word2vec model file \'{}\' doesn't exist!".format(trained_word2vec_model_file))
print("Using word2vec model file : {}".format(trained_word2vec_model_file))
# validate training params file
training_params_file = os.path.join(FLAGS.checkpoint_dir, "..", "training_params.pickle")
if not os.path.exists(training_params_file):
print("Training params file \'{}\' is missing!".format(training_params_file))
print("Using training params file : {}".format(training_params_file))
# Load params
params = data_helpers.loadDict(training_params_file)
num_labels = int(params['num_labels'])
max_document_length = int(params['max_document_length'])
# Load data
if FLAGS.eval_train:
x_raw, y_test = data_helpers.load_data_and_labels(FLAGS.input_text_file, FLAGS.input_label_file, num_labels)
else:
x_raw = ["a masterpiece four years in the making", "everything is off."]
y_test = [1, 0]
# Get Embedding vector x_test
sentences, max_document_length = data_helpers.padding_sentences(x_raw, '<PADDING>', padding_sentence_length = max_document_length)
x_test = np.array(word2vec_helpers.embedding_sentences(sentences, file_to_load = trained_word2vec_model_file))
print("x_test.shape = {}".format(x_test.shape))
# Evaluation
# ==================================================
print("\nEvaluating...\n")
checkpoint_file = tf.train.latest_checkpoint(FLAGS.checkpoint_dir)
graph = tf.Graph()
with graph.as_default():
session_conf = tf.ConfigProto(
allow_soft_placement=FLAGS.allow_soft_placement,
log_device_placement=FLAGS.log_device_placement)
sess = tf.Session(config=session_conf)
with sess.as_default():
# Load the saved meta graph and restore variables
saver = tf.train.import_meta_graph("{}.meta".format(checkpoint_file))
saver.restore(sess, checkpoint_file)
# Get the placeholders from the graph by name
input_x = graph.get_operation_by_name("input_x").outputs[0]
# input_y = graph.get_operation_by_name("input_y").outputs[0]
dropout_keep_prob = graph.get_operation_by_name("dropout_keep_prob").outputs[0]
# Tensors we want to evaluate
predictions = graph.get_operation_by_name("output/predictions").outputs[0]
# Generate batches for one epoch
batches = data_helpers.batch_iter(list(x_test), FLAGS.batch_size, 1, shuffle=False)
# Collect the predictions here
all_predictions = []
for x_test_batch in batches:
batch_predictions = sess.run(predictions, {input_x: x_test_batch, dropout_keep_prob: 1.0})
all_predictions = np.concatenate([all_predictions, batch_predictions])
# Print accuracy if y_test is defined
if y_test is not None:
correct_predictions = float(sum(all_predictions == y_test))
print("Total number of test examples: {}".format(len(y_test)))
print("Accuracy: {:g}".format(correct_predictions/float(len(y_test))))
# Save the evaluation to a csv
predictions_human_readable = np.column_stack((np.array([text.encode('utf-8') for text in x_raw]), all_predictions))
out_path = os.path.join(FLAGS.checkpoint_dir, "..", "prediction.csv")
print("Saving evaluation to {0}".format(out_path))
with open(out_path, 'w') as f:
csv.writer(f).writerows(predictions_human_readable)
1204
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
