tensorflow37《TensorFlow实战》笔记-07-01 TensorFlow实现Word2Vec code

# 《TensorFlow实战》07 TensorFlow实现循环神经网络及Word2Vec
# win10 Tensorflow1.0.1 python3.5.3
# CUDA v8.0 cudnn-8.0-windows10-x64-v5.1
# filename:sz07.01.py # TensorFlow实现Word2Vec

# https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/tutorials/word2vec/word2vec_basic.py
# tensorflow\tensorflow\examples\tutorials\word2vec\word2vec_basic.py

import collections
import math
import os
import random
import zipfile
import numpy as np
from six.moves import urllib
import tensorflow as tf

# 如果下载失败，就手动下载http://mattmahoney.net/dc/text8.zip到sz07.01.py同目录下
# http://mattmahoney.net/dc/text8.zip
url = 'http://mattmahoney.net/dc/'

def maybe_download(filename, expected_bytes):
    if not os.path.exists(filename):
        filename, _ = urllib.request.urlretrieve(url + filename, filename)
    statinfo = os.stat(filename)
    if statinfo.st_size == expected_bytes:
        print('Found and verified', filename)
    else:
        print(statinfo.st_size)
        raise Exception('Failed to verify ' + filename + '. Can you get to it with a browser?')
    return filename

filename = maybe_download('text8.zip', 31344016)

def read_data(filename):
    with zipfile.ZipFile(filename) as f:
        data = tf.compat.as_str(f.read(f.namelist()[0])).split()
    return data

words = read_data(filename)
print('Data size', len(words))

vocabulary_size = 50000

def build_dataset(words):
    count = [['UNK', -1]]
    count.extend(collections.Counter(words).most_common(vocabulary_size - 1))
    dictionary = dict()
    for word, _ in count:
        dictionary[word] = len(dictionary)
    data = list()
    unk_count = 0
    for word in words:
        if word in dictionary:
            index = dictionary[word]
        else:
            index = 0
            unk_count += 1
        data.append(index)
    count[0][1] = unk_count
    reverse_dictionary = dict(zip(dictionary.values(), dictionary.keys()))
    return data, count, dictionary, reverse_dictionary
data, count, dictionary, reverse_dictionary = build_dataset(words)

del words
print('Most common words (+UNK)', count[:5])
print('Sample data', data[:10], [reverse_dictionary[i] for i in data[:10]])

data_index = 0

def generate_batch(batch_size, num_skips, skip_window):
    global data_index
    assert batch_size % num_skips == 0
    assert num_skips <= 2 * skip_window
    batch = np.ndarray(shape=(batch_size), dtype=np.int32)
    labels = np.ndarray(shape=(batch_size, 1), dtype=np.int32)
    span = 2 * skip_window + 1
    buffer = collections.deque(maxlen=span)
    for _ in range(span):
        buffer.append(data[data_index])
        data_index = (data_index + 1) % len(data)
    for i in range(batch_size // num_skips):
        target = skip_window
        targets_to_avoid = [skip_window]
        for j in range(num_skips):
            while target in targets_to_avoid:
                target = random.randint(0, span - 1)
            targets_to_avoid.append(target)
            batch[i * num_skips + j] = buffer[skip_window]
            labels[i * num_skips + j, 0] = buffer[target]
        buffer.append(data[data_index])
        data_index = (data_index + 1) % len(data)
    return batch, labels

batch, labels = generate_batch(batch_size=8, num_skips=2, skip_window=1)
for i in range(8):
    print(batch[i], reverse_dictionary[batch[i]], '->', labels[i, 0], reverse_dictionary[labels[i, 0]])

batch_size = 128
embedding_size = 128
skip_window = 1
num_skips = 2

valid_size = 16
valid_window = 100
valid_examples = np.random.choice(valid_window, valid_size, replace=False)
num_sampled = 64

graph = tf.Graph()
with graph.as_default():
    train_inputs = tf.placeholder(tf.int32, shape=[batch_size])
    train_labels = tf.placeholder(tf.int32, shape=[batch_size, 1])
    valid_dataset = tf.constant(valid_examples, dtype=tf.int32)

    with tf.device('/cpu:0'):
        embeddings = tf.Variable(tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0))
        embed = tf.nn.embedding_lookup(embeddings, train_inputs)
        nce_weights = tf.Variable(
            tf.truncated_normal([vocabulary_size, embedding_size], stddev=1.0/math.sqrt(embedding_size)))
        nce_biases = tf.Variable(tf.zeros([vocabulary_size]))
    loss = tf.reduce_mean(tf.nn.nce_loss(weights=nce_weights,
                                         biases=nce_biases,
                                         labels=train_labels,
                                         inputs=embed,
                                         num_sampled=num_sampled,
                                         num_classes=vocabulary_size))

    optimizer = tf.train.GradientDescentOptimizer(1.0).minimize(loss)
    norm = tf.sqrt(tf.reduce_sum(tf.square(embeddings), 1, keep_dims=True))
    normalized_embeddings = embeddings / norm
    valid_embeddings = tf.nn.embedding_lookup(normalized_embeddings, valid_dataset)
    similarity = tf.matmul(valid_embeddings, normalized_embeddings, transpose_b=True)
    init = tf.global_variables_initializer()

num_steps = 100001
with tf.Session(graph=graph) as session:
    init.run()
    print('Initialized')

    average_loss = 0
    for step in range(num_steps):
        batch_inputs, batch_labels = generate_batch(batch_size, num_skips, skip_window)
        feed_dict = {train_inputs: batch_inputs, train_labels: batch_labels}
        _, loss_val = session.run([optimizer, loss], feed_dict=feed_dict)
        average_loss += loss_val

        if step % 2000 == 0:
            if step > 0:
                average_loss /= 2000
            print("Average loss at step ", step, ": ", average_loss)
            average_loss = 0

        if step % 10000 == 0:
            sim = similarity.eval()
            for i in range(valid_size):
                valid_word = reverse_dictionary[valid_examples[i]]
                top_k = 8
                nearest = (-sim[i, :]).argsort()[1:top_k+1]
                log_str = "Nearest to %s:" % valid_word
                for k in range(top_k):
                    close_word = reverse_dictionary[nearest[k]]
                    log_str = "%s %s," % (log_str, close_word)
                print(log_str)
    final_embeddings = normalized_embeddings.eval()

from sklearn.manifold import TSNE
import matplotlib.pyplot as plt
def plot_with_labels(low_dim_embs, labels, filename='tsne.png'):
    assert low_dim_embs.shape[0] >= len(labels), "More labels than embeddings"
    plt.figure(figsize=(18,18))
    for i, label in enumerate(labels):
        x, y = low_dim_embs[i, :]
        plt.scatter(x, y)
        plt.annotate(label,
                     xy=(x,y),
                     xytext=(5,2),
                     textcoords='offset points',
                     ha='right',
                     va='bottom')
    plt.savefig(filename)

tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
plot_only = 100
low_dim_embs = tsne.fit_transform(final_embeddings[:plot_only, :])
labels = [reverse_dictionary[i] for i in range(plot_only)]
plot_with_labels(low_dim_embs, labels)
'''
Found and verified text8.zip
Data size 17005207
Most common words (+UNK) [['UNK', 418391], ('the', 1061396), ('of', 593677), ('and', 416629), ('one', 411764)]
Sample data [5242, 3084, 12, 6, 195, 2, 3137, 46, 59, 156] ['anarchism', 'originated', 'as', 'a', 'term', 'of', 'abuse', 'first', 'used', 'against']
3084 originated -> 12 as
3084 originated -> 5242 anarchism
12 as -> 3084 originated
12 as -> 6 a
6 a -> 195 term
6 a -> 12 as
195 term -> 2 of
195 term -> 6 a

Initialized
Average loss at step  0 :  242.824676514
Nearest to this: beatrix, territorial, odo, eileen, bicameral, animists, baptize, malm,
...
Nearest to world: crb, alexander, diphthongs, adventist, persons, alabama, leftist, idol,
Nearest to would: will, may, can, could, might, must, should, to,
Average loss at step  92000 :  4.71387670171
Average loss at step  94000 :  4.62264616895
Average loss at step  96000 :  4.71346354306
Average loss at step  98000 :  4.62557934207
Average loss at step  100000 :  4.66860903692
Nearest to this: which, it, the, that, mico, spokesperson, one, dwell,
Nearest to however: but, although, that, dasyprocta, ssbn, and, opencyc, microsite,
Nearest to was: is, had, has, were, became, been, when, by,
Nearest to d: b, lomond, raions, gh, t, deemed, aon, microsite,
Nearest to use: thaler, histone, agouti, crb, primigenius, potomac, callithrix, upanija,
Nearest to other: many, agouti, including, these, some, bends, callithrix, clodius,
Nearest to over: about, lw, off, coolidge, absent, six, mileva, four,
Nearest to after: before, when, during, in, persisted, following, was, gave,
Nearest to called: UNK, used, subatomic, clo, aorta, and, emblems, implied,
Nearest to often: sometimes, usually, generally, still, commonly, now, frequently, also,
Nearest to of: nine, in, dasyprocta, and, including, callithrix, amo, thaler,
Nearest to have: had, has, are, were, be, include, aral, elevate,
Nearest to can: may, would, will, could, must, should, might, to,
Nearest to three: four, five, two, six, seven, eight, one, nine,
Nearest to world: alexander, diphthongs, crb, leftist, adventist, guerrilla, vortigern, appointment,
Nearest to would: will, may, can, could, might, must, should, to,
'''