Python 如何在中获取单词向量坐标
我试图制作单词嵌入图,我做到了,但我想得到图中每个单词向量的坐标。这意味着,例如,我想得到显示单词的每个坐标 我使用了python、tensorflow,下面是我的代码,这是中的一个示例,我应该如何找到坐标?为了安全起见,我附加了代码Python 如何在中获取单词向量坐标,python,word2vec,Python,Word2vec,我试图制作单词嵌入图,我做到了,但我想得到图中每个单词向量的坐标。这意味着,例如,我想得到显示单词的每个坐标 我使用了python、tensorflow,下面是我的代码,这是中的一个示例,我应该如何找到坐标?为了安全起见,我附加了代码 # -*- coding: UTF-8 -*- from __future__ import absolute_import from __future__ import print_function import collections import m
# -*- coding: UTF-8 -*-
from __future__ import absolute_import
from __future__ import print_function
import collections
import math
import os
import random
import zipfile
import numpy as np
from six.moves import urllib
from six.moves import xrange # pylint: disable=redefined-builtin
import tensorflow as tf
import matplotlib
import matplotlib.font_manager as fm
font_location = "c:\\windows\\fonts\\malgun.ttf"
font_name = fm.FontProperties(fname=font_location).get_name()
matplotlib.rc('font', family=font_name)
# Step 1
filename = "text8.zip"
def read_data(filename):
with zipfile.ZipFile(filename) as f:
data = f.read(f.namelist()[0]).split()
for i, item in enumerate(data):
data[i] = item.decode('utf-8')
return data
words = read_data(filename)
print('Data size', len(words))
# Step 2
vocabulary_size = 10000
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 # dictionary['UNK']
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 # Hint to reduce memory.
print('Most common words (+UNK)', count[:5])
print('Sample data', data[:10], [reverse_dictionary[i] for i in data[:10]])
data_index = 0
# Step 3
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 # [ skip_window target skip_window ]
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 # target label at the center of the buffer
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]])
# Step 4
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)
# Ops and variables pinned to the CPU because of missing GPU implementation
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(nce_weights, nce_biases, embed, train_labels,
num_sampled, 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)
# Step 5
num_steps = 100001
with tf.Session(graph=graph) as session:
tf.initialize_all_variables().run()
print("Initialized")
average_loss = 0
for step in xrange(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
# Note that this is expensive (~20% slowdown if computed every 500 steps)
if step % 10000 == 0:
sim = similarity.eval()
for i in xrange(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 xrange(top_k):
close_word = reverse_dictionary[nearest[k]]
log_str = "%s %s," % (log_str, close_word)
print(log_str)
final_embeddings = normalized_embeddings.eval()
# Step 6
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)) #in inches
for i, tmlab in enumerate(labels):
# slabel = tmlab.decode('utf-8')
x, y = low_dim_embs[i,:]
plt.scatter(x, y)
plt.annotate(tmlab,
xy=(x, y),
xytext=(5, 2),
textcoords='offset points',
ha='right',
va='bottom')
plt.savefig(filename)
try:
from sklearn.manifold import TSNE
import matplotlib.pyplot as plt
tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
plot_only = 750
low_dim_embs = tsne.fit_transform(final_embeddings[:plot_only,:])
labels = [reverse_dictionary[i] for i in xrange(plot_only)]
plot_with_labels(low_dim_embs, labels)
except ImportError:
print("Please install sklearn and matplotlib to visualize embeddings.")
#-*-编码:UTF-8-*-
从未来导入绝对导入
来自未来导入打印功能
导入集合
输入数学
导入操作系统
随机输入
进口拉链
将numpy作为np导入
从六个移动导入urllib
from six.moves导入xrange#pylint:disable=重新定义的内置
导入tensorflow作为tf
导入matplotlib
将matplotlib.font\u管理器作为fm导入
font\u location=“c:\\windows\\font\\malgun.ttf”
font\u name=fm.FontProperties(fname=font\u location).get\u name()
matplotlib.rc('font',family=font\u name)
#第一步
filename=“text8.zip”
def read_数据(文件名):
使用zipfile.zipfile(文件名)作为f:
data=f.read(f.namelist()[0]).split()
对于i,枚举中的项目(数据):
数据[i]=项目解码('utf-8')
返回数据
words=读取数据(文件名)
打印('数据大小',长度(字))
#步骤2
词汇表大小=10000
def build_数据集(文字):
计数=['UNK',-1]]
count.extend(集合.计数器(单词).最常见(词汇表大小-1))
dictionary=dict()
对于word,计数为:
字典[字]=len(字典)
数据=列表()
unk_计数=0
用文字表示:
如果字典中有单词:
索引=字典[单词]
其他:
索引=0#字典['UNK']
unk_计数+=1
data.append(索引)
计数[0][1]=未计数
reverse\u dictionary=dict(zip(dictionary.values(),dictionary.keys())
返回数据、计数、字典、反向字典
数据、计数、字典、反向字典=构建数据集(单词)
del words#提示减少内存。
打印(‘最常见的单词(+UNK)’,计数[:5])
打印('Sample data',data[:10],[reverse_dictionary[i]for i in data[:10]]
数据索引=0
#步骤3
def生成批次(批次大小、跳过数量、跳过窗口):
全球数据索引
断言批大小%num\u跳过==0
assert num_跳过0:
平均损失/=2000
打印(“步骤中的平均损耗”,步骤“:”,平均损耗)
平均损耗=0
#请注意,这是昂贵的(如果每500步计算约20%的减速)
如果步骤%10000==0:
sim=相似性。eval()
对于X范围内的i(有效的_大小):
有效单词=反向字典[有效示例[i]]
top_k=8
最近的=(-sim[i,:])。argsort()[1:top_k+1]
log\u str=“最接近%s:“%valid\u word”
对于X范围内的k(顶部):
关闭单词=反向字典[最近的[k]]
log_str=“%s%s,”%(log_str,close_word)
打印(日志)
final_embedings=标准化的_embedings.eval()
#步骤6
带有标签的def plot_(低尺寸EMB,标签,文件名='tsne.png'):
assert low_dim_embs.shape[0]>=len(标签),“标签多于嵌入”
plt.数字(figsize=(18,18))#英寸
对于i,枚举中的tmlab(标签):
#slabel=tmlab.decode('utf-8')
x、 y=低尺寸EMB[i,:]
plt.散射(x,y)
产品说明(tmlab,
xy=(x,y),
xytext=(5,2),
textcoords='offset points',
哈"对",,
va='bottom')
plt.savefig(文件名)
尝试:
从sklearn.manifold导入TSNE
将matplotlib.pyplot作为plt导入
tsne=tsne(困惑=30,n_分量=2,初始值=pca',n_iter=5000)
plot_only=750
low_dim_embs=tsne.fit_变换(最终嵌入[:仅绘图,:])
labels=[X范围内i的反向字典[i](仅绘图)]
使用标签(低尺寸EMB、标签)打印
除恐怖外:
打印(“请安装sklearn和matplotlib以可视化嵌入。”)
您的嵌入维度可以任意大(50-300或更多)。但是,你想在一个二维空间上绘制这些单词。因为,您的嵌入大小是128,比2大得多,所以您需要在二维空间中投影这些嵌入。解决这个问题最著名的方法是。您可以使用来自此的代码。这对我有用。如果这解决了你的问题,请投票 嵌入维度可以任意大(50-300或更多)。但是,你想在一个二维空间上绘制这些单词。因为,您的嵌入大小是128,比2大得多,所以您需要在二维空间中投影这些嵌入。解决这个问题最著名的方法是。您可以使用来自此的代码。这对我有用。如果这解决了你的问题,请投票 但是如果我们只想存储原始坐标以供进一步计算呢?它们保存在哪个对象中?但是如果我们只想存储原始坐标以供进一步计算呢?它们保存在哪个对象中?