简单demo的代码路径在tensorflow\tensorflow\g3doc\tutorials\word2vec\word2vec_basic.py
Sikp gram方式的model思路
http://tensorflow.org/tutorials/word2vec/index.md
另外可以参考cs224d课程的课件。
窗口设置为左右1个词
对应skip gram模型
就是一个单词预测其周围单词（cbow模型是
输入一系列context词，预测一个中心词)
Quick -> the quick -> brown
Skip gram的训练目标cost function是
对应
但是这样太耗时了
每一步训练时间代价都是O(VocabularySize)
于是我们采用了 nce(noise-contrastive estimation)的方式,也就是负样本采样，采用某种方式随机生成词作为负样本，比如 quick -> sheep ，sheep作为负样本，假设我们就取一个负样本
- 输入数据
这里是
分隔好的单词 - 读入单词存储到list中
-
统计词频 0号位置给 unknown, 其余按照频次由高到低排列，unknown的获取按照预设词典大小
比如50000，则频次排序靠后于50000的都视为unknown建立好 key->id id->key的双向索引map
4. 产生一组training batch
batch_size = 128
embedding_size = 128 # Dimension of the embedding vector.
skip_window = 1 # How many words to consider left and right.
num_skips = 2 # How many times to reuse an input to generate a label.
Batch_size每次sgd训练时候扫描的数据大小, embedding_size 词向量的大小，skip_window 窗口大小,
Num_skips = 2 表示input用了产生label的次数限制
demo中默认是2，
可以设置为1 对比下
默认2的时候
batch, labels = generate_batch(batch_size=8, num_skips=2, skip_window=1)
for i in range(8):
print(batch[i], '->', labels[i, 0])
print(reverse_dictionary[batch[i]], '->', reverse_dictionary[labels[i, 0]])
Sample data [5239, 3084, 12, 6, 195, 2, 3137, 46, 59, 156]
-> 5239
originated -> anarchism
-> 12
originated -> as
12 -> 6
as -> a
12 -> 3084
as -> originated
6 -> 195
a -> term
6 -> 12
a -> as
195 -> 2
term -> of
195 -> 6
term -> a
3084左侧出现2次，对应窗口左右各1
设置1的时候
batch, labels = generate_batch(batch_size=8, num_skips=1, skip_window=1)
for i in range(8):
print(batch[i], '->', labels[i, 0])
print(reverse_dictionary[batch[i]], '->', reverse_dictionary[labels[i, 0]])
Sample data [5239, 3084, 12, 6, 195, 2, 3137, 46, 59, 156]
-> 12
originated -> as
12 -> 3084
as -> originated
6 -> 12
a -> as
195 -> 2
term -> of
2 -> 3137
of -> abuse
3137 -> 46
abuse -> first
46 -> 59
first -> used
59 -> 156
3084左侧只出现1次
# Step 4: Function to generate a training batch for the skip-gram model.
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], '->', labels[i, 0])
print(reverse_dictionary[batch[i]], '->', reverse_dictionary[labels[i, 0]])
就是对于一个中心词
在window范围
随机选取 num_skips个词，产生一系列的
(input_id, output_id) 作为(batch_instance, label)
这些都是正样本
训练准备，
Input embedding W
Output embedding W^
后面code都比较容易理解，tf定义了nce_loss来自动处理，每次会自动添加随机负样本
num_sampled = 64 # Number of negative examples to sample.
graph = tf.Graph()
with graph.as_default():
# Input data.
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)
# Construct the variables.
embeddings = tf.Variable(
tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0))
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]))
# Look up embeddings for inputs.
embed = tf.nn.embedding_lookup(embeddings, train_inputs)
# Compute the average NCE loss for the batch.
# tf.nce_loss automatically draws a new sample of the negative labels each
# time we evaluate the loss.
loss = tf.reduce_mean(
tf.nn.nce_loss(nce_weights, nce_biases, embed, train_labels,
num_sampled, vocabulary_size))
# Construct the SGD optimizer using a learning rate of 1.0.
optimizer = tf.train.GradientDescentOptimizer(1.0).minimize(loss)
训练过程利用embedding矩阵的乘法计算了不同词向量的欧式距离
并计算了高频几个词对应的距离最近的词展示
最后调用 skitlearn的TSNE模块
进行降维到2元，绘图展示。