Cacher is the code snippet organizer for pro developers

data_index = 0  #全局索引值，代表当前窗口最左侧在整个data词库中的位置

## generate batch from all samples
def generate_batch(batch_size, num_skips, skip_window):  #num_skips代表每个中心词周围选取的context word数量，skip_window代
    #表半个窗口大小
    global data_index
    assert batch_size % num_skips == 0#每个中心词取num_skips个context word，一个批次取中心词数量是batch_size//num_skips
    assert num_skips <= 2 * skip_window#窗口大小2*skip_window+1，取词个数要小于窗口大小
    batch = np.ndarray(shape=(batch_size), dtype=np.int32)
    labels = np.ndarray(shape=(batch_size, 1), dtype=np.int32)#为什么是这种shape,因为后面用到的nce_loss规定需要是这种shape，否则也应该是batch的shape：[batch_size]
    span = 2 * skip_window + 1#窗口大小2*skip_window+1
    buffer = collections.deque(maxlen=span)#保存窗口中的值，用双端队列的进出来模拟窗口滑动
    if data_index + span > len(data):
        data_index = 0  #越界则回0
    buffer.extend(data[data_index:data_index + span])
    data_index += span
    for i in range(batch_size // num_skips):
        target = skip_window
        targets_to_avoid = [skip_window]  #避免重复取词
        for j in range(num_skips):
            while target in targets_to_avoid:
                target = random.randint(0, span - 1)  #随机在中心词周围取词，不能重复取词
            targets_to_avoid.append(target)
            batch[i * num_skips + j] = buffer[skip_window]
            labels[i * num_skips + j, 0] = buffer[target]
        if data_index == len(data):
            buffer[:] = data[:span]
            data_index = span
        else:
            buffer.append(data[data_index])
            data_index += 1  #窗口滑动，并改变data_index
    data_index = (data_index + len(data) - span) % len(data)  #需要更新data_index，设置data_index为全局变量是因为
    #该方法会多次调用，每次调用会产生一个批次大小的数据量。作为输入和标签值。
    return batch, labels

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()
## create 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)

    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)
    
    #为了测试使用求得的embeddings，需要norm化。然后取一些原始words，找距离最近的即相似度最高的。
    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 = 40001

## create session and begins to training
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:  #测试并使用embeddings
            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()