Python 尺寸必须相等，但为1和128表示'；采样的最大损耗/最大损耗'；（op:'；MatMul'；）带有输入形状：[128,1]，[64128]

尺寸必须相等，但输入形状为[128,1]、[64128]的“采样的软件最大损耗/MatMul”（op:“MatMul”）的尺寸为1和128

# -*- coding: utf-8 -*-
from __future__ import print_function
import collections
import math
import numpy as np
import random
import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()
from sklearn.manifold import TSNE
import pickle

sample = open("/Users/henry/Desktop/Cor.txt")
words = sample.read() 
print('Data size %d' % len(words))
sample.close()

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  # dictionary['UNK']
            unk_count = 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)
print('Most common words (+UNK)', count[:5])
print('Sample data', data[:10])
del words  # Hint to reduce memory.


data_index = 0

def generate_batch(batch_size, num_skips, skip_window):
    global data_index
    assert batch_size % num_skips == 0  # each word pair is a batch, so a training data [context target context] would increase batch number of 2.
    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

print('data:', [reverse_dictionary[di] for di in data[:8]])

for num_skips, skip_window in [(2, 1), (4, 2)]:
    data_index = 0
    batch, labels = generate_batch(batch_size=8, num_skips=num_skips, skip_window=skip_window)
    print('\nwith num_skips = %d and skip_window = %d:' % (num_skips, skip_window))
    print('    batch:', [reverse_dictionary[bi] for bi in batch])
    print('    labels:', [reverse_dictionary[li] for li in labels.reshape(8)])


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.
# We pick a random validation set to sample nearest neighbors. here we limit the
# validation samples to the words that have a low numeric ID, which by
# construction are also the most frequent. 
valid_size = 16 # Random set of words to evaluate similarity on.
valid_window = 100 # Only pick dev samples in the head of the distribution.
valid_examples = np.array(random.sample(range(valid_window), valid_size))
num_sampled = 64 # Number of negative examples to sample.

graph = tf.Graph()

with graph.as_default():

    # Input data.
    train_dataset = 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)

    # Variables.
    embeddings = tf.Variable(
                        tf.random_uniform([vocabulary_size, embedding_size], -1.0, 1.0))
    softmax_weights = tf.Variable(
                        tf.truncated_normal([vocabulary_size, embedding_size],
                                            stddev=1.0 / math.sqrt(embedding_size)))
    softmax_biases = tf.Variable(tf.zeros([vocabulary_size]))

    # Model.
    # Look up embeddings for inputs.
    embed = tf.nn.embedding_lookup(embeddings, train_dataset)
    # Compute the softmax loss, using a sample of the negative labels each time.
    loss = tf.reduce_mean(
                        tf.nn.sampled_softmax_loss(softmax_weights, softmax_biases, embed,
                                                    train_labels, num_sampled, vocabulary_size))

    # Optimizer.
    # Note: The optimizer will optimize the softmax_weights AND the embeddings.
    # This is because the embeddings are defined as a variable quantity and the
    # optimizer's `minimize` method will by default modify all variable quantities 
    # that contribute to the tensor it is passed.
    # See docs on `tf.train.Optimizer.minimize()` for more details.
    optimizer = tf.train.AdagradOptimizer(1.0).minimize(loss)

    # Compute the similarity between minibatch examples and all embeddings.
    # We use the cosine distance:
    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, tf.transpose(normalized_embeddings))
    embeddings_2 = (normalized_embeddings + softmax_weights)/2.0
    norm_ = tf.sqrt(tf.reduce_sum(tf.square(embeddings_2), 1, keep_dims=True))
    normalized_embeddings_2 = embeddings_2 / norm_


num_steps = 100001

with tf.Session(graph=graph) as session:
    if int(tf.VERSION.split('.')[1]) > 11:
        tf.global_variables_initializer().run()
    else:
        tf.initialize_all_variables().run()
    print('Initialized')

    average_loss = 0
    for step in range(num_steps):
        batch_data, batch_labels = generate_batch(batch_size, num_skips, skip_window)
        feed_dict = {train_dataset : batch_data, train_labels : batch_labels}
        _, l = session.run([optimizer, loss], feed_dict=feed_dict)
        average_loss += l
        if step % 2000 == 0:
            if step > 0:
                average_loss = average_loss / 2000
            # The average loss is an estimate of the loss over the last 2000 batches.
            print('Average loss at step %d: %f' % (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 range(valid_size):
                valid_word = reverse_dictionary[valid_examples[i]]
                top_k = 8 # number of nearest neighbors
                nearest = (-sim[i, :]).argsort()[1:top_k+1]  # let alone itself, so begin with 1
                log = 'Nearest to %s:' % valid_word
                for k in range(top_k):
                    close_word = reverse_dictionary[nearest[k]]
                    log = '%s %s,' % (log, close_word)
                print(log)
    final_embeddings = normalized_embeddings.eval()
    final_embeddings_2 = normalized_embeddings_2.eval()  # this is better


num_points = 400

tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000)
two_d_embeddings = tsne.fit_transform(final_embeddings[1:num_points+1, :])
two_d_embeddings_2 = tsne.fit_transform(final_embeddings_2[1:num_points+1, :])

with open('2d_embedding_skip_gram.pkl', 'wb') as f:
    pickle.dump([two_d_embeddings, two_d_embeddings_2, reverse_dictionary], f)

#-*-编码：utf-8-*-
来自未来导入打印功能
导入集合
输入数学
将numpy作为np导入
随机输入
将tensorflow.compat.v1导入为tf
tf.disable_v2_behavior（）
从sklearn.manifold导入TSNE
进口泡菜
示例=打开（“/Users/henry/Desktop/Cor.txt”）
words=sample.read（）
打印（'数据大小%d'%len（字））
sample.close（）
词汇表大小=50000
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=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）
打印（‘最常见的单词（+UNK）’，计数[：5]）
打印（'样本数据'，数据[：10]）
del words#提示减少内存。
数据索引=0
def生成批次（批次大小、跳过数量、跳过窗口）：
全球数据索引
断言批次大小%num_skips==0#每个单词对都是一个批次，因此训练数据[context target context]会将批次数增加2。
assert num_跳过根据错误，问题是关于相互相乘的矩阵的维数。为了解决这个问题，矩阵的维数应该是
[1,128]，[128,64]
我得到了以下错误，当我查看API时，我发现“输入”之前的“标签”，首先是在代码输入中

tf.nn.nce_loss(
    weights, biases, labels, inputs, num_sampled, num_classes, num_true=1,
    sampled_values=None, remove_accidental_hits=False, name='nce_loss'
)

谢谢，如何设置矩阵的维数？您可以使用转置操作进行更改。[128，1]到[1128]我认为维度问题在这里：相似性=tf.matmul（有效的嵌入，tf.transpose（规范化的嵌入）），因此，您可以转置第一个，这是有效的嵌入，而不转置第二个。相似性=tf.matmul（tf.transpose（valid_embedding），normalized_embedding）我这样做相似性=tf.matmul（valid_embed，tf.transpose（softmax_weights））+softmax_对相似性的偏差=tf.matmul（tf.transpose（valid_embedding），softmax_weights）+softmax_偏差，但结果与ValueError相同：维度必须相等，但是对于输入形状为[128,1]，[64128]的“sampled_softmax_loss/MatMul”（op:'MatMul'），则为1和128。代码块中的包装错误
tf.nn.nce_loss(
    weights, biases, labels, inputs, num_sampled, num_classes, num_true=1,
    sampled_values=None, remove_accidental_hits=False, name='nce_loss'
)