代码之家 › 专栏 › 技术社区 › Swamy

如何使用lstm为给定句子分配发生概率的语言模型

deep-learning keras neural-network machine-learning

Swamy · 技术社区 · 6 年前

目前,我正在使用三角学来做这件事。它指定给定句子出现的概率。但仅限于两个词的语境。但LSTM可以做得更多。那么,如何建立一个lstm模型来分配给定句子的出现概率呢?

1 回复 | 直到 6 年前

rvinas 5 年前

我刚刚编写了一个非常简单的示例,演示如何使用lstm模型计算句子出现的概率。完整的代码可以找到 here 是的。

鹅妈妈童谣集 在伦敦大约1765年):

# Data
data = ["Two little dicky birds",
        "Sat on a wall,",
        "One called Peter,",
        "One called Paul.",
        "Fly away, Peter,",
        "Fly away, Paul!",
        "Come back, Peter,",
        "Come back, Paul."]

首先,让我们用 keras.preprocessing.text.Tokenizer 要创建词汇表并标记句子:

# Preprocess data
tokenizer = Tokenizer()
tokenizer.fit_on_texts(data)
vocab = tokenizer.word_index
seqs = tokenizer.texts_to_sequences(data)

我们的模型将以一系列单词作为输入(上下文),并在给定上下文的情况下输出每个单词在词汇表中的条件概率分布。为此,我们通过填充序列并在序列上滑动窗口来准备训练数据:

def prepare_sentence(seq, maxlen):
    # Pads seq and slides windows
    x = []
    y = []
    for i, w in enumerate(seq):
        x_padded = pad_sequences([seq[:i]],
                                 maxlen=maxlen - 1,
                                 padding='pre')[0]  # Pads before each sequence
        x.append(x_padded)
        y.append(w)
    return x, y

# Pad sequences and slide windows
maxlen = max([len(seq) for seq in seqs])
x = []
y = []
for seq in seqs:
    x_windows, y_windows = prepare_sentence(seq, maxlen)
    x += x_windows
    y += y_windows
x = np.array(x)
y = np.array(y) - 1  # The word <PAD> does not constitute a class
y = np.eye(len(vocab))[y]  # One hot encoding

我决定为每一节分别滑动窗口,但这可以做得不同。

接下来,我们用keras定义并训练一个简单的lstm模型。该模型由嵌入层、lstm层和具有softmax激活的密集层(使用lstm最后一个时间步的输出生成给定上下文的词汇表中每个单词的概率)组成:

# Define model
model = Sequential()
model.add(Embedding(input_dim=len(vocab) + 1,  # vocabulary size. Adding an
                                               # extra element for <PAD> word
                    output_dim=5,  # size of embeddings
                    input_length=maxlen - 1))  # length of the padded sequences
model.add(LSTM(10))
model.add(Dense(len(vocab), activation='softmax'))
model.compile('rmsprop', 'categorical_crossentropy')

# Train network
model.fit(x, y, epochs=1000)

联合概率 P(w_1, ..., w_n) 句子的出现 w_1 ... w_n 可使用条件概率规则计算:

P(w_1, ..., w_n)=P(w_1)*P(w_2|w_1)*...*P(w_n|w_{n-1}, ..., w_1)

其中每个条件概率由lstm模型给出。请注意,它们可能非常小,因此在日志空间中工作以避免数值不稳定性问题是明智的。总而言之:

# Compute probability of occurence of a sentence
sentence = "One called Peter,"
tok = tokenizer.texts_to_sequences([sentence])[0]
x_test, y_test = prepare_sentence(tok, maxlen)
x_test = np.array(x_test)
y_test = np.array(y_test) - 1  # The word <PAD> does not constitute a class
p_pred = model.predict(x_test)  # array of conditional probabilities
vocab_inv = {v: k for k, v in vocab.items()}

#Â Compute product
# Efficient version: np.exp(np.sum(np.log(np.diag(p_pred[:, y_test]))))
log_p_sentence = 0
for i, prob in enumerate(p_pred):
    word = vocab_inv[y_test[i]+1]  # Index 0 from vocab is reserved to <PAD>
    history = ' '.join([vocab_inv[w] for w in x_test[i, :] if w != 0])
    prob_word = prob[y_test[i]]
    log_p_sentence += np.log(prob_word)
    print('P(w={}|h={})={}'.format(word, history, prob_word))
print('Prob. sentence: {}'.format(np.exp(log_p_sentence)))

注意 :这是一个非常小的玩具数据集,我们可能过于适合。