检查语言模型的复杂度

oku*_*oub 2 nlp language-model lstm keras perplexity

我使用 Keras LSTM 创建了一个语言模型,现在我想评估它是否良好,因此我想计算困惑度。

在 Python 中计算模型的复杂度的最佳方法是什么?

Ale*_*lex 5

我已经提出了两个版本并附上了相应的来源,请随时查看链接。

def perplexity_raw(y_true, y_pred):
    """
    The perplexity metric. Why isn't this part of Keras yet?!
    /sf/ask/2931691591/
    https://github.com/keras-team/keras/issues/8267
    """
#     cross_entropy = K.sparse_categorical_crossentropy(y_true, y_pred)
    cross_entropy = K.cast(K.equal(K.max(y_true, axis=-1),
                          K.cast(K.argmax(y_pred, axis=-1), K.floatx())),
                  K.floatx())
    perplexity = K.exp(cross_entropy)
    return perplexity

def perplexity(y_true, y_pred):
    """
    The perplexity metric. Why isn't this part of Keras yet?!
    /sf/ask/2931691591/
    https://github.com/keras-team/keras/issues/8267
    """
    cross_entropy = K.sparse_categorical_crossentropy(y_true, y_pred)
    perplexity = K.exp(cross_entropy)
    return perplexity
Run Code Online (Sandbox Code Playgroud)