RougeN度量
RougeN class
keras_nlp.metrics.RougeN(order=2, use_stemmer=False, name="rouge-n", **kwargs)
ROUGE-N metric.
This class implements the ROUGE-N variant of the ROUGE metric. The ROUGE-N metric is traditionally used for evaluating summarisation systems. Succinctly put, ROUGE-N is a score based on the number of matching n-grams between the reference text and the hypothesis text.
Note on input shapes:
For y_true and y_pred, this class supports scalar values and batch
inputs of shapes (), (batch_size,) and (batch_size, 1).
Arguments
- order: The order of n-grams which are to be matched. It should lie in
range [1, 9]. Defaults to
2. - use_stemmer: bool. Whether Porter Stemmer should be used to strip word
suffixes to improve matching. Defaults to
False. - dtype: string or tf.dtypes.Dtype. Precision of metric computation. If
not specified, it defaults to
"float32". - name: string. Name of the metric instance.
- **kwargs: Other keyword arguments.
References
Examples
- Python string.
>>> rouge_n = keras_nlp.metrics.RougeN(order=2)
>>> y_true = "the tiny little cat was found under the big funny bed"
>>> y_pred = "the cat was under the bed"
>>> rouge_n(y_true, y_pred)["f1_score"]
<tf.Tensor: shape=(), dtype=float32, numpy=0.26666668>
- List inputs.
>>> rouge_n = keras_nlp.metrics.RougeN(order=2)
>>> y_true = [
... "the tiny little cat was found under the big funny bed",
... "i really love contributing to KerasNLP",
... ]
>>> y_pred = [
... "the cat was under the bed",
... "i love contributing to KerasNLP",
... ]
>>> rouge_n(y_true, y_pred)["f1_score"]
<tf.Tensor: shape=(), dtype=float32, numpy=0.4666667>
- 2D inputs.
>>> rouge_n = keras_nlp.metrics.RougeN(order=2)
>>> y_true =[
... ["the tiny little cat was found under the big funny bed"],
... ["i really love contributing to KerasNLP"],
... ]
>>> y_pred =[
... ["the cat was under the bed"],
... ["i love contributing to KerasNLP"],
... ]
>>> rouge_n(y_true, y_pred)["f1_score"]
<tf.Tensor: shape=(), dtype=float32, numpy=0.4666667>
- Trigrams.
>>> rouge_n = keras_nlp.metrics.RougeN(order=3)
>>> y_true = [
... "the tiny little cat was found under the big funny bed",
... "i really love contributing to KerasNLP",
... ]
>>> y_pred = [
... "the cat was under the bed",
... "i love contributing to KerasNLP",
... ]
>>> rouge_n(y_true, y_pred)["f1_score"]
<tf.Tensor: shape=(), dtype=float32, numpy=0.2857143>