def __init__(self,
name=None,
topn=None,
dtype=None,
ragged=False,
**kwargs):
"""Constructor.
Args:
name: A string used as the name for this metric.
topn: A cutoff for how many examples to consider for this metric.
dtype: Data type of the metric output. See `tf.keras.metrics.Metric`.
ragged: A bool indicating whether the supplied tensors are ragged. If
True y_true, y_pred and sample_weight (if providing per-example weights)
need to be ragged tensors with compatible shapes.
**kwargs: Other keyward arguments used in `tf.keras.metrics.Metric`.
"""
super(PrecisionIAMetric, self).__init__(name=name,
dtype=dtype,
ragged=ragged,
**kwargs)
self._topn = topn
self._metric = metrics_impl.PrecisionIAMetric(name=name,
topn=topn,
ragged=ragged)
def __init__(self,
topn=None,
dtype=None,
name="precision_ia_metric",
**kwargs):
super(PrecisionIAMetric, self).__init__(name=name, dtype=dtype, **kwargs)
self._topn = topn
self._metric = metrics_impl.PrecisionIAMetric(name=name, topn=topn)
def __init__(self,
name=None,
topn=None,
dtype=None,
ragged=False,
**kwargs):
super(PrecisionIAMetric, self).__init__(name=name,
dtype=dtype,
**kwargs)
self._topn = topn
self._metric = metrics_impl.PrecisionIAMetric(name=name,
topn=topn,
ragged=ragged)
def precision_ia(labels, predictions, weights=None, topn=None, name=None):
"""Computes Intent-Aware Precision as weighted average of relevant examples.
Args:
labels: A `Tensor` with shape [batch_size, list_size, subtopic_size]. A
nonzero value means that the example covers the corresponding subtopic.
predictions: A `Tensor` with shape [batch_size, list_size]. Each value is
the ranking score of the corresponding example.
weights: A `Tensor` of the same shape of predictions or [batch_size, 1]. The
former case is per-example and the latter case is per-list.
topn: A cutoff for how many examples to consider for this metric.
name: A string used as the name for this metric.
Returns:
A metric for the weighted precision of the batch.
"""
metric = metrics_impl.PrecisionIAMetric(name, topn)
with tf.compat.v1.name_scope(metric.name, 'precision_ia',
(labels, predictions, weights)):
precision_at_k, per_list_weights = metric.compute(
labels, predictions, weights)
return tf.compat.v1.metrics.mean(precision_at_k, per_list_weights)
