|View source on GitHub|
Computes recall@k of top-k predictions with respect to sparse labels.
tf.contrib.metrics.sparse_recall_at_top_k( labels, top_k_predictions, class_id=None, weights=None, metrics_collections=None, updates_collections=None, name=None )
class_id is specified, we calculate recall by considering only the
entries in the batch for which
class_id is in the label, and computing
the fraction of them for which
class_id is in the top-k
class_id is not specified, we'll calculate recall as how often on
average a class among the labels of a batch entry is in the top-k
sparse_recall_at_top_k creates two local variables,
false_negative_at_<k>, that are used to compute the recall_at_k
frequency. This frequency is ultimately returned as
idempotent operation that simply divides
true_positive_at_<k> by total
For estimation of the metric over a stream of data, the function creates an
update_op operation that updates these variables and returns the
recall_at_<k>. Set operations applied to
labels calculate the
true positives and false negatives weighted by
false_negative_at_<k> using these
None, weights default to 1. Use weights of 0 to mask values.
SparseTensorwith shape [D1, ... DN, num_labels], where N >= 1 and num_labels is the number of target classes for the associated prediction. Commonly, N=1 and
labelshas shape [batch_size, num_labels]. [D1, ... DN] must match
top_k_predictions. Values should be in range [0, num_classes), where num_classes is the last dimension of
predictions. Values outside this range always count towards
Tensorwith shape [D1, ... DN, k] where N >= 1. Commonly, N=1 and top_k_predictions has shape [batch size, k]. The final dimension contains the indices of top-k labels. [D1, ... DN] must match
class_id: Integer class ID for which we want binary metrics. This should be in range [0, num_classes), where num_classes is the last dimension of
predictions. If class_id is outside this range, the method returns NAN.
Tensorwhose rank is either 0, or n-1, where n is the rank of
labels. If the latter, it must be broadcastable to
labels(i.e., all dimensions must be either
1, or the same as the corresponding
metrics_collections: An optional list of collections that values should be added to.
updates_collections: An optional list of collections that updates should be added to.
name: Name of new update operation, and namespace for other dependent ops.
Tensorwith the value of
true_positivesdivided by the sum of
false_negativesvariables appropriately, and whose value matches
Noneand its shape doesn't match
predictions, or if either
updates_collectionsare not a list or tuple.