def _compute_precision_recall(input_layer, labels, threshold,
per_example_weights):
"""Returns the numerator of both, the denominator of precision and recall."""
# To apply per_example_weights, we need to collapse each row to a scalar, but
# we really want the sum.
labels.get_shape().assert_is_compatible_with(input_layer.get_shape())
relevant = tf.to_float(tf.greater(labels, 0))
retrieved = tf.to_float(tf.greater(input_layer, threshold))
selected = relevant * retrieved
if per_example_weights:
per_example_weights = tf.convert_to_tensor(per_example_weights,
name='per_example_weights')
if selected.get_shape().dims:
per_example_weights.get_shape().assert_is_compatible_with(
[selected.get_shape().dims[0]])
else:
per_example_weights.get_shape().assert_is_compatible_with([None])
per_example_weights = tf.to_float(tf.greater(per_example_weights, 0))
selected = functions.reduce_batch_sum(selected) * per_example_weights
relevant = functions.reduce_batch_sum(relevant) * per_example_weights
retrieved = functions.reduce_batch_sum(retrieved) * per_example_weights
sum_relevant = tf.reduce_sum(relevant)
sum_retrieved = tf.reduce_sum(retrieved)
selected = tf.reduce_sum(selected)
return selected, sum_retrieved, sum_relevant
def _batch_sum_bce(x, target, name='binary_cross_entropy'):
logits = functions.binary_cross_entropy_loss_with_logits(x,
target,
name=name)
if per_output_weights is not None:
logits *= per_output_weights
return functions.reduce_batch_sum(logits)
def _compute_precision_recall(input_, labels, threshold, per_example_weights):
"""Returns the numerator of both, the denominator of precision and recall."""
# To apply per_example_weights, we need to collapse each row to a scalar, but
# we really want the sum.
labels.get_shape().assert_is_compatible_with(input_.get_shape())
relevant = tf.to_float(tf.greater(labels, 0))
retrieved = tf.to_float(tf.greater(input_, threshold))
selected = relevant * retrieved
if per_example_weights is not None:
per_example_weights = _convert_and_assert_per_example_weights_compatible(
input_, per_example_weights, dtype=None)
per_example_weights = tf.to_float(tf.greater(per_example_weights, 0))
selected = functions.reduce_batch_sum(selected) * per_example_weights
relevant = functions.reduce_batch_sum(relevant) * per_example_weights
retrieved = functions.reduce_batch_sum(retrieved) * per_example_weights
sum_relevant = tf.reduce_sum(relevant)
sum_retrieved = tf.reduce_sum(retrieved)
selected = tf.reduce_sum(selected)
return selected, sum_retrieved, sum_relevant
def _compute_precision_recall(input_, labels, threshold, per_example_weights):
"""Returns the numerator of both, the denominator of precision and recall."""
# To apply per_example_weights, we need to collapse each row to a scalar, but
# we really want the sum.
labels.get_shape().assert_is_compatible_with(input_.get_shape())
relevant = tf.to_float(tf.greater(labels, 0))
retrieved = tf.to_float(tf.greater(input_, threshold))
selected = relevant * retrieved
if per_example_weights is not None:
per_example_weights = _convert_and_assert_per_example_weights_compatible(
input_, per_example_weights, dtype=None
)
per_example_weights = tf.to_float(tf.greater(per_example_weights, 0))
selected = functions.reduce_batch_sum(selected) * per_example_weights
relevant = functions.reduce_batch_sum(relevant) * per_example_weights
retrieved = functions.reduce_batch_sum(retrieved) * per_example_weights
sum_relevant = tf.reduce_sum(relevant)
sum_retrieved = tf.reduce_sum(retrieved)
selected = tf.reduce_sum(selected)
return selected, sum_retrieved, sum_relevant
def _batch_sum_bce(x, target, name='binary_cross_entropy'):
return functions.reduce_batch_sum(
functions.binary_cross_entropy_loss_with_logits(x, target, name=name))
def _batch_sum_bce(x, target, name="binary_cross_entropy"):
logits = functions.binary_cross_entropy_loss_with_logits(x, target, name=name)
if per_output_weights is not None:
logits *= per_output_weights
return functions.reduce_batch_sum(logits)
