def m_ndp_post(self, p_novel: np.ndarray, gt_novel: np.ndarray) -> Dict:
"""
m_ndp_post function.
Args:
p_novel: detection predictions for N videos (Dimension: N X 1)
gt_novel: ground truth detections for N videos (Dimension: N X 1)
Returns:
Dictionary containing detection performance post novelty.
"""
return M_ndp(p_novel, gt_novel, mode="post_novelty")
def m_ndp(self, p_novel: np.ndarray, gt_novel: np.ndarray) -> Dict:
"""
m_ndp function.
Args:
p_novel: detection predictions for N videos (Dimension: N X 1)
gt_novel: ground truth detections for N videos (Dimension: N X 1)
Returns:
Dictionary containing novelty detection performance over the test.
"""
return M_ndp(p_novel, gt_novel, mode="full_test")
def m_ndp_post(self, p_novel: np.ndarray, gt_novel: np.ndarray) -> Dict:
"""
m_ndp_post function.
See :func:`~sail-on-client.evaluate.transcription.DocumentTranscriptionMetrics.m_ndp`
with post_novelty. This computes from the first GT novel sample.
Args:
p_novel: detection predictions (Dimension: [img X novel])
gt_novel: ground truth detections (Dimension: [img X detection])
Returns:
Dictionary containing detection performance post novelty.
"""
return M_ndp(p_novel, gt_novel, mode="post_novelty")
def m_ndp_post(self, p_novel: np.ndarray, gt_novel: np.ndarray) -> Dict:
"""
Novelty Detection Performance Post Red Light. m_ndp_post function.
See :func:`~sail-on-client.evaluation.ImageClassificationMetrics.m_ndp` with
post_novelty. This computes from the first GT novel sample
Args:
p_novel: detection predictions (Dimension: [img X novel])
gt_novel: ground truth detections (Dimension: [img X detection])
Returns:
Dictionary containing detection performance post novelty.
"""
return M_ndp(p_novel, gt_novel, mode="post_novelty")
def m_ndp(self, p_novel: np.ndarray, gt_novel: np.ndarray) -> Dict:
"""
m_ndp function.
Novelty detection performance. The method computes per-sample novelty
detection performance over the entire test.
Args:
p_novel: detection predictions (Dimension: [img X novel])
gt_novel: ground truth detections (Dimension: [img X detection])
Returns:
Dictionary containing novelty detection performance over the test.
"""
return M_ndp(p_novel, gt_novel)
def m_ndp_pre(self, p_novel: np.ndarray, gt_novel: np.ndarray) -> Dict:
"""
Novelty Detection Performance Pre Red Light. m_ndp_pre function.
See :func:`~sail-on-client.evaluation.ImageClassificationMetrics.m_ndp` with
post_novelty. This computes to the first GT novel sample. It really isn't useful
and is just added for completion. Should always be 0 since no possible TP.
Args:
p_novel: detection predictions (Dimension: [img X novel])
gt_novel: ground truth detections (Dimension: [img X detection])
Returns:
Dictionary containing detection performance pre novelty.
"""
return M_ndp(p_novel, gt_novel, mode="pre_novelty")
def m_ndp(self,
p_novel: np.ndarray,
gt_novel: np.ndarray,
mode: str = "full_test") -> Dict:
"""
Novelty Detection Performance: Program Metric.
Novelty detection performance. The method computes per-sample novelty detection performance.
Args:
p_novel: detection predictions (Dimension: [img X novel])
Nx1 vector with each element corresponding to probability of it being novel
gt_novel: ground truth detections (Dimension: [img X detection])
Nx1 vector with each element 0 (not novel) or 1 (novel)
mode: the mode to compute the test. if 'full_test' computes on all test samples,
if 'post_novelty' computes from first GT novel sample. If 'pre_novelty', only calculate
before first novel sample.
Returns:
Dictionary containing novelty detection performance over the test.
"""
return M_ndp(p_novel, gt_novel, mode="full_test")