diff options
author | Daniel Friesel <daniel.friesel@uos.de> | 2021-01-18 14:27:37 +0100 |
---|---|---|
committer | Daniel Friesel <daniel.friesel@uos.de> | 2021-01-18 14:27:37 +0100 |
commit | 72ff3b91864eb8afe44ff6e049ba591c84b30781 (patch) | |
tree | 466d548b8333a6a408a28f48fe52fe355b3f8db6 /lib/lennart | |
parent | a561a91f16ac97a8cabc04a5e99be277e0351982 (diff) |
energytrace drift compensation: allow arbitrarily long skips
Diffstat (limited to 'lib/lennart')
-rw-r--r-- | lib/lennart/DataProcessor.py | 62 |
1 files changed, 33 insertions, 29 deletions
diff --git a/lib/lennart/DataProcessor.py b/lib/lennart/DataProcessor.py index 1fea0db..589dde2 100644 --- a/lib/lennart/DataProcessor.py +++ b/lib/lennart/DataProcessor.py @@ -161,7 +161,7 @@ class DataProcessor: self.sync_timestamps[4:-8] = with_drift_compensation except ValueError: logger.error( - f"Iteration #{self.offline_index}: drift-compensated sequence is too short" + f"Iteration #{self.offline_index}: drift-compensated sequence is too short ({len(with_drift_compensation)}/{len(self.sync_timestamps[4:-8])-1})" ) raise @@ -240,12 +240,16 @@ class DataProcessor: et_timestamps_start : et_timestamps_end + 1 ] candidate_weight = dict() - if os.getenv("DFATOOL_DRIFT_COMPENSATION_PENALTY"): + if 0: + penalties = (None,) + elif os.getenv("DFATOOL_DRIFT_COMPENSATION_PENALTY"): penalties = (int(os.getenv("DFATOOL_DRIFT_COMPENSATION_PENALTY")),) else: penalties = (1, 2, 5, 10, 15, 20) for penalty in penalties: - for changepoint in pelt.get_changepoints(energy_data, penalty=penalty): + for changepoint in pelt.get_changepoints( + energy_data, penalty=penalty, num_changepoints=1 + ): if changepoint in candidate_weight: candidate_weight[changepoint] += 1 else: @@ -296,6 +300,12 @@ class DataProcessor: transition_by_node = dict() compensated_timestamps = list() + # up to two nodes may be skipped + max_skip_count = 2 + + if os.getenv("DFATOOL_DC_MAX_SKIP"): + max_skip_count = int(os.getenv("DFATOOL_DC_MAX_SKIP")) + for transition_index, candidates in enumerate( transition_start_candidate_weights ): @@ -342,32 +352,26 @@ class DataProcessor: for transition_index, candidates in enumerate( transition_start_candidate_weights ): - if transition_index < 2: - continue - for from_i, (_, from_drift, _) in enumerate( - transition_start_candidate_weights[transition_index - 2] - ): - for to_i, (_, to_drift, _) in enumerate(candidates): - # Penalize shortcut by the duration of one sample - # (~270 us) - edge_srcs.append( - nodes_by_transition_index[transition_index - 2][from_i] - ) - edge_dsts.append(nodes_by_transition_index[transition_index][to_i]) - csr_weights.append(np.abs(from_drift - to_drift) + 270e-6) - if transition_index < 3: - continue - for from_i, (_, from_drift, _) in enumerate( - transition_start_candidate_weights[transition_index - 3] - ): - for to_i, (_, to_drift, _) in enumerate(candidates): - # Penalize shortcut by the duration of one sample - # (~270 us) - edge_srcs.append( - nodes_by_transition_index[transition_index - 3][from_i] - ) - edge_dsts.append(nodes_by_transition_index[transition_index][to_i]) - csr_weights.append(np.abs(from_drift - to_drift) + 2 * 270e-6) + for skip_count in range(2, max_skip_count + 2): + if transition_index < skip_count: + continue + for from_i, (_, from_drift, _) in enumerate( + transition_start_candidate_weights[transition_index - skip_count] + ): + for to_i, (_, to_drift, _) in enumerate(candidates): + # Penalize shortcut by the duration of one sample + # (~270 us) + edge_srcs.append( + nodes_by_transition_index[transition_index - skip_count][ + from_i + ] + ) + edge_dsts.append( + nodes_by_transition_index[transition_index][to_i] + ) + csr_weights.append( + np.abs(from_drift - to_drift) + skip_count * 270e-6 + ) sm = scipy.sparse.csr_matrix( (csr_weights, (edge_srcs, edge_dsts)), shape=(new_node + 1, new_node + 1) |