Ngf!vdZddlZddlmZddlmZdefdZejj Gdd e Z dS) a'Argmax matcher implementation. This class takes a similarity matrix and matches columns to rows based on the maximum value per column. One can specify matched_thresholds and to prevent columns from matching to rows (generally resulting in a negative training example) and unmatched_theshold to ignore the match (generally resulting in neither a positive or negative training example). This matcher is used in Fast(er)-RCNN. Note: matchers are used in TargetAssigners. There is a create_target_assigner factory function for popular implementations. N)Match)Optional num_classesctj|d||jtj}|d|ddS)Nr)devicedtyper)torchzerossizerboolscatter_ unsqueeze)xronehots b/var/www/html/ai-engine/env/lib/python3.11/site-packages/effdet/object_detection/argmax_matcher.py one_hot_boolr!sH [K S S SF ??1akk!nna 0 00c \eZdZdZ ddedeededefd Zd Zd Z d Z d e fdZ dS) ArgMaxMatcheraMatcher based on highest value. This class computes matches from a similarity matrix. Each column is matched to a single row. To support object detection target assignment this class enables setting both matched_threshold (upper threshold) and unmatched_threshold (lower thresholds) defining three categories of similarity which define whether examples are positive, negative, or ignored: (1) similarity >= matched_threshold: Highest similarity. Matched/Positive! (2) matched_threshold > similarity >= unmatched_threshold: Medium similarity. Depending on negatives_lower_than_unmatched, this is either Unmatched/Negative OR Ignore. (3) unmatched_threshold > similarity: Lowest similarity. Depending on flag negatives_lower_than_unmatched, either Unmatched/Negative OR Ignore. For ignored matches this class sets the values in the Match object to -2. NTFmatched_thresholdunmatched_thresholdnegatives_lower_than_unmatchedforce_match_for_each_rowc ||td||_d|_|||_n||krtd||_|s+|j|jkrtd|j|j||_||_dS)aTConstruct ArgMaxMatcher. Args: matched_threshold: Threshold for positive matches. Positive if sim >= matched_threshold, where sim is the maximum value of the similarity matrix for a given column. Set to None for no threshold. unmatched_threshold: Threshold for negative matches. Negative if sim < unmatched_threshold. Defaults to matched_threshold when set to None. negatives_lower_than_unmatched: Boolean which defaults to True. If True then negative matches are the ones below the unmatched_threshold, whereas ignored matches are in between the matched and unmatched threshold. If False, then negative matches are in between the matched and unmatched threshold, and everything lower than unmatched is ignored. force_match_for_each_row: If True, ensures that each row is matched to at least one column (which is not guaranteed otherwise if the matched_threshold is high). Defaults to False. See argmax_matcher_test.testMatcherForceMatch() for an example. Raises: ValueError: if unmatched_threshold is set but matched_threshold is not set or if unmatched_threshold > matched_threshold. NzINeed to also define matched_threshold when unmatched_threshold is definedgzEunmatched_threshold needs to be smaller or equal to matched_thresholdzzWhen negatives are in between matched and unmatched thresholds, these cannot be of equal value. matched: %s, unmatched: %s) ValueError_matched_threshold_unmatched_threshold_force_match_for_each_row_negatives_lower_than_unmatched)selfrrrrs r__init__zArgMaxMatcher.__init__;s8  %,?,Khii i"3+-!  &(9D % %"%666 !hiii(;D %- U(D,CCC "X!%!8$:SUUU*B&/M,,,rcjdtj|jdtj|jzS)aKPerforms matching when the rows of similarity matrix are empty. When the rows are empty, all detections are false positives. So we return a tensor of -1's to indicate that the columns do not match to any rows. Returns: matches: int32 tensor indicating the row each column matches to. r)r r)r onesshapelongrr!similarity_matrixs r_match_when_rows_are_emptyz(ArgMaxMatcher._match_when_rows_are_emptyis/EJ06q9TeTlmmmmmrctj|d\}}|j|j|k}||jk|j|kz}|jr/|||d}|||d}n.|||d}|||d}|jr`tj|d}t||j d}tj|d\}} tj || |} | S|S)zPerforms matching when the rows of similarity matrix are non empty. Returns: matches: int32 tensor indicating the row each column matches to. rNr$r) r maxrrr _set_values_using_indicatorrargmaxrr&where) r!r) matched_valsmatchesbelow_unmatched_thresholdbetween_thresholdsforce_match_column_idsforce_match_column_indicatorsforce_match_column_maskforce_match_row_ids final_matchess r_match_when_rows_are_non_emptyz,ArgMaxMatcher._match_when_rows_are_non_emptyts;!& *;Q ? ? g  " .(,(AL(P %".$2K"K"&"9L"H"J 3 \::7D]_abb::7DVXZ[[::7D]_abb::7DVXZ[[  ) %*\2CQ%G%G ",89OQbQhijQk,l,l );@9Ebde;f;f 8 #%8!K(?ATV]^^M Nrc|jddkr"t||St||S)a Tries to match each column of the similarity matrix to a row. Args: similarity_matrix: tensor of shape [N, M] representing any similarity metric. Returns: Match object with corresponding matches for each of M columns. r)r&rr*r:r(s rmatchzArgMaxMatcher.matchsQ  "1 % * *889JKKLL L<<=NOOPP PrvalcT||j}|d|z z||zzS)zSet the indicated fields of x to val. Args: x: tensor. indicator: boolean with same shape as x. val: scalar with value to set. Returns: modified tensor. )r r)tor )r!r indicatorr=s rr.z)ArgMaxMatcher._set_values_using_indicators1LLqwL// A M"S9_44r)NTF) __name__ __module__ __qualname____doc__floatrr r"r*r:r<intr.rrrr's(9=8<27 ,N,N$),N&.uo,N26,N,0 ,N,N,N,N\ n n n@ Q Q Q 5S 5 5 5 5 5 5rr) rDr matcherrtypingrrFrjitscriptobjectrrGrrrMs   11111 F5F5F5F5F5FF5F5F5F5F5r