o siZ@sdZddlZddlmZddlZddlZddZddZd d Z dd d Z d ddZ dddZ  d!ddZ ddZd"ddZd"ddZddZdS)#a Pattern discovery involves the identification of musical patterns (i.e. short fragments or melodic ideas that repeat at least twice) both from audio and symbolic representations. The metrics used to evaluate pattern discovery systems attempt to quantify the ability of the algorithm to not only determine the present patterns in a piece, but also to find all of their occurrences. Based on the methods described here: T. Collins. MIREX task: Discovery of repeated themes & sections. http://www.music-ir.org/mirex/wiki/2013:Discovery_of_Repeated_Themes_&_Sections, 2013. Conventions ----------- The input format can be automatically generated by calling :func:`mir_eval.io.load_patterns`. This format is a list of a list of tuples. The first list collections patterns, each of which is a list of occurrences, and each occurrence is a list of MIDI onset tuples of ``(onset_time, mid_note)`` A pattern is a list of occurrences. The first occurrence must be the prototype of that pattern (i.e. the most representative of all the occurrences). An occurrence is a list of tuples containing the onset time and the midi note number. Metrics ------- * :func:`mir_eval.pattern.standard_FPR`: Strict metric in order to find the possibly transposed patterns of exact length. This is the only metric that considers transposed patterns. * :func:`mir_eval.pattern.establishment_FPR`: Evaluates the amount of patterns that were successfully identified by the estimated results, no matter how many occurrences they found. In other words, this metric captures how the algorithm successfully *established* that a pattern repeated at least twice, and this pattern is also found in the reference annotation. * :func:`mir_eval.pattern.occurrence_FPR`: Evaluation of how well an estimation can effectively identify all the occurrences of the found patterns, independently of how many patterns have been discovered. This metric has a threshold parameter that indicates how similar two occurrences must be in order to be considered equal. In MIREX, this evaluation is run twice, with thresholds .75 and .5. * :func:`mir_eval.pattern.three_layer_FPR`: Aims to evaluate the general similarity between the reference and the estimations, combining both the establishment of patterns and the retrieval of its occurrences in a single F1 score. * :func:`mir_eval.pattern.first_n_three_layer_P`: Computes the three-layer precision for the first N patterns only in order to measure the ability of the algorithm to sort the identified patterns based on their relevance. * :func:`mir_eval.pattern.first_n_target_proportion_R`: Computes the target proportion recall for the first N patterns only in order to measure the ability of the algorithm to sort the identified patterns based on their relevance. N)utilcCstdd|DS)a$Compute the number of onset_midi objects in a pattern Parameters ---------- patterns A list of patterns using the format returned by :func:`mir_eval.io.load_patterns()` Returns ------- n_onsets : int Number of onsets within the pattern. cSs$g|]}|D] }|D]}|q qqSr).0patocco_mrrD/home/ubuntu/.local/lib/python3.10/site-packages/mir_eval/pattern.py Ns$z!_n_onset_midi..)len)patternsrrr _n_onset_midi?sr cCst|dkr tdt|dkrtd||fD]%}|D] }t|dkr*td|D]}|D] }t|dkrz+_occurrence_intersection..cSrrrrrrr rrr)occ_Pocc_Qset_Pset_Qrrr _occurrence_intersectionqsr"cardinality_scorec Cstt|t|f}t|D]/\}}t|D]&\}}|dkr:ttt|t|g}tt||||||f<qtdq|S)aaCompute the score matrix between the patterns P and Q. Parameters ---------- P : list Pattern containing a list of occurrences. Q : list Pattern containing a list of occurrences. similarity_metric : str A string representing the metric to be used when computing the similarity matrix. Accepted values: - "cardinality_score": Count of the intersection between occurrences. (Default value = "cardinality_score") Returns ------- sm : np.array The score matrix between P and Q using the similarity_metric. r#zc Cst||t|}t|}d}t|dkst|dkrdS|D]D}t|d}|D]8}t|d} t|t| kr>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> F, P, R = mir_eval.pattern.standard_FPR(ref_patterns, est_patterns) Parameters ---------- reference_patterns : list The reference patterns using the format returned by :func:`mir_eval.io.load_patterns()` estimated_patterns : list The estimated patterns in the same format tol : float Tolerance level when comparing reference against estimation. Default parameter is the one found in the original matlab code by Tom Collins used for MIREX 2013. (Default value = 1e-5) Returns ------- f_measure : float The standard F1 Score precision : float The standard Precision recall : float The standard Recall rr3r3raxis) rr r r$asarrayr(absdiffr'r f_measure) rrtolnPnQk ref_patternr) est_patternr* precisionrecallr9rrr standard_FPRs* %@    rBcCst||t|}t|}t||f}t|dks t|dkr"dSt|D]\}}t|D]\}} t|| |} t| |||f<q.q&ttj|dd} ttj|dd} t | | } | | | fS)a;Compute the establishment F1 Score, Precision and Recall. Examples -------- >>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> F, P, R = mir_eval.pattern.establishment_FPR(ref_patterns, ... est_patterns) Parameters ---------- reference_patterns : list The reference patterns in the format returned by :func:`mir_eval.io.load_patterns()` estimated_patterns : list The estimated patterns in the same format similarity_metric : str A string representing the metric to be used when computing the similarity matrix. Accepted values: - "cardinality_score": Count of the intersection between occurrences. (Default value = "cardinality_score") Returns ------- f_measure : float The establishment F1 Score precision : float The establishment Precision recall : float The establishment Recall rr2r4r) rr r$r%r r&r0r(meanrr9)rrr+r;r<Sr-r>r.r?sr@rAr9rrr establishment_FPRs (   rF?c Cst||t|}t|}t||df}tjdtd}t|dks(t|dkr*dSt|D]A\}} t|D]8\} } t| | |} t | |krnt tj | dd||| df<t tj | dd||| df<t ||| gf}q6q.t|dkr{d} d}nN|dddddf}t tj |t |dddf|dddfdd} |dddddf}t tj |t |dddf|dddfdd}t | |}|| |fS) aCompute the occurrence F1 Score, Precision and Recall. Examples -------- >>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> F, P, R = mir_eval.pattern.occurrence_FPR(ref_patterns, ... est_patterns) Parameters ---------- reference_patterns : list The reference patterns in the format returned by :func:`mir_eval.io.load_patterns()` estimated_patterns : list The estimated patterns in the same format thres : float How similar two occurrences must be in order to be considered equal (Default value = .75) similarity_metric : str A string representing the metric to be used when computing the similarity matrix. Accepted values: - "cardinality_score": Count of the intersection between occurrences. (Default value = "cardinality_score") Returns ------- f_measure : float The occurrence F1 Score precision : float The occurrence Precision recall : float The occurrence Recall r)rr)dtyperr2r4rN)rr r$r%emptyintr r&r0r(rCvstackix_rr9)rrthresr+r;r<O_PRrel_idxr-r>r.r?rEr@rAr)Rr9rrr occurrence_FPR+s2 /  88  rQcst||ddfdddfdd t|dks#t|dkr%d S||d d }ttj|dd }ttj|dd }t||}|||fS)aThree Layer F1 Score, Precision and Recall. As described by Meridith. Examples -------- >>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> F, P, R = mir_eval.pattern.three_layer_FPR(ref_patterns, ... est_patterns) Parameters ---------- reference_patterns : list The reference patterns in the format returned by :func:`mir_eval.io.load_patterns()` estimated_patterns : list The estimated patterns in the same format Returns ------- f_measure : float The three-layer F1 Score precision : float The three-layer Precision recall : float The three-layer Recall cSs6tt||}|tt|}|tt|}||fS)a3Compute the first layer Precision and Recall values given the set of occurrences in the reference and the set of occurrences in the estimation. Parameters ---------- ref_occs est_occs Returns ------- precision recall )r r"r')ref_occsest_occsrEr@rArrr compute_first_layer_PRsz/three_layer_FPR..compute_first_layer_PRcs:||}ttj|dd}ttj|dd}||fS)a:Compute the second layer Precision and Recall values given the set of occurrences in the reference and the set of occurrences in the estimation. Parameters ---------- ref_pattern est_pattern Returns ------- precision recall rr4r)r$rCr()r>r?F_1r@rA) compute_layerrr compute_second_layer_PRs z0three_layer_FPR..compute_second_layer_PRrc s|dkr|dkrtd|t|}t|}t||f}t|D]+}t|D]$}|dkr0}n|dkr6}|||||\} } t| | |||f<q'q!|S)aCompute the F-measure matrix for a given layer. The reference and estimated elements can be either patterns or occurrences, depending on the layer. For layer 1, the elements must be occurrences. For layer 2, the elements must be patterns. Parameters ---------- ref_elements est_elements layer (Default value = 1) Returns ------- F : F-measure for the given layer rrz-Layer (%d) must be an integer between 1 and 2)rr r$r%rangerr9) ref_elements est_elementslayerr;r<Fr-r.funcr@rA)rTrWrr rVs    z&three_layer_FPR..compute_layerrr2r)r[r4N)r)rr r$rCr(rr9)rrF_2 precision_3recall_3 f_measure_3r)rTrVrWr three_layer_FPR~s  &  rbcCsPt||t|dkst|dkrdS|dtt||}t||\}}}|S)aFirst n three-layer precision. This metric is basically the same as the three-layer FPR but it is only applied to the first n estimated patterns, and it only returns the precision. In MIREX and typically, n = 5. Examples -------- >>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> P = mir_eval.pattern.first_n_three_layer_P(ref_patterns, ... est_patterns, n=5) Parameters ---------- reference_patterns : list The reference patterns in the format returned by :func:`mir_eval.io.load_patterns()` estimated_patterns : list The estimated patterns in the same format n : int Number of patterns to consider from the estimated results, in the order they appear in the matrix (Default value = 5) Returns ------- precision : float The first n three-layer Precision rr2N)rr minr rbrrnfn_est_patternsr\r)rPrrr first_n_three_layer_Ps rhcCsPt||t|dkst|dkrdS|dtt||}t||\}}}|S)aFirst n target proportion establishment recall metric. This metric is similar is similar to the establishment FPR score, but it only takes into account the first n estimated patterns and it only outputs the Recall value of it. Examples -------- >>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> R = mir_eval.pattern.first_n_target_proportion_R( ... ref_patterns, est_patterns, n=5) Parameters ---------- reference_patterns : list The reference patterns in the format returned by :func:`mir_eval.io.load_patterns()` estimated_patterns : list The estimated patterns in the same format n : int Number of patterns to consider from the estimated results, in the order they appear in the matrix. (Default value = 5) Returns ------- recall : float The first n target proportion Recall. rr2N)rr rdr rFrerrr first_n_target_proportion_R*s ricKs(t}tjt||fi|\|d<|d<|d<tjt||fi|\|d<|d<|d<d|d<tjt||fi|\|d <|d <|d <d |d<tjt||fi|\|d <|d<|d<tjt||fi|\|d<|d<|d<d|vrxd|d<tjt||fi||d<tjt ||fi||d<|S)a$Load data and perform the evaluation. Examples -------- >>> ref_patterns = mir_eval.io.load_patterns("ref_pattern.txt") >>> est_patterns = mir_eval.io.load_patterns("est_pattern.txt") >>> scores = mir_eval.pattern.evaluate(ref_patterns, est_patterns) Parameters ---------- ref_patterns : list The reference patterns in the format returned by :func:`mir_eval.io.load_patterns()` est_patterns : list The estimated patterns in the same format **kwargs Additional keyword arguments which will be passed to the appropriate metric or preprocessing functions. Returns ------- scores : dict Dictionary of scores, where the key is the metric name (str) and the value is the (float) score achieved. r\r)rPF_estP_estR_estg?threshzF_occ.5zP_occ.5zR_occ.5rGzF_occ.75zP_occ.75zR_occ.75F_3P_3R_3rfrcFFPFFTP_est) collections OrderedDictr filter_kwargsrBrFrQrbrhri) ref_patterns est_patternskwargsscoresrrr evaluateUsR  rz)r#)r1)rGr#)rc)__doc__numpyr$rrrsr rr"r0rBrFrQrbrhrirzrrrr s&8   %D @ S  - +