MnM Modules

• mnm.biqoefs ... Biquad coefficients
• mnm.biquad ... Biquad filtering
• mnm.delta ... Inter-frame regression
• mnm.diag ... Matrix diagonal
• mnm.dtw ... Dynamic Time Warping
• mnm.gmmem ... Expectation Maximization for Gaussian Mixture Models
• mnm.hist ... Histogram
• mnm.knn ... K-Nearest Neighbour search
• mnm.lu ... Lower-Upper decomposition
• mnm.mahalanobis ... Mahalanobis distance
• mnm.mean ... Mean filtering
• mnm.meanstd ... Mean and standard deviation
• mnm.median ... Median filtering
• mnm.minmax ... Minimum and maximum
• mnm.moments ... Statistical moments
• mnm.nmd ... Non-zero Matrix Decomposition
• mnm.nmf ... Non-zero Matrix Factorization
• mnm.obsprob ... Obsprob
• mnm.onepole ... Onepole filtering
• mnm.qr ... Orthogonal-Right decomposition
• mnm.stats ... Stats
• mnm.submat ... Sub-matrix
• mnm.sum ... Matrix sum
• mnm.svd ... Singular Value Decomposition
• mnm.transpose ... Matrix transposition
• mnm.viterbi ... Viterbi algorithm
• mnm.xdist2 ... Square of Euclidean distance
• mnm.xmul ... Matrix multiplication

---

mnm.biqoefs		Biquad coefficients
		Calculates biquad coefficients for various filter types.
		arguments: none attributes: mode <'lp'|'lowpass'|'hp'|'highpass'|'bpcskirt'|'resonant'|'bpcpeak'|'bandpass'|'peaking'|'peaknotch&#\ 039;|'notch'|'bs'|'bandstop'|'ap'|'allpass'|'ls'|'lowshelf'|'hs'|'highshelf': type> - filter type [lowpass] f0 <num: freq> - cutoff or centre frequency (default is half Nyquist frequency) unit <'nyquist'|'hz'|'hertz': unit> - set unit for freq as ratio to Nyquist frequency or in hz [nyquist] sr <num: sr> - sample rate for freq in Hertz [44100] q <num: q> - quality/resonance [1] qnorm <num: qnorm> - if qnorm = 1, divide q by 1./sqrt(2.) so as to get a monotonic filter response with q = 1. (instead of 1./sqrt(2.), which is the default with qnorm = 0) gain <num: gain> - linear gain [1] out <matrix: coeffs> - output biquad coefficients messages: postdoc - post external doc to console bang - output coefficients out <matrix: coeffs> - output biquad coefficients inlets: none outlets: 1 <matrix: coeffs> - output biquad coefficients

---

mnm.biquad		Biquad filtering
		Calculates biquad filtering over vectors (rows or columns) or stream of values (of any dimension).
		arguments: <num: input> [<num: filter>] - init input and filter size attributes: mode <'df1'|'df2t': struct> - set biquad structure [df1] dim <'col'|'row'|'auto'|'stream': mode> - set the dimension on which to operate [auto] out <matrix: values> - output filtered values outstate <matrix: state> - output state messages: postdoc - post external doc to console insize <num: input> [<num: filter>] - init input and filter size coefs <list: coeffs> - set the biquad coefficients getstate - get the biquad state clear - reset the internal state out <matrix: values> - output filtered values outstate <matrix: state> - output state inlets: 1 <matrix: coeffs> - input values outlets: 1 <matrix: values> - output filtered values 2 <matrix: state> - output state

---

mnm.delta		Inter-frame regression
		Calculates derivative of incoming matrices or vectors.
		arguments: 1 <num: input> [<num: filter>] - init input size 2 <num: filter>: init filter size attributes: insize <num: input> - set input size filtersize <num: filter> - set filter size inadddel <num: del> - add a delay to the delayed input norm <bool: normalize> - set normalization mode ([off] outdelayed <matrix: delayed> - output delayed inputs (in phase with deltas) out <matrix: delta> - output deltas outstate <matrix: state> - internal values messages: postdoc - post external doc to console clear - clear the memory of inputs getstate - get the internal weights vector getnorm - get the normalization factor getring - get the input ring buffer getdelay - get the filter delay outdelayed <matrix: delayed> - output delayed inputs (in phase with deltas) out <matrix: delta> - output deltas outstate <matrix: state> - internal values inlets: 1 <matrix: input> - multiply matrix with given right or left operand outlets: 1 <matrix: delayed> - output delayed inputs (in phase with deltas) 2 <matrix: delta> - output deltas 3 <matrix: state> - internal values

---

mnm.diag		Matrix diagonal
		Returns a copy of the diagonal of the incoming matrix in a vector.The length of the result is the minimum of the dimensions of the input.
		arguments: none attributes: out <vector: diagonal> - output diagonal vector messages: postdoc - post external doc to console out <vector: diagonal> - output diagonal vector inlets: 1 <matrix: input> - input matrix outlets: 1 <vector: diagonal> - output diagonal vector

---

mnm.dtw		Dynamic Time Warping
		Calculates DTW of incoming matrix or vector.
		arguments: 1 <matrix: op> - matrix to be used as right operand attributes: outa <matrix: A> - set A output matrix outb <matrix: B> - set B output matrix messages: postdoc - post external doc to console inlets: 1 <matrix: left> - left hand side matrix operand 2 <matrix: right> - right hand side matrix operand outlets: 1 <matrix: A> - A output matrix 2 <matrix: B> - B output matrix

---

mnm.gmmem		Expectation Maximization for Gaussian Mixture Models
		GMM EM has to be documented.
		arguments: 1 <num: num> - number of centers to use attributes: outcenters <matrix: centers> - centers output matrix outcov <matrix: covariance> - covariance output matrix outpriors <matrix: priors> - priors output matrix mode <diagonal|full|spherical: mode> covariance computation types criteria <num: criteria> - criteria ncenters <num: num> - number of centers messages: postdoc - post external doc to console inlets: 1 <matrix: input> - matrix of values outlets: 1 <matrix: centers> - centers output matrix 2 <matrix: covariance> - covariance output matrix 3 <matrix: priors> - priors output matrix

---

mnm.hist		Histogram
		Calculates histogram of incoming matrix elements. The input matrix, list or vector element's occurences are counted in the given number of bins in between the min and max value
		arguments: 1 <num: num> - number of bins attributes: out <vector: histogram> - output histogram vector bpf <matrix: histogram> - output two-column matrix with bin indices and histogram values norm <'off'|'max'|'sum': normalize> - normalise histogram so that max or sum is 1 messages: postdoc - post external doc to console out <vector: histogram> - output histogram vector set_n <num: num> - number of bins inlets: 1 <matrix: values> - intput matrix or list outlets: 1 <vector: histogram> - output histogram vector 2 <num: min> - output min data value 3 <num: max> - output max data value

---

mnm.knn		K-Nearest Neighbour search
		Find the k nearest neighbours and their distances to the query point in multi-dimensional data using an efficient multidimensional search tree with logarithmic time complexity.
		arguments: 1 <num: max k> - max number k of nearest neighbours to search 2 <num: max radius> - max radius of nearest neighbours to search (0 for unlimited) attributes: setdata <matrix: data> - matrix(N, D) of data setsigma <matrix: sigma> - matrix(1, D) of sigma = 1/weights dmode <'orthogonal'|'hyperplane'|'pca': mode> decomposition mode mmode <'mean'|'middle'|'median'> - pivot calculation mode sort <bool: switch> - sort output by distance height <num: height> - given tree height if positive, subtract from maxheight if negative outdist <vector: distance> - output vector of distances to n <= k nearest neighbours (n x 1) outind <vector: indices> - data row indices of n <= k nearest neighbours (n x 1) messages: postdoc - post external doc to console outdist <vector: distance> - output vector of distances to n <= k nearest neighbours (n x 1) outind <vector: indices> - data row indices of n <= k nearest neighbours (n x 1) setk <num: max> - max number k of nearest neighbours to search setradius <num: max> - max radius of nearest neighbours to search (0 for unlimited) getmeanvectors <matrix: out> - copy mean vectors for the M tree nodes to copy matrix (M x D) out getsplitplanes <matrix: out> - copy vectors perpendicular to the hyperplanes splitting the M tree nodes to matrix (M x D) out getprofile <dict: out> - copy profiling info to given dict and clear print [<'info'|'r\ aw'|'data'|'compact'|'nodes'|'profile': mode>] - print tree info of varying detail (default: nodes), print and clear profiling info if keyword 'profile' is given inlets: 1 <vector: x> - query vector to search k-nearest neighbours (1 x D) 2 <matrix: data> - data input matrix (N x D) 3 <vector: sigma> - sigma input matrix (1 x D) outlets: 1 <vector: distance> - output vector of distances to n <= k nearest neighbours (n x 1) 2 <vector: indices> - data row indices of n <= k nearest neighbours (n x 1)

---

mnm.lu		Lower-Upper decomposition
		Calculates LU decomposition of incoming matrix.
		arguments: none attributes: outl <matrix: L> - L output matrix outu <matrix: U> - U output matrix outpivot <matrix: pivot> - pivot output matrix outx <matrix: X> - X output matrix outdet <matrix: det> - determinant output matrix messages: postdoc - post external doc to console determinant - computes determinant of decomposed matrix solve <matrix: input> - solves system with incoming matrix and decomposed matrix outl <matrix: L> - L output matrix outu <matrix: U> - U output matrix outpivot <matrix: pivot> - pivot output matrix outx <matrix: X> - X output matrix outdet <matrix: det> - determinant output matrix inlets: 1 <matrix: input> - matrix to decompose outlets: 1 <matrix: L> - L output matrix 2 <matrix: U> - U output matrix 3 <matrix: pivot> - pivot output matrix 4 <matrix: X> - X output matrix 5 <matrix: det> - determinant output matrix

---

mnm.mahalanobis		Mahalanobis distance
		Calculates mahalanobis distance of incoming matrices or vectors.
		arguments: <matrix: mean> <matrix: covariance> - init mean and covariance attributes: out <matrix: output> - output mahalanobis distance messages: postdoc - post external doc to console set_mu <matrix: mu> - matrix of mean values set_sigma <matrix: sigma> - matrix of covariance out <matrix: output> - output mahalanobis distance inlets: 1 <vector: input> - query vector> 2 <matrix: mu> - mu input matrix 3 <matrix: sigma> - sigma input matrix outlets: 1 <matrix: output> - output mahalanobis distance

---

mnm.mean		Mean filtering
		Calculates mean filtering over vectors (rows or columns) or stream of values (of any dimension).
		arguments: <num: input> [<num: filter>] - init input and filter size attributes: filtersize <num: max> - set the maximum filter size (default is 0 for using the input size) dim <'col'|'row'|'auto'|'stream': mode> - set the dimension on which to operate [auto] out <matrix: output> - filtered values outstate <matrix: state> - output state messages: postdoc - post external doc to console insize <num: input> [<num: filter>] - set input and filter size getstate - get the mean state clear - reset the internal state out <matrix: output> - filtered values outstate <matrix: state> - output state inlets: 1 <matrix: input> - input values outlets: 1 <matrix: output> - filtered values 2 <matrix: state> - output state

---

mnm.meanstd		Mean and standard deviation
		Calculates the arithmetic mean and standard deviation of each column or row (depending on 'mode' argument) as one row or column vector
		arguments: 1 <1|2|'row'|'col': mode switch> - compute over rows or columns [col] attributes: mode <1|2|'row'|'col': mode switch> - compute over rows or columns [col] scalar <bool: switch> - output a simple float value (instead of 1 x 1 matrix) for scalar results [on] outmean <matrix: mean> - mean output vector or value outstd <matrix: stddev> - standard deviation output vector or value messages: postdoc - post external doc to console outmean <matrix: mean> - mean output vector or value outstd <matrix: stddev> - standard deviation output vector or value inlets: 1 <matrix: input> - input matrix outlets: 1 <matrix: mean> - mean output vector or value 2 <matrix: stddev> - standard deviation output vector or value

---

mnm.median		Median filtering
		Calculates median filtering over vectors (rows or columns) or stream of values (of any dimension).
		arguments: <num: input> [<num: filter>] - init input and filter size attributes: filtersize <num: filter> - set the maximum filter size (default is 0 for using the input size) dim <'col'|'row'|'auto'|'stream': mode> - set the dimension on which to operate [auto] out <matrix: output> - output filtered values outstate <matrix: output> - output state messages: postdoc - post external doc to console insize <num: input> [<num: filter>] - set input and filter size getstate - get the median state clear - reset the internal state out <matrix: output> - output filtered values outstate <matrix: output> - output state inlets: 1 <matrix: input> - input values outlets: 1 <matrix: output> - output filtered values 2 <matrix: output> - output state

---

mnm.minmax		Minimum and maximum
		Calculates the min, index of min, max, index of max of each column or row (depending on argument) as one row or column vector
		arguments: 1 <1|2|'row'|'col': mode switch> - compute over rows or columns [col] attributes: mode <1|2|'row'|'col': mode switch> - compute over rows or columns [col] scalar <bool: switch> - output a simple float value (instead of 1 x 1 matrix) for scalar results [on] outmin <vector: min> - minimum output vector or value outargmin <vector: min> - minimum index output vector or value outmax <vector: max> - maximum output vector or value outargmax <vector: max> - maximum index output vector or value messages: postdoc - post external doc to console outmin <vector: min> - minimum output vector or value outargmin <vector: min> - minimum index output vector or value outmax <vector: max> - maximum output vector or value outargmax <vector: max> - maximum index output vector or value inlets: 1 <matrix: input> - incoming matrix, vector, or list outlets: 1 <vector: min> - minimum output vector or value 2 <vector: min> - minimum index output vector or value 3 <vector: max> - maximum output vector or value 4 <vector: max> - maximum index output vector or value

---

mnm.moments		Statistical moments
		Calculates moments from first to specified order.
		arguments: 1 <num: order> - moments maximum order [1] attributes: std <bool: switch> - compute the standards moments for orders > 2 [on] mode <1|2|'row'|'col': mode switch> - compute over rows or columns [col] out <vector: moments> - output moments vector outsum <vector: moments> - output sums vector messages: postdoc - post external doc to console order <num: order> - set moments maximum order out <vector: moments> - output moments vector outsum <vector: moments> - output sums vector inlets: 1 <vector: input> - input vector outlets: 1 <vector: moments> - output moments vector 2 <vector: moments> - output sums vector

---

mnm.nmd		Non-zero Matrix Decomposition
		Calculates NMD of incoming matrix.
		arguments: none attributes: outh <matrix: H> - H output matrix criteria <num: criteria> - criteria sH <num: sH> - sH itermax <num: itermax> - maximum number of iterations messages: postdoc - post external doc to console inlets: 1 <matrix: input> - input matrix 2 <matrix: input> - input matrix 3 <num: L2> - L2 outlets: 1 <matrix: output> - output matrix

---

mnm.nmf		Non-zero Matrix Factorization
		Calculates NMF of incoming matrix.
		arguments: 1 <num: num> - number of components attributes: outw <matrix: W> - W output matrix outh <matrix: H> - H output matrix criteria <num: criteria> - stopping criteria rdim <num: num> - number of components itermax <num: max> - maximum number of iterations messages: postdoc - post external doc to console inlets: 1 <matrix: input> - matrix to be decomposed outlets: 1 <matrix: W> - W output matrix 2 <matrix: H> - H output matrix

---

mnm.obsprob		Obsprob
		Obsprob has to be documented.
		arguments: none attributes: none messages: postdoc - post external doc to console inlets: 1 <matrix: obs> - input observation frames matrix (D x 1) 2 <matrix: states> - input states matrix outlets: 1 <matrix: prob> - output log(B) (K x 1) 2 <num> - no description

---

mnm.onepole		Onepole filtering
		Calculates onepole filtering (low-pass or high-pass) over vectors (rows or columns) or stream of values (of any dimension).
		arguments: <num: input> [<num: filter>] - init input and filter size attributes: f0 <num: freq> - set the onepole frequency, normalised by the Nyquist frequency (default is 1.) dim <'col'|'row'|'auto'|'stream': mode> - set the dimension on which to operate [auto] mode <'lowpass'|'highpass': mode> - filter type [lowpass] out <matrix: output> - output filtered values outstate <matrix: state> - output state messages: postdoc - post external doc to console getstate - get the onepole state clear - reset the internal state insize <num: input> [<num: filter>] - init input and filter size out <matrix: output> - output filtered values outstate <matrix: state> - output state inlets: 1 <matrix: input> - input values outlets: 1 <matrix: output> - output filtered values 2 <matrix: state> - output state

---

mnm.qr		Orthogonal-Right decomposition
		Calculates QR decomposition of incoming matrix.
		arguments: none attributes: outq <matrix: Q> - Q output matrix outr <matrix: R> - R output matrix outx <matrix: X> - X output matrix messages: postdoc - post external doc to console solve <matrix: input> - solve system with QR decomposition outq <matrix: Q> - Q output matrix outr <matrix: R> - R output matrix outx <matrix: X> - X output matrix inlets: 1 <matrix: input> - matrix to be decomposed outlets: 1 <matrix: Q> - Q output matrix 2 <matrix: R> - R output matrix 3 <matrix: X> - X output matrix

---

mnm.stats		Stats
		Stats has to be documented.
		arguments: <vector: histogram> [<num: low> <num: high>] - set histogram vector and boundaries attributes: norm <bool: normalize> - switch normalization [off] messages: postdoc - post external doc to console bang - output stats clear - clear histogram set <vector: histogram> [<num: low> <num: high>] - set histogram vector and boundaries inlets: 1 <num: value> - data input outlets: 1 <num: average> - output average 2 <num: stddev> - output standard deviation 3 <num: count> - output count

---

mnm.submat		Sub-matrix
		Copies sub-matrix outof incoming matrix.
		arguments: <num: begin row> <num: begin col> <num: end row> <num: end col> - start end end indices of sub-matrix attributes: begin <num: row> <num: col> - start indices of sub-matrix end <num: row> <num: col> - end indices of sub-matrix out <matrix: output> - ouput sub-matrix messages: postdoc - post external doc to console out <matrix: output> - ouput sub-matrix inlets: 1 <matrix: input> - input matrix outlets: 1 <matrix: output> - ouput sub-matrix

---

mnm.sum		Matrix sum
		Calculates sum over entire matrix, matrix rows or matrix columns.
		arguments: 1 <1|2|'row'|'col': mode switch> - compute over rows or columns [col] attributes: mode <1|2|'row'|'col': mode switch> - compute over rows or columns [col] out <vector: sum> - output sum vector messages: postdoc - post external doc to console out <vector: sum> - output sum vector inlets: 1 <matrix: input> - incoming matrix to be summed outlets: 1 <vector: sum> - output sum vector

---

mnm.svd		Singular Value Decomposition
		Calculates SVD of incoming matrix.
		arguments: 1 <num: rank> - number of singular values attributes: mode <'manual'|'auto': mode> - automatically eliminate negligible singular values [manual] outu <matrix: U> - U output matrix outs <matrix: S> - S output matrix outvt <matrix: V'> - V' output matrix (transposed of V) messages: postdoc - post external doc to console outu <matrix: U> - U output matrix outs <matrix: S> - S output matrix outvt <matrix: V'> - V' output matrix (transposed of V) inlets: 1 <matrix: input> - matrix to be decomposed by SVD outlets: 1 <matrix: U> - U output matrix 2 <matrix: S> - S output matrix 3 <matrix: V'> - V' output matrix (transposed of V)

---

mnm.transpose		Matrix transposition
		Calculates transposed matrix.
		arguments: none attributes: out <matrix: output> - output transposed matrix messages: postdoc - post external doc to console out <matrix: output> - output transposed matrix inlets: 1 <matrix: input> - matrix to be transposed outlets: 1 <matrix: output> - output transposed matrix

---

mnm.viterbi		Viterbi algorithm
		Calculates Viterbi path of incoming matrices or vectors.
		arguments: none attributes: line <bool: online> - switch online mode on/off [off] verbose <bool: verbose> - verbose or not [off] latency <num: latency> - maximum latency (nb of frames) messages: postdoc - post external doc to console geta - get state matrix A getpi - get state matrix Pi getpsi - get state matrix Psi locpaths - get locpaths matrix reinit - input message reinit bang - decode inlets: 1 - messages only 2 <matrix: observations> - input observation matrix logB (T x K) 3 <matrix: pi> - input state prior distribution Pi (1 x K) 4 <matrix: A> - input state transition matrix A (K x K) outlets: 1 <matrix: path> - output decoded best path 2 <matrix: state> - output state matrices

---

mnm.xdist2		Square of Euclidean distance
		Calculates the square of the Euclidean distance between a vector and each line of a matrix.
		arguments: <matrix: right> - matrix to be used as right operand attributes: swap <bool: switch> swap operands out <matrix: output> - result squared distance matrix messages: postdoc - post external doc to console out <matrix: output> - result squared distance matrix inlets: 1 <matrix: left> - input left operand> 2 <matrix: right> - input right operand> outlets: 1 <matrix: output> - result squared distance matrix

---

mnm.xmul		Matrix multiplication
		Calculates matrix multiplication as in out = left x right. The left and right operands of the matrix multiplication are given by the respective inlets unless the swap option is enabled.The dimensions of the resulting output matrix are corresponding to the minimum dimensions of the two operators.
		arguments: <matrix: op> - set right or (with swap enabled) left matrix multiplication operand attributes: swap <bool: switch> swap operands out <matrix: output> - output matrix messages: postdoc - post external doc to console out <matrix: output> - output matrix inlets: 1 <matrix: input> - multiply matrix with given right or left operand 2 <matrix: op> - set right or (with swap enabled) left matrix multiplication operand outlets: 1 <matrix: output> - output matrix