File generated with thanks to "boutiques" (https://github.com/boutiques/boutiques) bosh pprint zenodo.4010742 ================================================================================ Tool name: BEst - cMEM (ver: 1.2) Tool description: EEG/MEG source localisation with Maximum Entropy on the Mean - cMEM (time series representation) Command-line: best [INPUT_DATA] [OUTPUT_DIR_NAME] [MEM_METHOD] [SENSORS_TYPES] [RECONSTRUCTION_WINDOW] [BASELINE_WINDOW] [BASELINE] [NORMALIZATION] [CLUSTERING_METHOD] [MSP_WINDOW] [MSP_THRESHOLD_METHOD] [MSP_THRESHOLD] [NEIGHBORHOOD_ORDER] [SPATIAL_SMOOTHING] [ACTIVE_MEAN_INIT] [ACTIVE_PROBA_INIT] [LAMBDA_INIT] [ACTIVE_PROBA_THRESHOLD] [ACTIVE_VAR_COEF] [INACTIVE_VAR_COEF] [NOISE_COV_METHOD] [OPTIM_METHOD] [USE_PARALLEL] [MAX_WORKERS] ================================================================================ Input Groups: Name: Data Definition Group Member IDs: sensors_types, reconstruction_window, baseline, baseline_window, normalization Name: Job Specifications Group Member IDs: use_parallel, max_workers Name: Clustering Group Member IDs: clustering_method, msp_window, msp_threshold_method, msp_threshold, neighborhood_order, spatial_smoothing Name: Model Priors Group Member IDs: active_mean_init, active_proba_init, lambda_init, active_proba_threshold, active_var_coef, inactive_var_coef Name: Solver Options Group Member IDs: noise_cov_method, optim_method ================================================================================ optional arguments: --baseline [BASELINE] ID: baseline Value Key: [BASELINE] Type: File List: False Optional: True Description: This is your baseline file (.mat, .tgz, .tar.gz, .tar) as exported from Brainstorm. If no baseline file is specified, then the baseline data will be extracted from within the (input) recording data. --normalization {adaptive,fixed} ID: normalization Value Key: [NORMALIZATION] Type: String List: False Optional: True Default Value: adaptive Description: Normalization strategy used for computing the solution. If set to 'adaptive', then a minimum norm solution will be used to normalize the data. --useParallel {true,false} ID: use_parallel Value Key: [USE_PARALLEL] Type: String List: False Optional: True Default Value: true Value Dependency: Value Disables Requires ------- ----------- ----------- true max_workers false max_workers Description: If set, then the samples will be reconstructed in parallel. --maxWorkers [MAX_WORKERS] ID: max_workers Value Key: [MAX_WORKERS] Type: Number List: False Optional: True Integer: True Range: [2, N/A] Default Value: 12 Description: Maximum number of workers for parallel processing. --clusteringMethod {static,blockwise} ID: clustering_method Value Key: [CLUSTERING_METHOD] Type: String List: False Optional: True Default Value: static Value Dependency: Value Disables Requires --------- ---------- ---------- blockwise msp_window static msp_window Description: With the method 'blockwise', cortical parcels are computed within consecutive time windows specified with the option: 'MSP window'. With the method 'static', only one set of cortical parcels is computed for the whole data. --mspWindow [MSP_WINDOW] ID: msp_window Value Key: [MSP_WINDOW] Type: Number List: False Optional: True Integer: False Range: [N/A, N/A] Default Value: 10 Description: Used when clustering method is set to 'blockwise', this is the size of the sliding window in millisecond (ms). --mspThresholdMethod {arbitrary,fdr} ID: msp_threshold_method Value Key: [MSP_THRESHOLD_METHOD] Type: String List: False Optional: True Default Value: arbitrary Value Dependency: Value Disables Requires --------- ------------- ------------- arbitrary msp_threshold fdr msp_threshold Description: Thresholding method applied to the MSP scores. If set to 'fdr', then thresholds will be learned from baseline. Otherwise, the option 'MSP scores threshold' is used. --mspThreshold [MSP_THRESHOLD] ID: msp_threshold Value Key: [MSP_THRESHOLD] Type: Number List: False Optional: True Integer: False Range: [N/A, 1] Description: This is used when 'MSP scores threshold method' is set to 'arbitrary'. A whole brain parcellation is done if this threshold is set to 0. --neighborhoodOrder [NEIGHBORHOOD_ORDER] ID: neighborhood_order Value Key: [NEIGHBORHOOD_ORDER] Type: Number List: False Optional: True Integer: True Range: [N/A, N/A] Default Value: 4 Description: This is used to set the maximal size of cortical parcels (initial source configuration for MEM). --spatialSmoothing [SPATIAL_SMOOTHING] ID: spatial_smoothing Value Key: [SPATIAL_SMOOTHING] Type: Number List: False Optional: True Integer: False Range: [N/A, 1] Default Value: 0.6 Description: Smoothness of MEM solution: spatial regularization of the MEM (linear decay of spatial source correlations). --activeMeanInit {1: regular minimum norm,2: null hypothesis,3: MSP-regularized minimum norm,4: L-curve optimized Minimum Norm Estimate} ID: active_mean_init Value Key: [ACTIVE_MEAN_INIT] Type: String List: False Optional: True Default Value: 2: null hypothesis Description: Initialization method of the active mean of each cluster. --activeProbaInit {1: mean MSP scores,2: max MSP scores,3: median MSP scores,4: equal to 0.5,5: equal to 1} ID: active_proba_init Value Key: [ACTIVE_PROBA_INIT] Type: String List: False Optional: True Default Value: 3: median MSP scores Description: Initialization method of the active probability of each cluster. --lambdaInit {0: null hypothesis (vector of zeros),1: random} ID: lambda_init Value Key: [LAMBDA_INIT] Type: String List: False Optional: True Default Value: 1: random Description: Initialization method of the sensor weights vector. --activeProbaThreshold [ACTIVE_PROBA_THRESHOLD] ID: active_proba_threshold Value Key: [ACTIVE_PROBA_THRESHOLD] Type: Number List: False Optional: True Integer: False Range: [N/A, 1] Description: A threshold used to exclude clusters with low probability from the computed solution. --activeVarCoef [ACTIVE_VAR_COEF] ID: active_var_coef Value Key: [ACTIVE_VAR_COEF] Type: Number List: False Optional: True Integer: False Range: [N/A, 1] Default Value: 0.05 Description: A weight applied to the active variance of each cluster. --inactiveVarCoef [INACTIVE_VAR_COEF] ID: inactive_var_coef Value Key: [INACTIVE_VAR_COEF] Type: Number List: False Optional: True Integer: False Range: [N/A, 1] Description: A weight applied to the inactive variance of each cluster. --noiseCovMethod {0: Identity matrix,1: Scalar matrix,2: Diagonal matrix,3: Full,4: Wavelet-based} ID: noise_cov_method Value Key: [NOISE_COV_METHOD] Type: String List: False Optional: True Default Value: 2: Diagonal matrix Description: The performance of the MEM is tied to a consistent estimation of the noise covariance matrix. We recommend using the method: '2: Diagonal matrix'. --optimMethod {fminunc,minfunc} ID: optim_method Value Key: [OPTIM_METHOD] Type: String List: False Optional: True Default Value: fminunc Description: 'fminunc': MATLAB standard unconstrained optimization routine. 'minfunc': (faster) Unconstrained optimization routine, copyright Mark Schmidt, INRIA. required arguments: --inputData [INPUT_DATA] ID: input_data Value Key: [INPUT_DATA] Type: File List: False Optional: False Description: The input data: a directory or a file (.mat, .tgz, .tar.gz, .tar) as exported from Brainstorm. --outputDirName [OUTPUT_DIR_NAME] ID: output_dir_name Value Key: [OUTPUT_DIR_NAME] Type: String List: False Optional: False Default Value: cbrain-cmem-sources Description: Name of the output directory --memMethod {cMEM} ID: mem_method Value Key: [MEM_METHOD] Type: String List: False Optional: False Default Value: cMEM --sensorsTypes {EEG,MEG,EEG+MEG} ID: sensors_types Value Key: [SENSORS_TYPES] Type: String List: False Optional: False Description: The data sensors types to process. --reconstructionWindow [RECONSTRUCTION_WINDOW] ID: reconstruction_window Value Key: [RECONSTRUCTION_WINDOW] Type: String List: False Optional: False Description: This is the portion of your input recording data to reconstruct. The time window should be specified by two (increasing) numbers in seconds separated by a blank space: 'TIME_BEGIN TIME_END'. For example: '-0.5 1' means from -0.5 to 1 s. --baselineWindow [BASELINE_WINDOW] ID: baseline_window Value Key: [BASELINE_WINDOW] Type: String List: False Optional: False Description: This is the portion of your baseline data to use for estimating a noise covariance matrix. The time window should be specified by two (increasing) numbers in seconds separated by a blank space: 'TIME_BEGIN TIME_END'. For example: '-1 0.5' means from -1 to 0.5 s. ================================================================================