def loading_data(img_pattern): loader = DataLoader(configuration_file=conf_file, loader='mat', task='fcmri', event_file=img_pattern[:-4] + ".txt", img_pattern=img_pattern, atlas='findlab') prepro = PreprocessingPipeline(nodes=[ Transformer(), #Detrender(), #SampleZNormalizer(), #FeatureZNormalizer() ]) #prepro = PreprocessingPipeline() ds = loader.fetch(prepro=prepro) return ds
data_path = '/media/robbis/DATA/meg/viviana-hcp/' conf_file = "/media/robbis/DATA/meg/viviana-hcp/bids.conf" loader = DataLoader( configuration_file=conf_file, data_path=data_path, subjects="/media/robbis/DATA/meg/viviana-hcp/participants.tsv", loader='bids-meg', task='blp', bids_atlas="complete", bids_correction="corr", bids_derivatives='True', load_fx='hcp-blp') ds = loader.fetch() nodes = ds.fa.nodes_1 matrix = np.zeros_like(ds.samples[0]) nanmask = np.logical_not(np.isnan(ds.samples).sum(0)) ds = ds[:, nanmask] networks = ['AN', 'CON', 'DAN', 'DMN', 'FPN', 'LN', 'MN', 'VAN', 'VFN', 'VPN'] networks = [[n] for n in networks] kwargs_list = [{'nodes_1': v, 'nodes_2': v} for v in networks] ############################################################################ base_config = dict(prepro=['sample_slicer', 'feature_slicer'], estimator=[('clf', LinearRegression())], analysis=TaskPredictionTavor, analysis__n_jobs=-1,
from pyitab.preprocessing.pipelines import PreprocessingPipeline from pyitab.analysis.iterator import AnalysisIterator from pyitab.analysis.configurator import AnalysisConfigurator from pyitab.analysis.pipeline import AnalysisPipeline from sklearn.feature_selection.univariate_selection import SelectKBest from sklearn.model_selection import * from sklearn.svm.classes import SVC import _pickle as pickle loader = DataLoader( configuration_file="/home/carlos/fmri/carlo_ofp/ofp_new.conf", task='OFP_NORES') ds = loader.fetch() decoding = RoiDecoding(n_jobs=20, scoring=['accuracy']) results = dict() for subject in np.unique(ds.sa.subject): results[subject] = [] for evidence in [1, 2, 3]: pipeline = PreprocessingPipeline(nodes=[ TargetTransformer('decision'), SampleSlicer(**{ 'subject': [subject], 'evidence': [evidence] }), Balancer(balancer=RandomUnderSampler(return_indices=True),
loader='mat', task='fcmri', atlas='findlab', event_file=m[:-4]+".txt", img_pattern=m) prepro = PreprocessingPipeline(nodes=[ #Transformer(), #Detrender(), SampleZNormalizer(), #FeatureZNormalizer() ]) #prepro = PreprocessingPipeline() ds = loader.fetch(prepro=prepro) _default_options = [ { 'prepro':['sample_slicer', 'target_transformer'], 'target_transformer__attr': 'expertise', 'sample_slicer__targets': ['Samatha'] }, { 'prepro':['sample_slicer', 'target_transformer'], 'target_transformer__attr': 'age', 'sample_slicer__targets': ['Samatha'] }, { 'prepro':['sample_slicer', 'target_transformer', 'sample_residual'], 'target_transformer__attr': 'expertise',
from pyitab.preprocessing.math import AbsoluteValueTransformer, SignTransformer from pyitab.preprocessing.base import Transformer from pyitab.analysis.states.base import Clustering from sklearn import cluster, mixture from joblib import Parallel, delayed conf_file = "/media/robbis/DATA/fmri/working_memory/working_memory.conf" conf_file = '/m/home/home9/97/guidotr1/unix/data/simulations/meg/simulations.conf' loader = DataLoader(configuration_file=conf_file, loader='simulations', task='simulations') ds = loader.fetch(prepro=Transformer()) _default_options = { 'estimator': [ [[('clf1', cluster.MiniBatchKMeans())]], [[('clf1', cluster.KMeans())]], [[('clf1', cluster.SpectralClustering())]], [[('clf1', cluster.AgglomerativeClustering())]], [[('clf5', mixture.GaussianMixture())]], ], 'sample_slicer__subject': [[trial] for trial in np.unique(ds.sa.subject)], 'estimator__clf1__n_clusters': range(2, 10), 'estimator__clf5__n_components': range(2, 10), }
from imblearn.over_sampling import SMOTE import numpy as np from imblearn.under_sampling import * from imblearn.over_sampling import * conf_file = "/home/carlos/mount/megmri03/fmri/carlo_ofp/ofp.conf" conf_file = "/media/robbis/DATA/fmri/carlo_ofp/ofp.conf" #conf_file = "/home/carlos/fmri/carlo_ofp/ofp_new.conf" if conf_file[1] == 'h': from mvpa_itab.utils import enable_logging root = enable_logging() loader = DataLoader(configuration_file=conf_file, task='OFP') ds = loader.fetch() return_ = True ratio = 'auto' _default_options = { 'sample_slicer__evidence': [[1]], 'sample_slicer__subject': [[s] for s in np.unique(ds.sa.subject)], 'balancer__balancer': [ AllKNN(return_indices=return_, ratio=ratio), CondensedNearestNeighbour(return_indices=return_, ratio=ratio), EditedNearestNeighbours(return_indices=return_, ratio=ratio), InstanceHardnessThreshold(return_indices=return_, ratio=ratio), NearMiss(return_indices=return_, ratio=ratio), OneSidedSelection(return_indices=return_, ratio=ratio), RandomUnderSampler(return_indices=return_, ratio=ratio,
from pyitab.analysis.configurator import AnalysisConfigurator from pyitab.preprocessing import SampleSlicer, FeatureSlicer from pyitab.analysis.roi import RoiAnalyzer import os import numpy as np conf_file = "/home/robbis/mount/permut1/sherlock/bids/bids.conf" loader = DataLoader(configuration_file=conf_file, loader='bids', task='preproc', bids_task=['day1']) subjects = ['marcer', 'matsim', 'simpas'] for s in subjects: ds = loader.fetch(subject_names=[s], prepro=[SampleSlicer(trial_type=np.arange(1, 32))]) roi_analyzer = RoiAnalyzer(analysis=GSBS()) roi_analyzer.fit(ds, roi=['aal'], kmax=50) roi_analyzer.save() ################## Resting state ########################## conf_file = path = "/home/robbis/mount/permut1/sherlock/bids/bids.conf" loader = DataLoader(configuration_file=conf_file, data_path="/home/robbis/mount/permut1/sherlock/bids/", subjects='participants.tsv', loader='bids', task='preproc', bids_task=['day1'])
conf_file = "/media/robbis/DATA/fmri/carlo_mdm/memory.conf" loader = DataLoader( configuration_file=conf_file, #loader=load_mat_ds, task='BETA_MVPA') prepro = PreprocessingPipeline(nodes=[ #Transformer(), Detrender(), SampleZNormalizer(), FeatureZNormalizer() ]) #prepro = PreprocessingPipeline() ds = loader.fetch(prepro=prepro) _default_options = { #'target_trans__target': ["decision"], 'sample_slicer__accuracy': [[1], [0]], } _default_config = { 'prepro': ['sample_slicer', 'target_transformer', 'balancer'], 'sample_slicer__decision': ['NEW', 'OLD'], 'sample_slicer__evidence': [1], 'sample_slicer__accuracy': [0], 'target_transformer__target': "decision", "balancer__attr": 'subject', 'estimator': [('clf', SVC(C=1, kernel='linear'))], 'estimator__clf__C': 1,
roi_labels=roi_labels, task='RESIDUALS_MVPA') prepro = PreprocessingPipeline(nodes=[ #Transformer(), #Detrender(attr='file'), Detrender(attr='chunks'), SampleZNormalizer(), FeatureZNormalizer(), SampleSlicer(frame=[1, 2, 3, 4, 5, 6, 7]), #TargetTransformer(attr='decision'), MemoryReducer(dtype=np.float16), #Balancer(attr='frame'), ]) ds = loader.fetch(prepro=prepro, n_subjects=8) ds = MemoryReducer(dtype=np.float16).transform(ds) labels = list(roi_labels.keys())[:-1] import sentry_sdk sentry_sdk.init("https://[email protected]/1439199") _default_options = [ { 'target_transformer__attr': "image_type", 'sample_slicer__attr': { 'image_type': ["I", "O"]
from pyitab.analysis.states.gsbs import GSBS from pyitab.io.loader import DataLoader from pyitab.analysis.configurator import AnalysisConfigurator from pyitab.preprocessing import SampleSlicer, FeatureSlicer from pyitab.preprocessing.connectivity import SpeedEstimator import os import numpy as np conf_file = path = "/home/robbis/mount/permut1/sherlock/bids/bids.conf" loader = DataLoader(configuration_file=conf_file, loader='bids', task='preproc', bids_task=['day1'], bids_run=['01', '02', '03']) ds = loader.fetch( subject_names=['matsim'], prepro=[SampleSlicer(trial_type=np.arange(1, 32)), FeatureSlicer(aal=[1])]) X = ds.samples speed = SpeedEstimator().transform(ds) peaks = speed > np.mean(speed) + 2 * np.std(speed) peaks_idx = np.nonzero(peaks.flatten())[0] X_ = np.split(X, peaks_idx, axis=0) cluster = [i * np.ones(x.shape[0]) for i, x in X_]
cross_val_multiscore, LinearModel, get_coef, Vectorizer, CSP) from sklearn.linear_model import LogisticRegression import warnings warnings.filterwarnings("ignore") conf_file = "/media/robbis/DATA/meg/c2b/meeting-december-data/bids.conf" loader = DataLoader(configuration_file=conf_file, loader='bids-meg', bids_window='300', bids_ses='01', task='power') ds = loader.fetch(subject_names=['sub-109123'], prepro=[Transformer()]) _default_options = { 'loader__bids_ses': ['01', '02'], 'sample_slicer__targets' : [ ['LH', 'RH'], ['LF', 'RF'], #['LH', 'RH', 'LF', 'RF'] ], 'estimator__clf': [ LinearModel(LogisticRegression(C=1, solver='liblinear')), SVC(C=1, kernel='linear', probability=True), SVC(C=1, gamma=1, kernel='rbf', probability=True),
from pyitab.preprocessing.pipelines import PreprocessingPipeline from pyitab.preprocessing.functions import Detrender, SampleSlicer, \ TargetTransformer, Transformer from pyitab.preprocessing.normalizers import SampleZNormalizer import warnings warnings.filterwarnings("ignore") conf_file = "/media/robbis/DATA/meg/reftep/bids.conf" loader = DataLoader(configuration_file=conf_file, task='reftep', load_fx='reftep-conn', loader='bids-meg', bids_pipeline='connectivity+lv') ds = loader.fetch(n_subjects=9) _default_options = { 'prepro': [ ['sample_slicer', 'target_transformer'], ['sample_slicer', 'feature_znormalizer', 'target_transformer'], ['sample_slicer', 'sample_znormalizer', 'target_transformer'], ], 'sample_slicer__subject': [[s] for s in np.unique(ds.sa.subject)], 'estimator__fsel__k': [50, 100, 150], 'estimator__clf': [ LogisticRegression(penalty='l1', solver='liblinear'), SVC(C=1, kernel='linear'), ], }