PHOTONAI WIZARD

Design your pipeline

Analysis Name

Specify a name for this analysis.

Project Description

Please provide a short description of your project.

Project Folder

Please give a local directory for your photon analysis

Design Choices

You optimize for balanced_accuracy using random_grid_search.

Optimizer	random_grid_search	n_configurations: 30
Inner Cross Validation Strategy	KFold	n_splits: 3 \|shuffle: True
Outer Cross Validation Strategy	KFold	n_splits: 5 \|shuffle: True
Metrics	accuracy
	balanced_accuracy
	specificity
	sensitivity
Best Config Metric	balanced_accuracy

Neuro Transformers

Your neuroimaging transformations consist of 1 elements with 1 hyperparameters.

HarvardOxford_Cortical_Threshold_25

roi_names

['all']

Transformers

Your pipeline contains 2 transformer elements specifying 1 hyperparameters.

LabelEncoder
PCA	n_components	FloatRange(0.2, 0.99)

Estimators

Your learning algorithm is SVC with 4 hyperparameters.

SVC	C	FloatRange(1e-7, 1e7, range_type="geomspace")
	gamma	scale
	max_iter	1000000.0
	kernel	['linear', 'rbf']

                        
                            
# Specify how results are going to be saved
# Define hyperpipe
hyperpipe = Hyperpipe('None',
                      project_folder = './results',
                      optimizer="random_grid_search",
                      optimizer_params={'n_configurations': 30},
                      metrics=['accuracy', 'balanced_accuracy', 'specificity', 'sensitivity'],
                      best_config_metric="balanced_accuracy",
                      outer_cv = KFold(n_splits=5,shuffle=True),
                      inner_cv = KFold(n_splits=3, shuffle=True),
                      verbosity=2)
        
# Add neuro elements
neuro_branch = NeuroBranch('Neuro', nr_of_processes=3)
neuro_branch += PipelineElement('BrainAtlas', hyperparameters={}, atlas_name='HarvardOxford_Cortical_Threshold_25',
                                rois=['all'], extract_mode='vec', batch_size=50)
hyperpipe += neuro_branch
            
# Add transformer elements
preprocessing_pipe = Preprocessing()
hyperpipe += preprocessing_pipe
preprocessing_pipe += PipelineElement("LabelEncoder") 
                        
hyperpipe += PipelineElement("PCA", hyperparameters={'n_components': FloatRange(0.2, 0.99)}, 
                             test_disabled=False)
hyperpipe += PipelineElement("SVC", hyperparameters={'C': FloatRange(1e-7, 1e7, range_type="geomspace"), 'kernel': ['linear', 'rbf']}, gamma='scale', max_iter=1000000.0)

PHOTONAI Wizard