Fmri to dataset

tests/data/BIDS_example/derivatives/SPM_example/sub-01/sub-01_ses-perceptionTest_noise.csv

@@ -0,0 +1 @@
+0,res_0001_run_01

tests/data/BIDS_example/derivatives/SPM_example/sub-01/sub-01_ses-perceptionTest_signal.csv

@@ -0,0 +1,2 @@
+0,cond_Face_run_01
+1,cond_House_run_01

tests/test_fmri.py

@@ -0,0 +1,140 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Tests for BidsDerivatives class and subclasses
+
+"""
+
+import unittest
+import os
+import rsatoolbox.io.fmri as fmri
+import numpy as np
+
+
+def get_fmri_data():
+    fmri_dir = os.path.join(
+            os.path.dirname(__file__), "data", "BIDS_example", "derivatives",
+            "SPM_example")
+    fmri_data = fmri.BidsDerivatives(fmri_dir)
+    return fmri_data
+
+
+def get_subject_data():
+    fmri_data = get_fmri_data()
+    subjects = fmri_data.get_subjects()
+    subject_data = fmri_data.subset_subject(subjects[0])                        # subset data for first subject
+    return subject_data
+
+
+def get_subject_session_data():
+    subject_data = get_subject_data()
+    session_types = subject_data.get_session_types()
+    subject_data_s = subject_data.subset_session_type(session_types[0])         # further subset data for first session type
+    return subject_data_s
+
+
+class TestParsing(unittest.TestCase):
+
+    def test_full_parsing(self):
+        fmri_data = get_fmri_data()
+        np.testing.assert_array_equal(
+            fmri_data.get_subjects(),
+            ['sub-01', 'sub-02'])
+        np.testing.assert_array_equal(
+            fmri_data.get_session_types(),
+            ['ses-perceptionTest', 'ses-perceptionTraining'])
+        np.testing.assert_array_equal(
+            fmri_data.runs_total, 2)
+
+    def test_sub_parsing(self):
+        subject_data = get_subject_data()
+        np.testing.assert_array_equal(
+            subject_data.get_subjects(),
+            ['sub-01'])
+        np.testing.assert_array_equal(
+            subject_data.get_sessions(),
+            ['ses-perceptionTest01'])
+        np.testing.assert_array_equal(
+            subject_data.runs_total, 1)
+        np.testing.assert_array_equal(
+            [os.path.isdir(run_dir)
+             for run_dir in subject_data.get_runs()],
+            [True])
+
+    def test_session_parsing(self):
+        subject_data = get_subject_session_data()
+        np.testing.assert_array_equal(
+            subject_data.get_sessions(),
+            ['ses-perceptionTest01'])
+        np.testing.assert_array_equal(
+            subject_data.runs_total, 1)
+        np.testing.assert_array_equal(
+            [os.path.isdir(run_dir)
+             for run_dir in subject_data.get_runs()],
+            [True])
+
+
+class TestPoolingSPM(unittest.TestCase):
+
+    def test_beta_pooling(self):
+        subject_data = get_subject_session_data()
+        beta_array, beta_descriptors = subject_data.load_betas_SPM()
+        np.testing.assert_array_equal(
+            beta_array.shape, (55, 60, 51, 2))
+        np.testing.assert_array_equal(
+            beta_array.shape[3], len(beta_descriptors))
+        beta_descriptors.sort()
+        np.testing.assert_array_equal(
+            np.unique(beta_descriptors),
+            beta_descriptors)
+
+    def test_beta_pooling_w_dict(self):
+        stim_ids_dict = {"Face": 1, "House": 2}
+        subject_data = get_subject_session_data()
+        beta_array, beta_descriptors = subject_data.load_betas_SPM(
+            stim_ids_dict=stim_ids_dict)
+        np.testing.assert_array_equal(
+            beta_array.shape, (55, 60, 51, 2))
+        np.testing.assert_array_equal(
+            beta_array.shape[3], len(beta_descriptors))
+        beta_descriptors.sort()
+        np.testing.assert_array_equal(
+            np.unique(beta_descriptors),
+            beta_descriptors)
+
+    def test_res_pooling(self):
+        subject_data = get_subject_session_data()
+        res_array, res_descriptors = subject_data.load_residuals_SPM()
+        np.testing.assert_array_equal(
+            res_array.shape, (55, 60, 51, 1))
+        np.testing.assert_array_equal(
+            res_array.shape[3], len(res_descriptors))
+        res_descriptors.sort()
+        np.testing.assert_array_equal(
+            np.unique(res_descriptors),
+            res_descriptors)
+
+
+class TestSaving(unittest.TestCase):
+
+    def test_saving_combo_signal(self):
+        stim_ids_dict = {"Face": 1, "House": 2}
+        subject_data = get_subject_session_data()
+        beta_array, beta_descriptors = subject_data.load_betas_SPM(
+            stim_ids_dict=stim_ids_dict)
+        subject_data.save2combo(
+            beta_array, beta_descriptors, data_type="signal")
+        assert os.path.isfile(subject_data.nifti_filename)
+        assert os.path.isfile(subject_data.csv_filename)
+
+    def test_saving_combo_noise(self):
+        subject_data = get_subject_session_data()
+        res_array, res_descriptors = subject_data.load_residuals_SPM()
+        subject_data.save2combo(
+            res_array, res_descriptors, data_type="noise")
+        assert os.path.isfile(subject_data.nifti_filename)
+        assert os.path.isfile(subject_data.csv_filename)
+
+
+if __name__ == '__main__':
+    unittest.main()