Note
Go to the end to download the full example code
8.7. Run junifer and julearn.#
This example uses a ParcelAggregation marker to compute the mean of each parcel using the Schaefer parcellation (100 rois, 7 Yeo networks) for a 3D nifti to extract some features for machine learning using julearn to predict some other data.
Authors: Leonard Sasse, Sami Hamdan, Nicolas Nieto, Synchon Mandal
License: BSD 3 clause
import tempfile
import nilearn
import pandas as pd
from julearn import run_cross_validation, PipelineCreator
import junifer.testing.registry # noqa: F401
from junifer.api import collect, run
from junifer.storage.sqlite import SQLiteFeatureStorage
from junifer.utils import configure_logging
Set the logging level to info to see extra information:
configure_logging(level="INFO")
2023-10-17 10:32:20,390 - JUNIFER - INFO - ===== Lib Versions =====
2023-10-17 10:32:20,391 - JUNIFER - INFO - numpy: 1.25.2
2023-10-17 10:32:20,391 - JUNIFER - INFO - scipy: 1.11.3
2023-10-17 10:32:20,391 - JUNIFER - INFO - pandas: 2.0.3
2023-10-17 10:32:20,391 - JUNIFER - INFO - nilearn: 0.10.2
2023-10-17 10:32:20,391 - JUNIFER - INFO - nibabel: 5.1.0
2023-10-17 10:32:20,391 - JUNIFER - INFO - junifer: 0.0.4.dev96
2023-10-17 10:32:20,391 - JUNIFER - INFO - ========================
Define the markers you want:
marker_dicts = [
{
"name": "Schaefer100x17_TrimMean80",
"kind": "ParcelAggregation",
"parcellation": "Schaefer100x17",
"method": "trim_mean",
"method_params": {"proportiontocut": 0.2},
},
{
"name": "Schaefer200x17_Mean",
"kind": "ParcelAggregation",
"parcellation": "Schaefer200x17",
"method": "mean",
},
]
Define target and confounds for julearn machine learning:
Load the VBM phenotype data for machine learning data: - Fetch the Oasis dataset
oasis_dataset = nilearn.datasets.fetch_oasis_vbm()
age = oasis_dataset.ext_vars[y][:10]
sex = (
pd.Series(oasis_dataset.ext_vars["mf"][:10])
.map(lambda x: 1 if x == "F" else 0)
.values
)
/opt/hostedtoolcache/Python/3.10.13/x64/lib/python3.10/site-packages/nilearn/datasets/struct.py:852: UserWarning: `legacy_format` will default to `False` in release 0.11. Dataset fetchers will then return pandas dataframes by default instead of recarrays.
warnings.warn(_LEGACY_FORMAT_MSG)
Create a temporary directory for junifer feature extraction:
with tempfile.TemporaryDirectory() as tmpdir:
storage = {"kind": "SQLiteFeatureStorage", "uri": f"{tmpdir}/test.sqlite"}
# run the defined junifer feature extraction pipeline
run(
workdir="/tmp",
datagrabber={"kind": "OasisVBMTestingDataGrabber"},
markers=marker_dicts,
storage=storage,
)
# read in extracted features and add confounds and targets
# for julearn run cross validation
collect(storage)
db = SQLiteFeatureStorage(uri=storage["uri"])
df_vbm = db.read_df(feature_name="VBM_GM_Schaefer200x17_Mean")
oasis_subjects = [x[0] for x in df_vbm.index]
df_vbm.index = oasis_subjects
/opt/hostedtoolcache/Python/3.10.13/x64/lib/python3.10/site-packages/nilearn/datasets/struct.py:852: UserWarning: `legacy_format` will default to `False` in release 0.11. Dataset fetchers will then return pandas dataframes by default instead of recarrays.
warnings.warn(_LEGACY_FORMAT_MSG)
2023-10-17 10:32:20,416 - JUNIFER - INFO - Getting element sub-01
2023-10-17 10:32:20,416 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:20,416 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0001_MR1/mwrc1OAS1_0001_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:20,416 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:20,417 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:20,418 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:20,418 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:20,418 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:20,418 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:20,418 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:20,418 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:20,420 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:21,258 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:21,308 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:21,308 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:21,308 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:21,308 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:21,308 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:21,308 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:21,308 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:21,311 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:22,215 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:22,288 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:22,288 - JUNIFER - INFO - Getting element sub-02
2023-10-17 10:32:22,288 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:22,288 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0002_MR1/mwrc1OAS1_0002_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:22,288 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:22,289 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:22,289 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:22,290 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:22,290 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:22,290 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:22,290 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:22,290 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:22,292 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:23,158 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:23,202 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:23,203 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:23,203 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:23,203 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:23,203 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:23,203 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:23,203 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:23,205 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:24,075 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:24,149 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:24,149 - JUNIFER - INFO - Getting element sub-03
2023-10-17 10:32:24,149 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:24,149 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0003_MR1/mwrc1OAS1_0003_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:24,149 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:24,151 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:24,151 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:24,151 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:24,151 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:24,151 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:24,151 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:24,151 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:24,153 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:25,027 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:25,073 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:25,073 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:25,074 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:25,074 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:25,074 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:25,074 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:25,074 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:25,076 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:25,976 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:26,049 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:26,049 - JUNIFER - INFO - Getting element sub-04
2023-10-17 10:32:26,049 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:26,049 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0004_MR1/mwrc1OAS1_0004_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:26,049 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:26,050 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:26,051 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:26,051 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:26,051 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:26,051 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:26,051 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:26,051 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:26,053 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:26,950 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:26,995 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:26,995 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:26,995 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:26,995 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:26,995 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:26,995 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:26,995 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:26,997 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:27,833 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:27,903 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:27,904 - JUNIFER - INFO - Getting element sub-05
2023-10-17 10:32:27,904 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:27,904 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0005_MR1/mwrc1OAS1_0005_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:27,904 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:27,905 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:27,905 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:27,905 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:27,905 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:27,905 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:27,905 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:27,905 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:27,907 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:28,758 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:28,803 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:28,804 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:28,804 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:28,804 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:28,804 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:28,804 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:28,804 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:28,807 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:29,715 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:29,784 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:29,785 - JUNIFER - INFO - Getting element sub-06
2023-10-17 10:32:29,785 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:29,785 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0006_MR1/mwrc1OAS1_0006_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:29,785 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:29,786 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:29,786 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:29,787 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:29,787 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:29,787 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:29,787 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:29,787 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:29,789 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:30,603 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:30,650 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:30,650 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:30,650 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:30,651 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:30,651 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:30,651 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:30,651 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:30,653 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:31,541 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:31,611 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:31,611 - JUNIFER - INFO - Getting element sub-07
2023-10-17 10:32:31,611 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:31,611 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0007_MR1/mwrc1OAS1_0007_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:31,611 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:31,613 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:31,613 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:31,613 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:31,613 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:31,613 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:31,613 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:31,613 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:31,616 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:32,493 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:32,547 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:32,548 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:32,548 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:32,548 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:32,548 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:32,548 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:32,548 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:32,550 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:33,397 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:33,467 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:33,467 - JUNIFER - INFO - Getting element sub-08
2023-10-17 10:32:33,467 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:33,467 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0009_MR1/mwrc1OAS1_0009_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:33,467 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:33,468 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:33,468 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:33,469 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:33,469 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:33,469 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:33,469 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:33,469 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:33,471 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:34,268 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:34,314 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:34,314 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:34,314 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:34,314 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:34,314 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:34,314 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:34,314 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:34,317 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:35,166 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:35,240 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:35,240 - JUNIFER - INFO - Getting element sub-09
2023-10-17 10:32:35,240 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:35,240 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0010_MR1/mwrc1OAS1_0010_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:35,240 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:35,241 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:35,241 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:35,242 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:35,242 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:35,242 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:35,242 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:35,242 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:35,244 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:36,093 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:36,138 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:36,138 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:36,138 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:36,138 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:36,138 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:36,138 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:36,138 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:36,140 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:37,018 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:37,089 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:37,089 - JUNIFER - INFO - Getting element sub-10
2023-10-17 10:32:37,089 - JUNIFER - INFO - Fitting pipeline
2023-10-17 10:32:37,089 - JUNIFER - INFO - Reading VBM_GM from /home/runner/nilearn_data/oasis1/OAS1_0011_MR1/mwrc1OAS1_0011_MR1_mpr_anon_fslswapdim_bet.nii.gz
2023-10-17 10:32:37,089 - JUNIFER - INFO - VBM_GM is type NIFTI
2023-10-17 10:32:37,090 - JUNIFER - INFO - Fitting marker Schaefer100x17_TrimMean80
2023-10-17 10:32:37,091 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:37,091 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:37,091 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:37,091 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:37,091 - JUNIFER - INFO - n_rois: 100
2023-10-17 10:32:37,091 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:37,093 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_100Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:37,914 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:37,958 - JUNIFER - INFO - Fitting marker Schaefer200x17_Mean
2023-10-17 10:32:37,958 - JUNIFER - INFO - Computing VBM_GM
2023-10-17 10:32:37,959 - JUNIFER - INFO - Fetching one of Schaefer parcellations.
2023-10-17 10:32:37,959 - JUNIFER - INFO - Parcellation parameters:
2023-10-17 10:32:37,959 - JUNIFER - INFO - resolution: 2.0
2023-10-17 10:32:37,959 - JUNIFER - INFO - n_rois: 200
2023-10-17 10:32:37,959 - JUNIFER - INFO - yeo_networks: 17
2023-10-17 10:32:37,961 - JUNIFER - INFO - Loading parcellation /home/runner/junifer/data/parcellations/schaefer_2018/Schaefer2018_200Parcels_17Networks_order_FSLMNI152_2mm.nii.gz
2023-10-17 10:32:38,832 - JUNIFER - INFO - Storing in <SQLiteFeatureStorage @ /tmp/tmpdm6q0sln/test.sqlite (multiple output)>
2023-10-17 10:32:38,902 - JUNIFER - INFO - Marker collection fitting done
2023-10-17 10:32:38,902 - JUNIFER - INFO - Collecting data using SQLiteFeatureStorage
2023-10-17 10:32:38,902 - JUNIFER - INFO - Collecting data from /tmp/tmpdm6q0sln/*test.sqlite
file: 0it [00:00, ?it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 9.49it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 9.47it/s]
file: 1it [00:00, 4.38it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.55it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.53it/s]
file: 2it [00:00, 4.62it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.60it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.58it/s]
file: 3it [00:00, 4.73it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 6.18it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 6.18it/s]
file: 4it [00:00, 3.81it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.59it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.57it/s]
file: 5it [00:01, 4.13it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.57it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.55it/s]
file: 6it [00:01, 4.35it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.75it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.72it/s]
file: 7it [00:01, 4.51it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.67it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.65it/s]
file: 8it [00:01, 4.62it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 6.04it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 6.03it/s]
file: 9it [00:02, 3.88it/s]
feature: 0%| | 0/2 [00:00<?, ?it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.64it/s]
feature: 100%|██████████| 2/2 [00:00<00:00, 10.62it/s]
file: 10it [00:02, 4.14it/s]
file: 10it [00:02, 4.24it/s]
2023-10-17 10:32:41,261 - JUNIFER - INFO - Collect done
Using julearn for machine learning: We predict the age given our vbm features and sex as a confound.
X = list(df_vbm.columns)
df_vbm[y] = age
df_vbm[confound] = sex
X_types = {
"features": X,
"confound": confound,
}
creator = PipelineCreator(problem_type="regression", apply_to="features")
creator.add("zscore", apply_to=["features", "confound"])
creator.add("confound_removal", apply_to="features", confounds="confound")
creator.add("ridge")
scores = run_cross_validation(
X=X + [confound],
y=y,
X_types=X_types,
data=df_vbm,
model=creator,
cv=3,
)
print(scores)
fit_time score_time ... fold cv_mdsum
0 0.232204 0.028238 ... 0 509497eb21da473048117e0c6704d3ea
1 0.227732 0.027601 ... 1 509497eb21da473048117e0c6704d3ea
2 0.225315 0.027693 ... 2 509497eb21da473048117e0c6704d3ea
[3 rows x 8 columns]
Interpretation of results: Doing machine learning with only 10 datapoints is not meaningful. This explains the big variation in scores for different cross-validation folds.
Total running time of the script: (0 minutes 21.935 seconds)