DOC: add a tutorial based on a group report I've developed

README.rst

@@ -29,3 +29,5 @@
 * **Assembler**: end-user *NiPreps* write out reportlets to a predetermined folder, which is then queried by the assembler using *PyBIDS*.
   The assembler follows a *report specification* in YAML format, which states the query to find specific reportlets and their corresponding metadata and text annotations.
   As a result, one HTML file with a concatenation of reportlets is produced.
+
+  A tutorial is available under the `docs/notebooks` directory of the repository to help you build your own visual report using *NiReports*.

docs/notebooks/data/fMRI_duration_after_censoring.csv

@@ -0,0 +1,4 @@
+filename,duration
+sub-001_ses-001_task-qct_dir-LR_part-mag_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz,156.8
+sub-001_ses-003_task-qct_dir-LR_part-mag_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz,156.8
+sub-001_ses-004_task-qct_dir-RL_part-mag_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz,145.6

docs/notebooks/data/reports-spec.yml

@@ -0,0 +1,38 @@
+package: funconn
+sections:
+- name: Group report 
+  reportlets:
+  - bids:
+      desc: censoring
+      extension: [.html]
+      suffix: bold
+    caption: |
+      fMRI duration after censoring computed as the number of good timepoints multiplied by the repetition time (TR). 
+      The red line corresponds to the QC cutoff of 5 minutes (300 seconds). Less than 5 minutes of fMRI is not enough to 
+      reliably estimate connectivity.
+    subtitle: Report about censoring.
+  - bids: {desc: fcdist, suffix: bold}
+    caption: Density distributions of within-session FC strengths. The FC distribution from each session are overlaid.
+    subtitle: Functional connectivity density distributions.
+    style:
+      max-width: 1200px
+
+  - bids: {desc: qcfc, suffix: bold}
+    caption: |
+      QC-FC plots tested functional connectivity associations with three image quality metrics (fd_mean, fd_num, and
+      fd_perc). Plots were generated from functional data from all sessions. Red dotted lines represent a theoretical 
+      artifact-free null-hypothesis distribution obtained through permutation analyses. QC-FC percent match level represents the distance
+      between the observed and the null-hypothesis distribution and is computed using the R implementation of the two-sample Kolmogorov-Smirnov 
+      test for goodness of fit. Red label indicates that the QC-FC distribution did not reach above the 95% cutoff.
+    subtitle: QC-FC correlation distributions.
+    style:
+      max-width: 1350px
+
+  - bids: {desc: qcfcvseuclidean, suffix: bold}
+    caption: |
+      Correlation between QC-FC and Euclidean distance separating nodes. The euclidean distance is computed from the centers of mass 
+      of each region in the atlas used to compute the FC.
+    subtitle: Distance-dependent effect of motion.
+    style:
+      max-width: 1350px
+