MNT rename package voxelwise_tutorials

Author: TomDLT

README.rst

@@ -13,14 +13,14 @@
 This repository contains some tutorials describing how to perform voxelwise
 modeling, based for instance on visual imaging experiments.
 The best way to explore these tutorials is to go to the
-`website <https://gallantlab.github.io/tutorials/>`_.
+`website <https://gallantlab.github.io/voxelwise_tutorials/>`_.
 
-The ``voxelwise`` package
+The ``voxelwise_tutorials`` package
 =========================
 
 On top of tutorials, this repository also contains a small Python package
-called ``voxelwise``, which contains useful fonctions to download the data sets,
-load the files, process the data, and visualize the results.
+called ``voxelwise_tutorials``, which contains useful fonctions to download the
+data sets, load the files, process the data, and visualize the results.
 
 Installation
 ------------
@@ -29,8 +29,8 @@ To install this package, run
 
 .. code-block:: bash
 
-   git clone https://github.com/gallantlab/tutorials.git
-   cd tutorials
+   git clone https://github.com/gallantlab/voxelwise_tutorials.git
+   cd voxelwise_tutorials
    pip install .
 
 
@@ -44,7 +44,7 @@ Developers can also install the package in editable mode via:
 Requirements
 ------------
 
-The Python package ``voxelwise`` requires the following dependencies:
+The Python package ``voxelwise_tutorials`` requires the following dependencies:
 
 - `numpy <https://github.com/numpy/numpy>`_
 - `scipy <https://github.com/scipy/scipy>`_
@@ -57,11 +57,12 @@ documentations, such as:
 - `h5py <https://github.com/h5py/h5py>`_
 - `himalaya <https://github.com/gallantlab/himalaya>`_
 - `pymoten <https://github.com/gallantlab/pymoten>`_
-- `voxelwise <https://github.com/gallantlab/tutorials>`_ (this repository)
+- `voxelwise_tutorials <https://github.com/gallantlab/voxelwise_tutorials>`_
+   (this repository)
 
 
 .. |Github| image:: https://img.shields.io/badge/github-tutorials-blue
-   :target: https://github.com/gallantlab/tutorials
+   :target: https://github.com/gallantlab/voxelwise_tutorials
 
 .. |Python| image:: https://img.shields.io/badge/python-3.7%2B-blue
    :target: https://www.python.org/downloads/release/python-370

doc/_static/custom.css

@@ -11,4 +11,4 @@
 }
 .sphx-glr-thumbcontainer a.internal {
     padding: 150px 10px 0 !important;   /*default = 150px 10px 0 */
-}
+}

doc/index.rst

@@ -16,12 +16,13 @@ The tutorials consist of Python scripts, which are rendered in a
 The tutorials are best explored in order, starting with the
 "Movies 3T tutorial".
 
-To run the tutorials yourself, we recommend to download the project on
-github at `gallantlab/tutorials <https://github.com/gallantlab/tutorials>`_.
-On top of the tutorials, the github repository contains a Python package
-called ``voxelwise``, which contains useful fonctions to download the data
-sets, load the files, process the data, and visualize the results.
-Install instructions are available `here <voxelwise_package.html>`_.
+To run the tutorials yourself, we recommend to download the project on github
+at `gallantlab/voxelwise_tutorials
+<https://github.com/gallantlab/voxelwise_tutorials>`_. On top of the tutorials,
+the github repository contains a Python package called ``voxelwise_tutorials``,
+which contains useful fonctions to download the data sets, load the files,
+process the data, and visualize the results. Install instructions are available
+`here <voxelwise_package.html>`_.
 
 Tutorials
 ---------

setup.py

@@ -1,8 +1,8 @@
 import re
 from setuptools import find_packages, setup
 
-# get version from voxelwise/__init__.py
-with open('voxelwise/__init__.py') as f:
+# get version from voxelwise_tutorials/__init__.py
+with open('voxelwise_tutorials/__init__.py') as f:
     infos = f.readlines()
 __version__ = ''
 for line in infos:
@@ -19,7 +19,7 @@
 
 if __name__ == "__main__":
     setup(
-        name='voxelwise',
+        name='voxelwise_tutorials',
         maintainer="Tom Dupre la Tour",
         maintainer_email="tom.dupre-la-tour@m4x.org",
         description="Tools and tutorials for voxelwise modeling",

tutorials/movies_3T/00_download_vim4.py

@@ -34,7 +34,7 @@
 
 ###############################################################################
 # We will only use the first subject in this tutorial, but you can run the same
-# analysis on the four other subjects. Uncomment the lines in `DATAFILES` to
+# analysis on the four other subjects. Uncomment the lines in ``DATAFILES``` to
 # download more subjects.
 #
 # We also skip the stimuli files, since the dataset provides two preprocessed
@@ -43,7 +43,7 @@
 
 import getpass
 
-from voxelwise.io import download_crcns
+from voxelwise_tutorials.io import download_crcns
 
 DATAFILES = [
     "TBD/features/motion_energy.hdf",

tutorials/movies_3T/01_plot_explainable_variance.py

@@ -32,7 +32,7 @@
 import os
 import numpy as np
 
-from voxelwise.io import load_hdf5_array
+from voxelwise_tutorials.io import load_hdf5_array
 
 ###############################################################################
 # First, we load the fMRI responses on the test set, which contains 10 repeats.
@@ -48,7 +48,7 @@
 # Finally, a correction can be applied to account for small numbers of repeat
 # (parameter ``bias_correction``).
 
-from voxelwise.utils import explainable_variance
+from voxelwise_tutorials.utils import explainable_variance
 ev = explainable_variance(Y_test, bias_correction=False)
 
 ###############################################################################
@@ -91,7 +91,7 @@
 # in a flatmap. To ease its use, we provide here an example function
 # ``plot_flatmap_from_mapper``.
 
-from voxelwise.viz import plot_flatmap_from_mapper
+from voxelwise_tutorials.viz import plot_flatmap_from_mapper
 
 mapper_file = os.path.join(directory, 'mappers', f'{subject}_mappers.hdf')
 plot_flatmap_from_mapper(ev, mapper_file, vmin=0, vmax=0.7)
@@ -121,7 +121,7 @@
 # Then, we load the fsaverage mapper. The mapper is a sparse CSR matrix, which
 # map each voxel to some vertices in the fsaverage surface.
 # The mapper is applied with a dot product ``@``.
-from voxelwise.io import load_hdf5_sparse_array
+from voxelwise_tutorials.io import load_hdf5_sparse_array
 voxel_to_fsaverage = load_hdf5_sparse_array(mapper_file,
                                             key='voxel_to_fsaverage')
 ev_projected = voxel_to_fsaverage @ ev

tutorials/movies_3T/02_plot_wordnet_model.py

@@ -40,7 +40,7 @@
 # We first load the fMRI responses.
 import os
 import numpy as np
-from voxelwise.io import load_hdf5_array
+from voxelwise_tutorials.io import load_hdf5_array
 
 file_name = os.path.join(directory, "responses", f"{subject}_responses.hdf")
 Y_train = load_hdf5_array(file_name, key="Y_train")
@@ -82,7 +82,7 @@
 # cross-validation split, which preserves each recording run.
 
 from sklearn.model_selection import check_cv
-from voxelwise.utils import generate_leave_one_run_out
+from voxelwise_tutorials.utils import generate_leave_one_run_out
 
 # indice of first sample of each run
 run_onsets = load_hdf5_array(file_name, key="run_onsets")
@@ -113,7 +113,7 @@
 # delayed feature with a different weight, to maximize the predictions.
 # With a sample every 2 seconds, we typically use 4 delays [1, 2, 3, 4] to
 # cover the most part of the hemodynamic response peak.
-from voxelwise.delayer import Delayer
+from voxelwise_tutorials.delayer import Delayer
 delayer = Delayer(delays=[1, 2, 3, 4])
 
 ###############################################################################
@@ -197,7 +197,7 @@
 # (Check previous example to see how to use the mapper to Freesurfer average
 # surface.)
 import matplotlib.pyplot as plt
-from voxelwise.viz import plot_flatmap_from_mapper
+from voxelwise_tutorials.viz import plot_flatmap_from_mapper
 
 mapper_file = os.path.join(directory, "mappers", f"{subject}_mappers.hdf")
 ax = plot_flatmap_from_mapper(scores, mapper_file, vmin=0, vmax=0.4)
@@ -264,7 +264,7 @@
 # from the diagonal means that one model has better predictive performances
 # than the other.
 
-from voxelwise.viz import plot_hist2d
+from voxelwise_tutorials.viz import plot_hist2d
 
 ax = plot_hist2d(scores_nodelay, scores)
 ax.set(title='Generalization R2 scores', xlabel='model without delays',

tutorials/movies_3T/03_plot_motion_energy_model.py

@@ -45,7 +45,7 @@
 import os
 import numpy as np
 
-from voxelwise.io import load_hdf5_array
+from voxelwise_tutorials.io import load_hdf5_array
 
 file_name = os.path.join(directory, "responses", f"{subject}_responses.hdf")
 Y_train = load_hdf5_array(file_name, key="Y_train")
@@ -82,7 +82,7 @@
 # previous example.
 
 from sklearn.model_selection import check_cv
-from voxelwise.utils import generate_leave_one_run_out
+from voxelwise_tutorials.utils import generate_leave_one_run_out
 
 # indice of first sample of each run
 run_onsets = load_hdf5_array(file_name, key="run_onsets")
@@ -101,7 +101,7 @@
 
 from sklearn.pipeline import make_pipeline
 from sklearn.preprocessing import StandardScaler
-from voxelwise.delayer import Delayer
+from voxelwise_tutorials.delayer import Delayer
 from himalaya.kernel_ridge import KernelRidgeCV
 from himalaya.backend import set_backend
 backend = set_backend("torch_cuda")
@@ -139,7 +139,7 @@
 # The performances are computed using the math:`R^2` scores.
 
 import matplotlib.pyplot as plt
-from voxelwise.viz import plot_flatmap_from_mapper
+from voxelwise_tutorials.viz import plot_flatmap_from_mapper
 
 mapper_file = os.path.join(directory, "mappers", f"{subject}_mappers.hdf")
 ax = plot_flatmap_from_mapper(scores_motion_energy, mapper_file, vmin=0,
@@ -189,7 +189,7 @@
 # from the diagonal means that one model has better predictive performances
 # than the other.
 
-from voxelwise.viz import plot_hist2d
+from voxelwise_tutorials.viz import plot_hist2d
 
 ax = plot_hist2d(scores_wordnet, scores_motion_energy)
 ax.set(title='Generalization R2 scores', xlabel='semantic wordnet model',
@@ -202,7 +202,7 @@
 # To further describe these differences, we can plot both performances on the
 # same flatmap.
 
-from voxelwise.viz import plot_2d_flatmap_from_mapper
+from voxelwise_tutorials.viz import plot_2d_flatmap_from_mapper
 
 mapper_file = os.path.join(directory, "mappers", f"{subject}_mappers.hdf")
 ax = plot_2d_flatmap_from_mapper(scores_wordnet, scores_motion_energy,

tutorials/movies_3T/04_plot_banded_ridge_model.py

@@ -32,7 +32,7 @@
 import os
 import numpy as np
 
-from voxelwise.io import load_hdf5_array
+from voxelwise_tutorials.io import load_hdf5_array
 
 file_name = os.path.join(directory, "responses", f"{subject}_responses.hdf")
 Y_train = load_hdf5_array(file_name, key="Y_train")
@@ -81,7 +81,7 @@
 # We define again a leave-one-run-out cross-validation split scheme.
 
 from sklearn.model_selection import check_cv
-from voxelwise.utils import generate_leave_one_run_out
+from voxelwise_tutorials.utils import generate_leave_one_run_out
 
 # define a cross-validation splitter, compatible with scikit-learn
 n_samples_train = X_train.shape[0]
@@ -97,7 +97,7 @@
 # and add delays with a ``Delayer``.
 from sklearn.pipeline import make_pipeline
 from sklearn.preprocessing import StandardScaler
-from voxelwise.delayer import Delayer
+from voxelwise_tutorials.delayer import Delayer
 from himalaya.backend import set_backend
 backend = set_backend("torch_cuda")
 
@@ -268,7 +268,7 @@
 # from the diagonal means that one model has better predictive performances
 # than the other.
 import matplotlib.pyplot as plt
-from voxelwise.viz import plot_hist2d
+from voxelwise_tutorials.viz import plot_hist2d
 
 ax = plot_hist2d(scores_baseline, scores)
 ax.set(title='Generalization R2 scores', xlabel='KernelRidgeCV',
@@ -309,7 +309,7 @@
 ###############################################################################
 # We can then plot the split scores on a flatmap with a 2D colormap.
 
-from voxelwise.viz import plot_2d_flatmap_from_mapper
+from voxelwise_tutorials.viz import plot_2d_flatmap_from_mapper
 
 mapper_file = os.path.join(directory, "mappers", f"{subject}_mappers.hdf")
 ax = plot_2d_flatmap_from_mapper(split_scores[0], split_scores[1],

tutorials/movies_3T/05_extract_motion_energy.py

@@ -56,8 +56,8 @@
 import numpy as np
 from moten.io import imagearray2luminance
 
-from voxelwise.progress_bar import bar
-from voxelwise.io import load_hdf5_array
+from voxelwise_tutorials.progress_bar import bar
+from voxelwise_tutorials.io import load_hdf5_array
 
 
 def compute_luminance(run_name, size=(96, 96)):
@@ -149,7 +149,7 @@ def compute_motion_energy(luminance,
 # approximately equal to the "motion-energy" features already precomputed in
 # the public data set.
 
-from voxelwise.io import save_hdf5_dataset
+from voxelwise_tutorials.io import save_hdf5_dataset
 
 features_directory = os.path.join(directory, "features")
 if not os.path.exists(features_directory):