DOC move references to the end of the examples

Author: TomDLT

tutorials/movies_3T/00_download_vim4.py

@@ -15,14 +15,6 @@
 This tutorial is based on publicly available data `published on CRCNS
 <https://crcns.org/data-sets/vc/TBD>`_. If you publish any work using this data
 set, please cite the original publication [1]_, and the data set [2]_.
-
-.. [1] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2012). A
-    continuous semantic space describes the representation of thousands of
-    object and action categories across the human brain. Neuron, 76(6),
-    1210-1224.
-
-.. [2] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2020): Gallant
-    Lab Natural Movie 3T fMRI Data. CRCNS.org. http://dx.doi.org/10.6080/TBD
 """
 # sphinx_gallery_thumbnail_path = "static/crcns.png"
 
@@ -89,3 +81,16 @@
 for datafile in DATAFILES:
     local_filename = download_crcns(datafile, username, password,
                                     destination=directory)
+
+###############################################################################
+# References
+# ----------
+#
+# .. [1] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2012). A
+#     continuous semantic space describes the representation of thousands of
+#     object and action categories across the human brain. Neuron, 76(6),
+#     1210-1224.
+#
+# .. [2] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2020):
+#     Gallant Lab Natural Movie 3T fMRI Data. CRCNS.org.
+#     http://dx.doi.org/10.6080/TBD

tutorials/movies_3T/04_plot_motion_energy_model.py

@@ -14,6 +14,7 @@
 *Motion-energy features:* Motion-energy features result from filtering a video
 stimulus with spatio-temporal Gabor filters. A pyramid of filters is used to
 compute the motion-energy features at multiple spatial and temporal scales.
+Motion-energy features were introduced in [1]_.
 
 *Summary:* As in the previous example, we first concatenate the features with
 multiple delays, to account for the slow hemodynamic response. A linear

tutorials/movies_3T/06_extract_motion_energy.py

@@ -15,16 +15,14 @@
 Motion-energy features were introduced in [1]_.
 
 The motion-energy extraction is performed by the package `pymoten
-<https://github.com/gallantlab/pymoten>`_.
+<https://github.com/gallantlab/pymoten>`_. Check the pymoten `gallery of
+examples <https://gallantlab.github.io/pymoten/auto_examples/index.html>`_ for
+visualizing motion-energy filters, and for pymoten API usage examples.
 
 Running time
 ------------
 Extracting motion energy is a bit longer than the other examples. It typically
 takes a couple hours to run.
-
-.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
-    Gallant, J. L. (2011). Reconstructing visual experiences from brain
-    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.
 """
 # sphinx_gallery_thumbnail_path = "static/moten.png"
 ###############################################################################

tutorials/movies_4T/00_download_vim2.py

@@ -13,14 +13,6 @@
 `published on CRCNS <https://crcns.org/data-sets/vc/vim-2/about-vim-2>`_.
 If you publish any work using this data set, please cite the original
 publication [1]_, and the data set [2]_.
-
-.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
-    Gallant, J. L. (2011). Reconstructing visual experiences from brain
-    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.
-
-.. [2] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
-    Gallant, J. L. (2014): Gallant Lab Natural Movie 4T fMRI Data. CRCNS.org.
-    http://dx.doi.org/10.6080/K00Z715X
 """
 # sphinx_gallery_thumbnail_path = "static/crcns.png"
 ###############################################################################
@@ -60,3 +52,14 @@
 for datafile in DATAFILES:
     local_filename = download_crcns(datafile, username, password,
                                     destination=directory, unpack=True)
+###############################################################################
+# References
+# ----------
+#
+# .. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
+#     Gallant, J. L. (2011). Reconstructing visual experiences from brain
+#     activity evoked by natural movies. Current Biology, 21(19), 1641-1646.
+#
+# .. [2] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
+#     Gallant, J. L. (2014): Gallant Lab Natural Movie 4T fMRI Data. CRCNS.org.
+#     http://dx.doi.org/10.6080/K00Z715X

tutorials/movies_4T/01_extract_motion_energy.py

@@ -5,23 +5,20 @@
 
 This script describes how to extract motion-energy features from the stimuli.
 
-*Motion-energy features:*
-Motion-energy features result from filtering a video stimulus with
-spatio-temporal Gabor filters. A pyramid of filters is used to compute the
-motion-energy features at multiple spatial and temporal scales. Motion-energy
-features were introduced in [1]_.
+*Motion-energy features:* Motion-energy features result from filtering a video
+stimulus with spatio-temporal Gabor filters. A pyramid of filters is used to
+compute the motion-energy features at multiple spatial and temporal scales.
+Motion-energy features were introduced in [1]_.
 
 The motion-energy extraction is performed by the package `pymoten
-<https://github.com/gallantlab/pymoten>`_.
+<https://github.com/gallantlab/pymoten>`_. Check the pymoten `gallery of
+examples <https://gallantlab.github.io/pymoten/auto_examples/index.html>`_ for
+visualizing motion-energy filters, and for pymoten API usage examples.
 
 Running time
 ------------
 Extracting motion energy is a bit longer than the other examples. It typically
 takes a couple hours to run.
-
-.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
-    Gallant, J. L. (2011). Reconstructing visual experiences from brain
-    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.
 """
 # sphinx_gallery_thumbnail_path = "static/moten.png"
 ###############################################################################
@@ -165,3 +162,11 @@ def compute_motion_energy(luminance,
 save_hdf5_dataset(
     os.path.join(features_directory, "motion_energy.hdf"),
     dataset=dict(X_train=motion_energy_train, X_test=motion_energy_test))
+
+###############################################################################
+# References
+# ----------
+#
+# .. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu,
+#     B., & Gallant, J. L. (2011). Reconstructing visual experiences from brain
+#     activity evoked by natural movies. Current Biology, 21(19), 1641-1646.

tutorials/movies_4T/02_plot_ridge_model.py

@@ -14,6 +14,7 @@
 *Motion-energy features:* Motion-energy features result from filtering a video
 stimulus with spatio-temporal Gabor filters. A pyramid of filters is used to
 compute the motion-energy features at multiple spatial and temporal scales.
+Motion-energy features were introduced in [1]_.
 
 *Summary:* We first concatenate the features with multiple delays, to account
 for the slow hemodynamic response. A linear regression model then weights each
@@ -24,10 +25,6 @@
 cross-validation. Finally, the model generalization performance is evaluated on
 a held-out test set, comparing the model predictions with the ground-truth fMRI
 responses.
-
-.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &
-    Gallant, J. L. (2011). Reconstructing visual experiences from brain
-    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.
 """
 # sphinx_gallery_thumbnail_number = 2
 ###############################################################################
@@ -241,15 +238,23 @@
 scores_nodelay = backend.to_numpy(scores_nodelay)
 
 ###############################################################################
-# Here we plot the comparison of model performances with a 2D histogram.
-# All ~70k voxels are represented in this histogram, where the diagonal
-# corresponds to identical performance for both models. A distibution deviating
-# from the diagonal means that one model has better predictive performances
-# than the other.
+# Here we plot the comparison of model performances with a 2D histogram. All
+# ~70k voxels are represented in this histogram, where the diagonal corresponds
+# to identical performance for both models. A distibution deviating from the
+# diagonal means that one model has better predictive performances than the
+# other.
 
 from voxelwise_tutorials.viz import plot_hist2d
 
 ax = plot_hist2d(scores_nodelay, scores)
 ax.set(title='Generalization R2 scores', xlabel='model without delays',
        ylabel='model with delays')
 plt.show()
+
+###############################################################################
+# References
+# ----------
+#
+# .. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu,
+#     B., & Gallant, J. L. (2011). Reconstructing visual experiences from brain
+#     activity evoked by natural movies. Current Biology, 21(19), 1641-1646.

tutorials/notebooks/movies_3T/00_download_vim4.ipynb

@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Download the data set from CRCNS\n\n\nIn this script, we download the data set from CRCNS. A (free) account is\nrequired.\n\n.. Warning:: The data has not been publicly released yet, so this notebook will\n    not work !\n\nCite this data set\n------------------\n\nThis tutorial is based on publicly available data `published on CRCNS\n<https://crcns.org/data-sets/vc/TBD>`_. If you publish any work using this data\nset, please cite the original publication [1]_, and the data set [2]_.\n\n.. [1] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2012). A\n    continuous semantic space describes the representation of thousands of\n    object and action categories across the human brain. Neuron, 76(6),\n    1210-1224.\n\n.. [2] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2020): Gallant\n    Lab Natural Movie 3T fMRI Data. CRCNS.org. http://dx.doi.org/10.6080/TBD\n"
+        "\n# Download the data set from CRCNS\n\n\nIn this script, we download the data set from CRCNS. A (free) account is\nrequired.\n\n.. Warning:: The data has not been publicly released yet, so this notebook will\n    not work !\n\nCite this data set\n------------------\n\nThis tutorial is based on publicly available data `published on CRCNS\n<https://crcns.org/data-sets/vc/TBD>`_. If you publish any work using this data\nset, please cite the original publication [1]_, and the data set [2]_.\n"
       ]
     },
     {
@@ -75,6 +75,13 @@
       "source": [
         "username = input(\"CRCNS username: \")\npassword = getpass.getpass(\"CRCNS password: \")\n\nfor datafile in DATAFILES:\n    local_filename = download_crcns(datafile, username, password,\n                                    destination=directory)"
       ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "References\n----------\n\n.. [1] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2012). A\n    continuous semantic space describes the representation of thousands of\n    object and action categories across the human brain. Neuron, 76(6),\n    1210-1224.\n\n.. [2] Huth, A. G., Nishimoto, S., Vu, A. T., & Gallant, J. L. (2020):\n    Gallant Lab Natural Movie 3T fMRI Data. CRCNS.org.\n    http://dx.doi.org/10.6080/TBD\n\n"
+      ]
     }
   ],
   "metadata": {

tutorials/notebooks/movies_3T/04_plot_motion_energy_model.ipynb

@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Fit a ridge model with motion energy features\n\n\nIn this example, we model the fMRI responses with motion-energy features\nextracted from the movie stimulus. The model is a regularized linear regression\nmodel.\n\nThis tutorial reproduces part of the analysis described in Nishimoto et al\n(2011) [1]_. See this publication for more details about the experiment, the\nmotion-energy features, along with more results and more discussions.\n\n*Motion-energy features:* Motion-energy features result from filtering a video\nstimulus with spatio-temporal Gabor filters. A pyramid of filters is used to\ncompute the motion-energy features at multiple spatial and temporal scales.\n\n*Summary:* As in the previous example, we first concatenate the features with\nmultiple delays, to account for the slow hemodynamic response. A linear\nregression model then weights each delayed feature with a different weight, to\nbuild a predictive model of BOLD activity. Again, the linear regression is\nregularized to improve robustness to correlated features and to improve\ngeneralization. The optimal regularization hyperparameter is selected\nindependently on each voxel over a grid-search with cross-validation. Finally,\nthe model generalization performance is evaluated on a held-out test set,\ncomparing the model predictions with the ground-truth fMRI responses.\n"
+        "\n# Fit a ridge model with motion energy features\n\n\nIn this example, we model the fMRI responses with motion-energy features\nextracted from the movie stimulus. The model is a regularized linear regression\nmodel.\n\nThis tutorial reproduces part of the analysis described in Nishimoto et al\n(2011) [1]_. See this publication for more details about the experiment, the\nmotion-energy features, along with more results and more discussions.\n\n*Motion-energy features:* Motion-energy features result from filtering a video\nstimulus with spatio-temporal Gabor filters. A pyramid of filters is used to\ncompute the motion-energy features at multiple spatial and temporal scales.\nMotion-energy features were introduced in [1]_.\n\n*Summary:* As in the previous example, we first concatenate the features with\nmultiple delays, to account for the slow hemodynamic response. A linear\nregression model then weights each delayed feature with a different weight, to\nbuild a predictive model of BOLD activity. Again, the linear regression is\nregularized to improve robustness to correlated features and to improve\ngeneralization. The optimal regularization hyperparameter is selected\nindependently on each voxel over a grid-search with cross-validation. Finally,\nthe model generalization performance is evaluated on a held-out test set,\ncomparing the model predictions with the ground-truth fMRI responses.\n"
       ]
     },
     {

tutorials/notebooks/movies_3T/06_extract_motion_energy.ipynb

@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Extract motion energy features from the stimuli\n\n\nThis script describes how to extract motion-energy features from the stimuli.\n\n.. Note:: This public data set already contains precomputed motion-energy.\n    Therefore, you do not need to run this script to fit motion-energy models\n    in other part of this tutorial.\n\n*Motion-energy features:* Motion-energy features result from filtering a video\nstimulus with spatio-temporal Gabor filters. A pyramid of filters is used to\ncompute the motion-energy features at multiple spatial and temporal scales.\nMotion-energy features were introduced in [1]_.\n\nThe motion-energy extraction is performed by the package `pymoten\n<https://github.com/gallantlab/pymoten>`_.\n\nRunning time\n------------\nExtracting motion energy is a bit longer than the other examples. It typically\ntakes a couple hours to run.\n\n.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &\n    Gallant, J. L. (2011). Reconstructing visual experiences from brain\n    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.\n"
+        "\n# Extract motion energy features from the stimuli\n\n\nThis script describes how to extract motion-energy features from the stimuli.\n\n.. Note:: This public data set already contains precomputed motion-energy.\n    Therefore, you do not need to run this script to fit motion-energy models\n    in other part of this tutorial.\n\n*Motion-energy features:* Motion-energy features result from filtering a video\nstimulus with spatio-temporal Gabor filters. A pyramid of filters is used to\ncompute the motion-energy features at multiple spatial and temporal scales.\nMotion-energy features were introduced in [1]_.\n\nThe motion-energy extraction is performed by the package `pymoten\n<https://github.com/gallantlab/pymoten>`_. Check the pymoten `gallery of\nexamples <https://gallantlab.github.io/pymoten/auto_examples/index.html>`_ for\nvisualizing motion-energy filters, and for pymoten API usage examples.\n\nRunning time\n------------\nExtracting motion energy is a bit longer than the other examples. It typically\ntakes a couple hours to run.\n"
       ]
     },
     {

tutorials/notebooks/movies_4T/00_download_vim2.ipynb

@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Download the data set from CRCNS\n\n\nIn this script, we download the data set from CRCNS.\nA (free) account is required.\n\nCite this data set\n------------------\n\nThis tutorial is based on publicly available data\n`published on CRCNS <https://crcns.org/data-sets/vc/vim-2/about-vim-2>`_.\nIf you publish any work using this data set, please cite the original\npublication [1]_, and the data set [2]_.\n\n.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &\n    Gallant, J. L. (2011). Reconstructing visual experiences from brain\n    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.\n\n.. [2] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &\n    Gallant, J. L. (2014): Gallant Lab Natural Movie 4T fMRI Data. CRCNS.org.\n    http://dx.doi.org/10.6080/K00Z715X\n"
+        "\n# Download the data set from CRCNS\n\n\nIn this script, we download the data set from CRCNS.\nA (free) account is required.\n\nCite this data set\n------------------\n\nThis tutorial is based on publicly available data\n`published on CRCNS <https://crcns.org/data-sets/vc/vim-2/about-vim-2>`_.\nIf you publish any work using this data set, please cite the original\npublication [1]_, and the data set [2]_.\n"
       ]
     },
     {
@@ -75,6 +75,13 @@
       "source": [
         "username = input(\"CRCNS username: \")\npassword = getpass.getpass(\"CRCNS password: \")\n\nfor datafile in DATAFILES:\n    local_filename = download_crcns(datafile, username, password,\n                                    destination=directory, unpack=True)"
       ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "References\n----------\n\n.. [1] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &\n    Gallant, J. L. (2011). Reconstructing visual experiences from brain\n    activity evoked by natural movies. Current Biology, 21(19), 1641-1646.\n\n.. [2] Nishimoto, S., Vu, A. T., Naselaris, T., Benjamini, Y., Yu, B., &\n    Gallant, J. L. (2014): Gallant Lab Natural Movie 4T fMRI Data. CRCNS.org.\n    http://dx.doi.org/10.6080/K00Z715X\n\n"
+      ]
     }
   ],
   "metadata": {