Figure scaling tutorials

MelJan · Oct 14, 2018 · 71a5f4a · 71a5f4a
1 parent 1c099ff
commit 71a5f4a
Show file tree

Hide file tree

Showing 5 changed files with 16 additions and 0 deletions.
diff --git a/docs/tutorials/GRBM_2D_example.rst b/docs/tutorials/GRBM_2D_example.rst
@@ -22,12 +22,14 @@ The code_ given below produces the following output.
 Visualization of the weight vectors learned by the GRBM with 4 hidden units together with the contour plot of the learned probability density function (PDF).
 
 .. figure:: images/GRBM_2D_PDF.png
+   :scale: 75 %
    :align: center
    :alt: Visualization of the PDF learned by the GRBM
 
 For a better visualization also the log-PDF.
 
 .. figure:: images/GRBM_2D_logPDF.png
+   :scale: 75 %
    :align: center
    :alt: Visualization of the log PDF learned by the GRBM
 
@@ -63,10 +65,12 @@ The exact log-likelihood, annealed importance sampling estimation, and reverse a
 For comparison here is the original mixing matrix an the corresponding ICA estimation.
 
 .. image:: images/GRBM_2D_MIX.png
+   :scale: 75 %
    :align: center
    :alt: Examples of mixing matrix 2D
 
 .. image:: images/GRBM_2D_ICA.png
+   :scale: 75 %
    :align: center
    :alt: Examples ICA estimation of the mixing matrix.
 

diff --git a/docs/tutorials/ICA_2D_example.rst b/docs/tutorials/ICA_2D_example.rst
@@ -32,6 +32,7 @@ The code_ given below produces the following output.
 Visualization of the data and true mixing matrix projected to the whitened space.
 
 .. figure:: images/ICA_2D_mixing_whitened.png
+   :scale: 75 %
    :align: center
    :alt: Examples of mixing matrix 2D in whitened space
 
@@ -45,16 +46,19 @@ Amari distanca between true mixing matrix and estimated mixing matrix:
    0.00989836830489
 
 .. image:: images/ICA_2D_estimate_whitened.png
+   :scale: 75 %
    :align: center
    :alt: Examples of ICA 2D in whitened space
 
 We can also project the ICA projection matrix back to the original space and compare the results in the original space.
 
 .. image:: images/ICA_2D_mixing.png
+   :scale: 75 %
    :align: center
    :alt: Examples of mixing matrix 2D
 
 .. image:: images/ICA_2D_estimate.png
+   :scale: 75 %
    :align: center
    :alt: Examples of ICA 2D
 

diff --git a/docs/tutorials/PCA_2D_example.rst b/docs/tutorials/PCA_2D_example.rst
@@ -24,18 +24,21 @@ The code_ given below produces the following output.
 The data is plotted with the extracted principal components.
 
 .. figure:: images/PCA_2D.png
+   :scale: 75 %
    :align: center
    :alt: Examples of PCA 2D
 
 Data and extracted principal components can also be plotted in the projected space.
 
 .. image:: images/PCA_2D_projected.png
+   :scale: 75 %
    :align: center
    :alt: Examples of PCA 2D in projected space
 
 The PCA-class can also perform whitening. Data and extracted principal components are plotted in the whitened space.
 
 .. image:: images/PCA_2D_whitened.png
+   :scale: 75 %
    :align: center
    :alt: Examples of PCA 2D in whitened space
 

diff --git a/docs/tutorials/PCA_eigenfaces.rst b/docs/tutorials/PCA_eigenfaces.rst
@@ -30,6 +30,7 @@ The first 100 principal components extracted from the dataset. The components fo
 The cumulative sum of the Eigenvalues show how 'compressable' the dataset is.
 
 .. image:: images/eigenspectrum_faces.png
+   :scale: 75 %
    :align: center
    :alt: Eigenspectrum of the face dataset
 

diff --git a/docs/tutorials/RBM_MNIST_small.rst b/docs/tutorials/RBM_MNIST_small.rst
@@ -38,6 +38,7 @@ Learned filters of a centered binary RBM on the MNIST dataset.
 The filters have been normalized such that the structure is more prominent.
 
 .. figure:: images/BRBM_small_centered_weights.png
+   :scale: 50 %
    :align: center
    :alt: weights centered
 
@@ -64,6 +65,7 @@ The code can also be executed without centering by setting
 Resulting in the following weights and sampling steps.
 
 .. figure:: images/BRBM_small_normal_weights.png
+   :scale: 50 %
    :alt: weights normal
    :align: center
 
@@ -94,6 +96,7 @@ While the centered model has a similar performance on the flipped version,
    reverse AIS Partition:  309.188366165 (LL train: -141.755241282, LL test: -141.029631984)
 
 .. figure:: images/BRBM_small_centered_weights_flipped.png
+   :scale: 50 %
    :alt: flipped filters centered
    :align: center
 
@@ -110,6 +113,7 @@ The normal RBM has not.
    reverse AIS Partition:  3495.20117625 (LL train: -183.188469308, LL test: -183.289157393)
 
 .. figure:: images/BRBM_small_normal_weights_flipped.png
+   :scale: 50 %
    :alt: flipped filters normal
    :align: center