Added disclaimer, change links such as to build from top root directory

Author: rougier

README.md

@@ -5,3 +5,11 @@ This work is licensed under a
 (http://creativecommons.org/licenses/by/4.0/).
 
 **Work in progress. No due date yet.**
+
+If you want to contribute to this book, you can:
+
+* Review chapters (please contact me)
+* Report issues (https://github.com/rougier/numpy-book/issues)
+* Suggest improvements (https://github.com/rougier/numpy-book/pulls)
+* Correct English (https://github.com/rougier/numpy-book/issues)
+* Design a better and more responsive html template for the book.

book/beyond-numpy.rst

@@ -14,7 +14,7 @@ specific cases, it might be more powerful than Numpy. Let's consider for
 example an interesting exercise that has been proposed by Tucker Balch in his
 `Coursera's Computational Investing
 <https://www.coursera.org/learn/computational-investing>`_ course. The exercise
-can be written as:
+is written as:
 
 Write the most succinct code possible to compute all "legal" allocations to 4
 stocks such that:

book/book.rst

@@ -16,13 +16,13 @@
 
 .. container:: title-logos
 
-   .. image:: ../data/cc.large.png
+   .. image:: data/cc.large.png
       :width: 40px
-   .. image:: ../data/by.large.png
+   .. image:: data/by.large.png
       :width: 40px
-   .. image:: ../data/sa.large.png
+   .. image:: data/sa.large.png
       :width: 40px
-   .. image:: ../data/nc.large.png
+   .. image:: data/nc.large.png
       :width: 40px
 
    (J. Kerouac style)
@@ -31,7 +31,7 @@
 .. ----------------------------------------------------------------------------
 .. container:: title-logos
 
-   .. image:: ../data/cubes.png
+   .. image:: data/cubes.png
       :width: 100%
 
 .. ----------------------------------------------------------------------------
@@ -51,6 +51,14 @@ make you acquire experience in the process.
    :class: main-content
    :depth: 2
 
+|
+|
+
+**Disclaimer** All external pictures should have associated credits. If there
+are missing credits, please tell me, I will correct it. Similarly, all excerpts
+should be sourced (wikipedia mostly). If not, this is an error and I will
+correct it as soon as you tell me.
+           
 .. ----------------------------------------------------------------------------
 .. |WIP| image:: https://img.shields.io/badge/status-WIP-orange.svg?style=flat-square
 

book/code-vectorization.rst

@@ -85,7 +85,7 @@ see how to vectorize it.
    Conus textile snail exhibits a cellular automaton pattern on its shell.
    Image by `Richard Ling <https://commons.wikimedia.org/wiki/File:Textile_cone.JPG>`_, 2005.
 
-.. image:: ../data/Textile-Cone-cropped.jpg
+.. image:: data/Textile-Cone-cropped.jpg
    :width: 100%
    :class: bordered
 
@@ -191,7 +191,7 @@ discovered by Richard K. Guy in 1970.
    The glider pattern is known to replicate itself one step diagonally in 4
    iterations.
 
-.. image:: ../data/glider.png
+.. image:: data/glider.png
    :width: 100%
 
 
@@ -287,7 +287,7 @@ cleared it.
 .. raw:: html
 
          <video width="100%" controls>
-         <source src="../data/game-of-life.mp4" type="video/mp4">
+         <source src="data/game-of-life.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
 
@@ -358,25 +358,25 @@ The figure below show some animation of the model for a specific set of paramete
 .. raw:: html
 
          <video width="33%" controls>
-         <source src="../data/gray-scott-1.mp4" type="video/mp4">
+         <source src="data/gray-scott-1.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
          <video width="33%" controls>
-         <source src="../data/gray-scott-2.mp4" type="video/mp4">
+         <source src="data/gray-scott-2.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
          <video width="33%" controls>
-         <source src="../data/gray-scott-3.mp4" type="video/mp4">
+         <source src="data/gray-scott-3.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
 
 
 Sources
 +++++++
 
-* `game-of-life-python.py <../code/game-of-life-python.py>`_
-* `game-of-life-numpy.py <../code/game-of-life-numpy.py>`_
-* `gray-scott.py <../code/gray-scott.py>`_ (solution to the exercise)
+* `game-of-life-python.py <code/game-of-life-python.py>`_
+* `game-of-life-numpy.py <code/game-of-life-numpy.py>`_
+* `gray-scott.py <code/gray-scott.py>`_ (solution to the exercise)
 
 
 References
@@ -423,7 +423,7 @@ divergent. Of course, the more iteration you do, the more precision you get.
    Romanesco broccoli, showing self-similar form approximating a natural fractal.
    Image by `Jon Sullivan <https://commons.wikimedia.org/wiki/File:Fractal_Broccoli.jpg>`_, 2004.
 
-.. image:: ../data/Fractal-Broccoli-cropped.jpg
+.. image:: data/Fractal-Broccoli-cropped.jpg
    :width: 100%
    :class: bordered
 
@@ -561,7 +561,7 @@ and added a power normalized colormap (gamma=0.3) as well as light shading.
    The Mandelbrot as rendered by maplotlib using recount normalization, power
    normalized colormap (gamma=0.3) and light shading.
 
-.. figure:: ../data/mandelbrot.png
+.. figure:: data/mandelbrot.png
    :width: 100%
    :class: bordered
 
@@ -590,18 +590,18 @@ normalized (i.e. all values are between 0 and 1).
    The Minkowski–Bouligand dimension of the Great Britain coastlines is
    approximately 1.24.
 
-.. figure:: ../data/fractal-dimension.png
+.. figure:: data/fractal-dimension.png
    :width: 100%
 
 
 Sources
 +++++++
 
-* `mandelbrot.py <../code/mandelbrot.py>`_
-* `mandelbrot-python.py <../code/mandelbrot-python.py>`_
-* `mandelbrot-numpy-1.py <../code/mandelbrot-numpy-1.py>`_
-* `mandelbrot-numpy-2.py <../code/mandelbrot-numpy-2.py>`_
-* `fractal-dimension.py <../code/fractal-dimension.py>`_ (solution to the exercise)
+* `mandelbrot.py <code/mandelbrot.py>`_
+* `mandelbrot-python.py <code/mandelbrot-python.py>`_
+* `mandelbrot-numpy-1.py <code/mandelbrot-numpy-1.py>`_
+* `mandelbrot-numpy-2.py <code/mandelbrot-numpy-2.py>`_
+* `fractal-dimension.py <code/fractal-dimension.py>`_ (solution to the exercise)
 
 References
 ++++++++++
@@ -629,7 +629,7 @@ boids that simulate flocking behaviors.
    Flocking birds are an example of self-organization in biology.
    Image by `Christoffer A Rasmussen <https://commons.wikimedia.org/wiki/File:Fugle,_ørnsø_073.jpg>`_, 2012.
 
-.. image:: ../data/Fugle-cropped.jpg
+.. image:: data/Fugle-cropped.jpg
    :width: 100%
    :class: bordered
 
@@ -662,7 +662,7 @@ rules. The rules applied in the simplest Boids world are as follows:
    Boids are governed by a set of three local rules (separation, cohesion and
    alignment) that serve as computing velocity and acceleration.
 
-.. image:: ../data/boids.png
+.. image:: data/boids.png
    :width: 100%
 
 Python implementation
@@ -883,7 +883,7 @@ We finally visualize the result using a custom oriented scatter plot.
 .. raw:: html
 
          <video width="100%" class="bordered" controls>
-         <source src="../data/boids.mp4" type="video/mp4">
+         <source src="data/boids.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
 
@@ -930,8 +930,8 @@ How would you write the `translate`, `scale` and `rotate` functions ?
 Sources
 +++++++
 
-* `boid-python.py <../code/boid-python.py>`_
-* `boid-numpy.py <../code/boid-numpy.py>`_ (solution to the exercise)
+* `boid-python.py <code/boid-python.py>`_
+* `boid-numpy.py <code/boid-numpy.py>`_ (solution to the exercise)
 
 References
 ++++++++++

book/preface.rst

@@ -34,13 +34,13 @@ About this book
 This book has been written in |ReST|_ format and generated using the
 `rst2html.py` command line available from the docutils_ python package.
 
-If you want to rebuild the html output, use:
+If you want to rebuild the html output, from top direcotry, type:
 
 .. code-block::
 
    $ rst2html.py --link-stylesheet --cloak-email-addresses \
-                 --toc-top-backlinks --stylesheet=book.css \
-                   book.rst book.html
+                 --toc-top-backlinks --stylesheet=book/book.css \
+                   book/book.rst book.html
 
 The sources are available from https://github.com/rougier/numpy-book.
 

book/problem-vectorization.rst

@@ -177,7 +177,7 @@ maze. First task is thus to build a maze.
    A hedge maze at Longleat stately home in England.
    Image by `Prince Rurik <https://commons.wikimedia.org/wiki/File:Longleat_maze.jpg>`_, 2005.
 
-.. image:: ../data//Longleat-maze-cropped.jpg
+.. image:: data//Longleat-maze-cropped.jpg
    :width: 100%
    :class: bordered
 
@@ -255,7 +255,7 @@ Here is an animation showing the generation process.
 .. raw:: html
 
    <video width="100%" controls>
-   <source src="../data/maze-build.mp4" type="video/mp4">
+   <source src="data/maze-build.mp4" type="video/mp4">
    Your browser does not support the video tag. </video>
 
 Breadth-first
@@ -294,11 +294,11 @@ it into an actual graph as proposed by `Valentin Bryukhanov
    shortest path, only to find a path (if it exists).
 
 Once this is done, writing the breadth first algorithm is straightforward. We
-start from the starting node and we visit nodes one level (breadth first,
-remember?) and we iterate the process until reaching the final node, if
+start from the starting node and we visit nodes at current depth only (breadth
+first, remember?) and we iterate the process until reaching the final node, if
 possible. The question is then, do we get the shortest path exploring the graph
 this way? In this specific case yes because we don't have a e-weighted graph,
-i.e. all the edges have the same weight (or cost). 
+i.e. all the edges have the same weight (or cost).
 
 .. code:: python
 
@@ -334,26 +334,26 @@ done, we can ascent the gradient from the starting node. You can check on the
 figure this leads to the shortest path.
 
 
-.. image:: ../data/value-iteration-1.pdf
+.. image:: data/value-iteration-1.pdf
    :width: 19%
-.. image:: ../data/value-iteration-2.pdf
+.. image:: data/value-iteration-2.pdf
    :width: 19%
-.. image:: ../data/value-iteration-3.pdf
+.. image:: data/value-iteration-3.pdf
    :width: 19%
-.. image:: ../data/value-iteration-4.pdf
+.. image:: data/value-iteration-4.pdf
    :width: 19%
-.. image:: ../data/value-iteration-5.pdf
+.. image:: data/value-iteration-5.pdf
    :width: 19%
 
-.. image:: ../data/value-iteration-6.pdf
+.. image:: data/value-iteration-6.pdf
    :width: 19%
-.. image:: ../data/value-iteration-7.pdf
+.. image:: data/value-iteration-7.pdf
    :width: 19%
-.. image:: ../data/value-iteration-8.pdf
+.. image:: data/value-iteration-8.pdf
    :width: 19%
-.. image:: ../data/value-iteration-9.pdf
+.. image:: data/value-iteration-9.pdf
    :width: 19%
-.. image:: ../data/value-iteration-10.pdf
+.. image:: data/value-iteration-10.pdf
    :width: 19%
 
 We start by setting the exit node to the value 1 while every other nodes are
@@ -424,16 +424,16 @@ illustrated on the figure below:
    propagated values from the end-point of the maze (bottom-right). Path is
    found by ascending gradient from the goal.
 
-.. image:: ../data/maze.png
+.. image:: data/maze.png
    :width: 100%
 
 
 
 Sources
 +++++++
 
-* `maze-build.py <../code/maze-build.py>`_
-* `maze-numpy.py <../code/maze-numpy.py>`_
+* `maze-build.py <code/maze-build.py>`_
+* `maze-numpy.py <code/maze-numpy.py>`_
 
 References
 ++++++++++
@@ -455,7 +455,7 @@ Fluid Dynamics
    Germany. Image by `Steven Mathey
    <https://commons.wikimedia.org/wiki/File:Self_Similar_Turbulence.png>`_, 2012.
 
-.. image:: ../data/Self-similar-turbulence.png
+.. image:: data/Self-similar-turbulence.png
    :width: 100%
 
 
@@ -530,25 +530,25 @@ using this technique.
 .. raw:: html
 
          <video width="33%" controls>
-         <source src="../data/smoke-1.mp4" type="video/mp4">
+         <source src="data/smoke-1.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
          <video width="33%" controls>
-         <source src="../data/smoke-2.mp4" type="video/mp4">
+         <source src="data/smoke-2.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
          <video width="33%" controls>
-         <source src="../data/smoke-gpu.mp4" type="video/mp4">
+         <source src="data/smoke-gpu.mp4" type="video/mp4">
          Your browser does not support the video tag. </video>
 
 
 Sources
 +++++++
 
-* `smoke-1.py <../code/smoke-1.py>`_
-* `smoke-2.py <../code/smoke-2.py>`_
-* `smoke_solver.py <../code/smoke_solver.py>`_
-* `smoke-interactive.py <../code/smoke-interactive.py>`_
+* `smoke-1.py <code/smoke-1.py>`_
+* `smoke-2.py <code/smoke-2.py>`_
+* `smoke_solver.py <code/smoke_solver.py>`_
+* `smoke-interactive.py <code/smoke-interactive.py>`_
 
 
 References
@@ -569,7 +569,7 @@ Blue noise
 
    Detail of "The Starry Night", Vincent van Gogh, 1889.
 
-.. image:: ../data/mosaic.png
+.. image:: data/mosaic.png
    :width: 100%
    :class: bordered
 
@@ -587,15 +587,15 @@ Numpy implementation
    Comparison of uniform, grid-jittered and Poisson disc sampling.
 
 
-.. image:: ../data/sampling.png
+.. image:: data/sampling.png
    :width: 100%
 
 Sources
 +++++++
 
-* `sampling.py <../code/sampling.py>`_
-* `mosaic.py <../code/mosaic.py>`_
-* `voronoi.py <../code/voronoi.py>`_
+* `sampling.py <code/sampling.py>`_
+* `mosaic.py <code/mosaic.py>`_
+* `voronoi.py <code/voronoi.py>`_
 
 References
 ++++++++++

code/random_walk.py

@@ -50,12 +50,12 @@ def find_crossing_2(seq, subseq):
 
 
 if __name__ == "__main__":
-    from tools import print_timeit
+    from tools import timeit
 
-    print_timeit("random_walk_1(n=10000)", globals())
-    print_timeit("random_walk_2(n=10000)", globals())
-    print_timeit("random_walk_3(n=10000)", globals())
+    timeit("random_walk_1(n=10000)", globals())
+    timeit("random_walk_2(n=10000)", globals())
+    timeit("random_walk_3(n=10000)", globals())
     print()
     W = random_walk_3(n=1000)
-    print_timeit("find_crossing_1(list(W), [+1,0,-1])", globals())
-    print_timeit("find_crossing_2(W, [+1,0,-1])", globals())
+    timeit("find_crossing_1(list(W), [+1,0,-1])", globals())
+    timeit("find_crossing_2(W, [+1,0,-1])", globals())