@@ -58,7 +58,7 @@
 # number of Bakeries to Cafés in a City (in this case Manhattan). We did a
 # quick google map search in Manhattan for bakeries and Cafés:
 #
-# .. image:: images/bak.png
+# .. image:: ../_static/images/bak.png
 #     :align: center
 #     :alt: bakery-cafe-manhattan
 #     :width: 600px
@@ -233,7 +233,7 @@
 # The Sinkhorn algorithm is very simple to code. You can implement it directly
 # using the following pseudo-code
 #
-# .. image:: images/sinkhorn.png
+# .. image:: ../_static/images/sinkhorn.png
 #     :align: center
 #     :alt: Sinkhorn algorithm
 #     :width: 440px
 
@@ -69,7 +69,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "## First OT Problem\n\nWe will solve the Bakery/Caf\u00e9s problem of transporting croissants from a\nnumber of Bakeries to Caf\u00e9s in a City (in this case Manhattan). We did a\nquick google map search in Manhattan for bakeries and Caf\u00e9s:\n\n<img src=\"file://images/bak.png\" align=\"center\" alt=\"bakery-cafe-manhattan\" width=\"600px\" height=\"280px\">\n\nWe extracted from this search their positions and generated fictional\nproduction and sale number (that both sum to the same value).\n\nWe have acess to the position of Bakeries ``bakery_pos`` and their\nrespective production ``bakery_prod`` which describe the source\ndistribution. The Caf\u00e9s where the croissants are sold are defined also by\ntheir position ``cafe_pos`` and ``cafe_prod``, and describe the target\ndistribution. For fun we also provide a\nmap ``Imap`` that will illustrate the position of these shops in the city.\n\n\nNow we load the data\n\n\n\n"
+        "## First OT Problem\n\nWe will solve the Bakery/Caf\u00e9s problem of transporting croissants from a\nnumber of Bakeries to Caf\u00e9s in a City (in this case Manhattan). We did a\nquick google map search in Manhattan for bakeries and Caf\u00e9s:\n\n<img src=\"file://../_static/images/bak.png\" align=\"center\" alt=\"bakery-cafe-manhattan\" width=\"600px\" height=\"280px\">\n\nWe extracted from this search their positions and generated fictional\nproduction and sale number (that both sum to the same value).\n\nWe have acess to the position of Bakeries ``bakery_pos`` and their\nrespective production ``bakery_prod`` which describe the source\ndistribution. The Caf\u00e9s where the croissants are sold are defined also by\ntheir position ``cafe_pos`` and ``cafe_prod``, and describe the target\ndistribution. For fun we also provide a\nmap ``Imap`` that will illustrate the position of these shops in the city.\n\n\nNow we load the data\n\n\n\n"
       ]
     },
     {
@@ -198,7 +198,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "## Regularized OT with Sinkhorn\n\nThe Sinkhorn algorithm is very simple to code. You can implement it directly\nusing the following pseudo-code\n\n<img src=\"file://images/sinkhorn.png\" align=\"center\" alt=\"Sinkhorn algorithm\" width=\"440px\" height=\"240px\">\n\nIn this algorithm, $\\oslash$ corresponds to the element-wise division.\n\nAn alternative is to use the POT toolbox with\n`ot.sinkhorn <https://pythonot.github.io/all.html#ot.sinkhorn>`_\n\nBe careful of numerical problems. A good pre-processing for Sinkhorn is to\ndivide the cost matrix ``C`` by its maximum value.\n\n"
+        "## Regularized OT with Sinkhorn\n\nThe Sinkhorn algorithm is very simple to code. You can implement it directly\nusing the following pseudo-code\n\n<img src=\"file://../_static/images/sinkhorn.png\" align=\"center\" alt=\"Sinkhorn algorithm\" width=\"440px\" height=\"240px\">\n\nIn this algorithm, $\\oslash$ corresponds to the element-wise division.\n\nAn alternative is to use the POT toolbox with\n`ot.sinkhorn <https://pythonot.github.io/all.html#ot.sinkhorn>`_\n\nBe careful of numerical problems. A good pre-processing for Sinkhorn is to\ndivide the cost matrix ``C`` by its maximum value.\n\n"
       ]
     },
     {
Original file line number	Diff line number	Diff line change
`@@ -69,7 +69,7 @@`
`69`	`69`	`"cell_type": "markdown",`
`70`	`70`	`"metadata": {},`
`71`	`71`	`"source": [`
`72`		- "## First OT Problem\n\nWe will solve the Bakery/Caf\u00e9s problem of transporting croissants from a\nnumber of Bakeries to Caf\u00e9s in a City (in this case Manhattan). We did a\nquick google map search in Manhattan for bakeries and Caf\u00e9s:\n\n<img src=\"file://images/bak.png\" align=\"center\" alt=\"bakery-cafe-manhattan\" width=\"600px\" height=\"280px\">\n\nWe extracted from this search their positions and generated fictional\nproduction and sale number (that both sum to the same value).\n\nWe have acess to the position of Bakeries ``bakery_pos`` and their\nrespective production ``bakery_prod`` which describe the source\ndistribution. The Caf\u00e9s where the croissants are sold are defined also by\ntheir position ``cafe_pos`` and ``cafe_prod``, and describe the target\ndistribution. For fun we also provide a\nmap ``Imap`` that will illustrate the position of these shops in the city.\n\n\nNow we load the data\n\n\n\n"
	`72`	+ "## First OT Problem\n\nWe will solve the Bakery/Caf\u00e9s problem of transporting croissants from a\nnumber of Bakeries to Caf\u00e9s in a City (in this case Manhattan). We did a\nquick google map search in Manhattan for bakeries and Caf\u00e9s:\n\n<img src=\"file://../_static/images/bak.png\" align=\"center\" alt=\"bakery-cafe-manhattan\" width=\"600px\" height=\"280px\">\n\nWe extracted from this search their positions and generated fictional\nproduction and sale number (that both sum to the same value).\n\nWe have acess to the position of Bakeries ``bakery_pos`` and their\nrespective production ``bakery_prod`` which describe the source\ndistribution. The Caf\u00e9s where the croissants are sold are defined also by\ntheir position ``cafe_pos`` and ``cafe_prod``, and describe the target\ndistribution. For fun we also provide a\nmap ``Imap`` that will illustrate the position of these shops in the city.\n\n\nNow we load the data\n\n\n\n"
`73`	`73`	`]`
`74`	`74`	`},`
`75`	`75`	`{`
`@@ -198,7 +198,7 @@`
`198`	`198`	`"cell_type": "markdown",`
`199`	`199`	`"metadata": {},`
`200`	`200`	`"source": [`
`201`		- "## Regularized OT with Sinkhorn\n\nThe Sinkhorn algorithm is very simple to code. You can implement it directly\nusing the following pseudo-code\n\n<img src=\"file://images/sinkhorn.png\" align=\"center\" alt=\"Sinkhorn algorithm\" width=\"440px\" height=\"240px\">\n\nIn this algorithm, $\\oslash$ corresponds to the element-wise division.\n\nAn alternative is to use the POT toolbox with\n`ot.sinkhorn <https://pythonot.github.io/all.html#ot.sinkhorn>`_\n\nBe careful of numerical problems. A good pre-processing for Sinkhorn is to\ndivide the cost matrix ``C`` by its maximum value.\n\n"
	`201`	+ "## Regularized OT with Sinkhorn\n\nThe Sinkhorn algorithm is very simple to code. You can implement it directly\nusing the following pseudo-code\n\n<img src=\"file://../_static/images/sinkhorn.png\" align=\"center\" alt=\"Sinkhorn algorithm\" width=\"440px\" height=\"240px\">\n\nIn this algorithm, $\\oslash$ corresponds to the element-wise division.\n\nAn alternative is to use the POT toolbox with\n`ot.sinkhorn <https://pythonot.github.io/all.html#ot.sinkhorn>`_\n\nBe careful of numerical problems. A good pre-processing for Sinkhorn is to\ndivide the cost matrix ``C`` by its maximum value.\n\n"
`202`	`202`	`]`
`203`	`203`	`},`
`204`	`204`	`{`