Correction mise en page, liens cliquables, image workflow modifiée, c…

…orrection explications

Melanoma_starting_kit.ipynb

@@ -25,15 +25,17 @@
     "2. [Data](#Data)\n",
     "3. [Workflow](#Workflow)\n",
     "4. [Evaluation](#Scoring)\n",
-    "5. [Local testing/exploration](#Local-testing)\n",
-    "6. [Submission](#Submitting-to-ramp.studio)"
+    "5. [Baseline Model](#Baseline-Model)\n",
+    "6. [Local testing](#Local-testing)\n",
+    "7. [Submission](#Submitting-to-ramp.studio)\n",
+    "8. [More information](#More-information)"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Introduction\n",
+    "# Introduction\n",
     "\n",
     "Skin diseases are very common, especially melanoma and require full attention. Indeed, the statistics are striking : Melanoma skin cancer is the 5th most common cancer in the UK, accounting for 4% of all new cancer cases (2015). \n",
     "1 person dies of melanoma every hour in North America, 1 out of 5 people in the USA will eventually contract a skin cancer during their lifetime and an interesting fact motivating our challenge : 70% of skin cancers are first detected by the patients themselves or their relatives.\n",
@@ -147,7 +149,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Data"
+    "# Data"
    ]
   },
   {
@@ -579,22 +581,29 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "<img src=\"./img/workflow.png\" />\n",
-    "<div style=\"text-align: center; font-style: italic\">Example of workflow</div>\n",
-    "<br />"
+    "# Workflow"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<img src=\"./img/workflow.png\" />\n"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "This Workflow allows to recognize the melanomas of the skins with a classification. If the model is well trained, wich means that the percentage of false-negative is low and the accurate is better than 90% then it does not need to be trained very often. In the fact, it has to be retrained when a new category of melanoma is discover. Indeed if the model is not retrained in this case, the rate of false-positive will increase and a lot of people will believe they are safe whearas they have a melanoma."
+    "This example of Workflow allows to build a full machine learning pipeline to recognize abnormal images of moles. If the model is well trained and robust enough, wich roughly means that the percentage of false-negative is low enough and the accuracy is satisfactory, then the model doesn't need further training and can remain still. In the case of a newly type of melanoma or mole disease discovery, then the algorithm needs to be retrained to include this new type. Ideally, since the classifier should be included in a smartphone application, the database should be regularly enriched with newly amateur photos to reflect as much as possible real life situations."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "# Scoring\n",
+    "\n",
     "## Prediction value measurement and quantitative score(s)"
    ]
   },
@@ -845,7 +854,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Image processing example"
+    "# Baseline Model"
    ]
   },
   {
@@ -1649,7 +1658,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Local testing/exploration\n",
+    "# Local testing\n",
     "\n",
     "It is <b><span style=\"color:red\">important that you test your submission files before submitting them</span></b>. For this we provide a unit test. Note that the test runs on your files in [`submissions/starting_kit`](/tree/submissions/starting_kit), not on the classes defined in the cells of this notebook.\n",
     "\n",
@@ -1743,7 +1752,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Submitting to [ramp.studio](http://ramp.studio)\n",
+    "# Submitting to [ramp.studio](http://ramp.studio)\n",
     "\n",
     "Once you found a good model, you can submit them to [ramp.studio](https://www.ramp.studio). First, if it is your first time using RAMP, [sign up](https://www.ramp.studio/sign_up), otherwise [log in](https://www.ramp.studio/login). Then find an open event on the particular problem, the event [mars_craters](https://www.ramp.studio/events/mars_craters_cifar) for this RAMP. Sign up for the event. Both signups are controled by RAMP administrators, so there **can be a delay between asking for signup and being able to submit**.\n",
     "\n",
@@ -1780,7 +1789,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## More information\n",
+    "# More information\n",
     "\n",
     "You can find more information in the [README](https://github.com/paris-saclay-cds/ramp-workflow/blob/master/README.md) of the [ramp-workflow library](https://github.com/paris-saclay-cds/ramp-workflow)."
    ]
@@ -1803,7 +1812,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.6"
+   "version": "3.6.7"
   },
   "nav_menu": {},
   "toc": {