balouf · October 13, 2021 08:34
diff --git a/first_notebook.ipynb b/first_notebook.ipynb
 {
  "cells": [
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "# Python 101"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "## Introduction"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "# Why Python?"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- A very popular language (on the podium, sometimes #1)\n- Easy to learn:\n  - `1+1`\n  - `print(\"Hello World\")`\n- Powerful to master:\n  - Build complex packages\n  - Publish them!\n- Huge ecosystem of packages and stackoverflow questions\n  - Including all ML stuff (Pytorch)\n  - Also some esoteric stuff"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "# Why NOT Python?"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- Slower than C (but you can learn to mitigate)\n- I need a very specific feature not available (rare but may occur)\n- I am fully happy with language X (no problem!)"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "# Python 2.X vs Python 3.Y"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- Big changes have been introduced in Python 3\n- Many legacy softwares have stayed 2.X for simplicity\n- As a beginner in Python:\n  - Use Python 3.X. Period.\n  - Avoid to be involved in any 2.X project if you can.\n  - Be careful with 2.X solutions in stackoverflow\n"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "# How to program in Python?"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Very roughly, you have three main ways to use Python:\n- Directly inside a Python interpreter\n  - Check something, make a one-liner\n- Write a python file (.py) that contains instructions to be executed\n  - All-purposes, from small scripts to huge packages\n- Inside an interactive Python notebook (.ipynb)\n  - Ideal to create new code and play with it"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "# How to program in Python?"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "To do python, you need:\n- A Python distribution\n- IDE for your notebooks / python files\n\nMany (many many) possibilities available. We focus on:\n- Jupyter Notebook\n- PyCharm"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "## Installation"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "# The Anaconda distribution"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "https://www.anaconda.com/products/individual\n- Pros:\n  - All-in-one\n  - Very easy to install\n  - Can be installed on user account (highly recommended for beginners)\n- Cons:\n  - 400 Mo\n  - Easy to break ;)\n  - Need to re-install for changing Python version"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "# Installing Anaconda"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- Tip #1: no space in the install directory\n- Tip #2: adding Anaconda to Path is not recommended, but it helps!\n- When you’re done, start an interpreter\n  - `1+1`\n  - `print(\"Hello World\")`\n- Optional: tour of the Anaconda Navigator\n  - Possibly better for beginners?\n  - Bad heaviness / usefulness trade-off"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "# Updating"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- Update is tricky:\n  - Necessary to get latest features\n  - Easy to inadvertently break things\n  - Safest moment to update is on a fresh install!\n- Multiple ways to update/install packages in Python\n  - Conda: the Anaconda internal package manager\n  - Pip: universal package manager\n  - ...\n- My rule of thumb:\n  - Conda-related stuff (e.g. notebooks) -> conda\n  - All the rest -> pip if possible"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "# Update intructions"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- `conda update conda`\n- `conda update --all`\n\nWhen ready, you can start your first notebook:\n\n`jupyter notebook`"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "## Your first notebook"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "A notebook is made of cells. A cell is a unit of content."
    },
    {
      "metadata": {},
      "cell_type": "raw",
      "source": "This is a raw Cell."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": ""
        }
      },
      "cell_type": "markdown",
      "source": "This is a *markdown* cell."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T19:34:42.895182Z",
          "start_time": "2021-10-12T19:34:42.868203Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "# This is a code cell\nprint(\"Hello world!\")",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### Navigation and shortcuts"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Jupyter has two main modes:\n- Command mode to navigate and manipulate entire cells;\n- Edit mode to change the content of a cell."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Recognize and switch mode:\n- Command mode (blue box): Select a cell (e.g. with arrows) and hit enter to switch to edit mode.\n- Edit mode (green box): Hit Esc to switch to command mode."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Shortcuts:\n- hit h when in command mode to see them (most are for command mode).\n- recommendation: try them all at least once!"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### Raw cells (shortcut `r` in command mode)"
    },
    {
      "metadata": {},
      "cell_type": "raw",
      "source": "You can ignore them for now (most users never need to use them)."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### Markdown cell (shortcut `m` in command mode)"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Markdown cells are used to enter all kind of contents but actual python code with a simple and nice formatting system."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "- This is a list\n 1. This is a numbered sublist\n 1. This is another numbered item. The actual numbers are not used.\n- This is **bold**, this is *italic*, this is ***both***.\n- You can write nice LaTeX formula in Markdown Cells, e.g. $\\int_2^3 x dx =\\frac{5}{2}$.\n- URL are automatically linked, like in https://www.google.com\n- Many other features that you will learn as you use it."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Recommendation: when writing a plain .py file, it is a good practice to describe what you do with *comments*. On notebook, it is preferred to use Markdown cells. This is especially nice when you implement some math stuff."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Headings cells are Markdown cells that start with one or more #. They translate into `<h1>`, `<h2>`... HTML structure."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "# First level heading"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "#### Fourth level heading"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "##### Fifth level heading"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "###### Last (sixth) level heading"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### Code cell (shortcut `y` in command mode)"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Code cells are where you actually write Python code. Each code cell is made of two parts:\n- input: the code to execute\n- output: the result (if any)"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "To execute a code cell:\n- click the *play* icon on the left of the cell\n- ctrl+enter executes the cell (goes command)\n- shift+enter executes the cell and move to the next one (goes command)\n- alt+enter executes the cell and create a new code cell (edit mode)"
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:02:31.265347Z",
          "start_time": "2021-10-12T20:02:31.254415Z"
        },
        "slideshow": {
          "slide_type": "subslide"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "1+1",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:02:27.612670Z",
          "start_time": "2021-10-12T20:02:27.596217Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "print(\"Hello World!\")",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Important feature: all cells in a given notebook share the same space (Kernel). This allows to build your code one step at a time"
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:04:12.112142Z",
          "start_time": "2021-10-12T20:04:12.100677Z"
        },
        "slideshow": {
          "slide_type": ""
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "x = 2",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:04:16.317605Z",
          "start_time": "2021-10-12T20:04:16.309616Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "y = 3",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:04:58.989804Z",
          "start_time": "2021-10-12T20:04:58.968143Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "x+y",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Notes :\n\n- The cell/kernel menu allow to execute multiple/all cells of a notebook, reset your notebook...\n- Each notebook has its own kernel"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### First steps in Python"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Python uses building blocks that are very similar to what you find in other languages: if, then, for, while... and print!"
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:23:13.091311Z",
          "start_time": "2021-10-12T21:23:13.067376Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "name = \"Python Workshop\"\nprint(\"Hello, \"+ name + \"!\")",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "In most languages, the keyword **end** is used to tell when a loop/condition/definition ends.\n\nOr some curly brackets.\n\nIn Python, there is no **end**. Instead, the ranges are defined according to indentation."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:30:06.599671Z",
          "start_time": "2021-10-12T21:30:06.578395Z"
        },
        "slideshow": {
          "slide_type": "subslide"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "# In most languages, indentation is optional and only for lisibility\nx = 0\nfor i in range(5):\n    x = x + i\n    print(\"x is \"+str(x))",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:30:16.318091Z",
          "start_time": "2021-10-12T21:30:16.291552Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "# In python, indentation (hence lisibility) is part of the language itself!\n# The price: indentation errors translate into code errors\nx = 0\nfor i in range(5):\n    x = x + i\nprint(\"x is \"+str(x))",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "When you make a mistake, you get some error message that (may?) help you."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:30:19.290312Z",
          "start_time": "2021-10-12T21:30:19.239216Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "print(\"This line is correct\")\nfor i in range(-3, 3):\n    inv_i = 1/i\n    print(f\"The inverse of {i} is {inv_i}\")\nprint(\"Will I get there?\")",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Defining functions is performed with **def**. "
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:21:49.313073Z",
          "start_time": "2021-10-12T21:21:49.268373Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "def my_division(x, y):\n    return x / y\nmy_division(10, 2)",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "You can assign default values, which allows highly flexible function calls."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:27:57.811460Z",
          "start_time": "2021-10-12T21:27:57.779934Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "def my_division(x=10, y=2):\n    return x / y\nprint(my_division(10, 2))\nprint(my_division(7))\nprint(my_division(y=4))\nprint(my_division(y=3, x=21))",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Packages can offer additional features. Many useful packages are shipped with Anaconda."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T21:41:31.806035Z",
          "start_time": "2021-10-12T21:41:31.315333Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "import numpy as np # numerical computation package. Very useful so we shorten its name as np\nfrom matplotlib import pyplot as plt # pyplot is a submodule (part of a module) to easily display figures\nx = np.linspace(-4, 4)\nplt.plot(x, np.sin(x), label='sin(x)')\nplt.legend()\nplt.show()",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "If you need another package, follows the package instructions (prefer `pip` if available).\nhttps://balouf.github.io/smup/installation.html"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### Magics"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Magics are used at the start of a cell or a line to change the way the code cell is interpreted. There are a lot of them by default (and a lot that can be added)."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Time it... to make stats about the execution time."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:53:56.502416Z",
          "start_time": "2021-10-12T20:53:54.196177Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "%timeit 6*7",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Time it is useful for fast benchmarking."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:21:45.502247Z",
          "start_time": "2021-10-12T20:21:39.028130Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "import numpy as np\n%timeit x = np.sum([np.random.rand()**2 for _ in range(10000)])",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:22:28.663652Z",
          "start_time": "2021-10-12T20:22:20.470206Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "%timeit x = np.sum(np.random.rand(10000)**2)",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Another one: write some javascript."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:18:37.444700Z",
          "start_time": "2021-10-12T20:18:37.425570Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "%%js\nif (true) alert(\"This is a javascript cell\")",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "### Extensions"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Jupyter extensions are not mandatory but they can facilitate your life. \n- Once installed, they are managed in the Nbextensions tab\n- You may have to close/reopen a notebook to see changes"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "Installation:\n- `conda install -c conda-forge jupyter_contrib_nbextensions`\n- ? (conda install -c conda-forge jupyter_nbextensions_configurator)\n- ? jupyter contrib nbextension install --user\n- `conda install -c conda-forge rise` (bonus)"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "#### Execution time"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Automatic benchmark (not as precise as timeit) AND date of last execution."
    },
    {
      "metadata": {
        "ExecuteTime": {
          "end_time": "2021-10-12T20:27:00.746534Z",
          "start_time": "2021-10-12T20:27:00.731790Z"
        },
        "trusted": false
      },
      "cell_type": "code",
      "source": "x = np.sum(np.random.rand(10000)**2)",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "#### Scratchpad (ctr+b)"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "When you want to test something in a hurry."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "#### Table of contents"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "#### Spellchecker"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "I don't neede that."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "subslide"
        }
      },
      "cell_type": "markdown",
      "source": "#### Others"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "- Gist-it\n- Rise\n- Split Cells\n- Initialization cells\n- ..."
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "## Jupyter Lab"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "A notebook file (`.ipynb`) has a simple structure that can be opened/used in many contexts:\n- Jupyter Lab (the replacement for jupyter notebook).\n- Colaboratory (online notebooks)\n- Github (display only)\n- PyCharm (although PC main focus is `.py` files)"
    },
    {
      "metadata": {
        "slideshow": {
          "slide_type": "slide"
        }
      },
      "cell_type": "markdown",
      "source": "## PyCharm"
    }
  ],
  "metadata": {
    "celltoolbar": "Slideshow",
    "hide_input": false,
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.7",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "toc": {
      "nav_menu": {},
      "number_sections": true,
      "sideBar": true,
      "skip_h1_title": true,
      "base_numbering": 1,
      "title_cell": "Table of Contents",
      "title_sidebar": "Contents",
      "toc_cell": false,
      "toc_position": {},
      "toc_section_display": true,
      "toc_window_display": false
    },
    "gist": {
      "id": "",
      "data": {
        "description": "first_notebook.ipynb",
        "public": true
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
 }
	{
	"cells": [
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "# Python 101"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "## Introduction"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "# Why Python?"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- A very popular language (on the podium, sometimes #1)\n- Easy to learn:\n - `1+1`\n - `print(\"Hello World\")`\n- Powerful to master:\n - Build complex packages\n - Publish them!\n- Huge ecosystem of packages and stackoverflow questions\n - Including all ML stuff (Pytorch)\n - Also some esoteric stuff"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "# Why NOT Python?"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- Slower than C (but you can learn to mitigate)\n- I need a very specific feature not available (rare but may occur)\n- I am fully happy with language X (no problem!)"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "# Python 2.X vs Python 3.Y"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- Big changes have been introduced in Python 3\n- Many legacy softwares have stayed 2.X for simplicity\n- As a beginner in Python:\n - Use Python 3.X. Period.\n - Avoid to be involved in any 2.X project if you can.\n - Be careful with 2.X solutions in stackoverflow\n"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "# How to program in Python?"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Very roughly, you have three main ways to use Python:\n- Directly inside a Python interpreter\n - Check something, make a one-liner\n- Write a python file (.py) that contains instructions to be executed\n - All-purposes, from small scripts to huge packages\n- Inside an interactive Python notebook (.ipynb)\n - Ideal to create new code and play with it"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "# How to program in Python?"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "To do python, you need:\n- A Python distribution\n- IDE for your notebooks / python files\n\nMany (many many) possibilities available. We focus on:\n- Jupyter Notebook\n- PyCharm"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "## Installation"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "# The Anaconda distribution"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "https://www.anaconda.com/products/individual\n- Pros:\n - All-in-one\n - Very easy to install\n - Can be installed on user account (highly recommended for beginners)\n- Cons:\n - 400 Mo\n - Easy to break ;)\n - Need to re-install for changing Python version"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "# Installing Anaconda"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- Tip #1: no space in the install directory\n- Tip #2: adding Anaconda to Path is not recommended, but it helps!\n- When you’re done, start an interpreter\n - `1+1`\n - `print(\"Hello World\")`\n- Optional: tour of the Anaconda Navigator\n - Possibly better for beginners?\n - Bad heaviness / usefulness trade-off"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "# Updating"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- Update is tricky:\n - Necessary to get latest features\n - Easy to inadvertently break things\n - Safest moment to update is on a fresh install!\n- Multiple ways to update/install packages in Python\n - Conda: the Anaconda internal package manager\n - Pip: universal package manager\n - ...\n- My rule of thumb:\n - Conda-related stuff (e.g. notebooks) -> conda\n - All the rest -> pip if possible"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "# Update intructions"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- `conda update conda`\n- `conda update --all`\n\nWhen ready, you can start your first notebook:\n\n`jupyter notebook`"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "## Your first notebook"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "A notebook is made of cells. A cell is a unit of content."
	},
	{
	"metadata": {},
	"cell_type": "raw",
	"source": "This is a raw Cell."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": ""
	}
	},
	"cell_type": "markdown",
	"source": "This is a markdown cell."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T19:34:42.895182Z",
	"start_time": "2021-10-12T19:34:42.868203Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "# This is a code cell\nprint(\"Hello world!\")",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### Navigation and shortcuts"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Jupyter has two main modes:\n- Command mode to navigate and manipulate entire cells;\n- Edit mode to change the content of a cell."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Recognize and switch mode:\n- Command mode (blue box): Select a cell (e.g. with arrows) and hit enter to switch to edit mode.\n- Edit mode (green box): Hit Esc to switch to command mode."
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Shortcuts:\n- hit h when in command mode to see them (most are for command mode).\n- recommendation: try them all at least once!"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### Raw cells (shortcut `r` in command mode)"
	},
	{
	"metadata": {},
	"cell_type": "raw",
	"source": "You can ignore them for now (most users never need to use them)."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### Markdown cell (shortcut `m` in command mode)"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Markdown cells are used to enter all kind of contents but actual python code with a simple and nice formatting system."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "- This is a list\n 1. This is a numbered sublist\n 1. This is another numbered item. The actual numbers are not used.\n- This is bold, this is italic, this is *both*.\n- You can write nice LaTeX formula in Markdown Cells, e.g. $\\int_2^3 x dx =\\frac{5}{2}$.\n- URL are automatically linked, like in https://www.google.com\n- Many other features that you will learn as you use it."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Recommendation: when writing a plain .py file, it is a good practice to describe what you do with comments. On notebook, it is preferred to use Markdown cells. This is especially nice when you implement some math stuff."
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Headings cells are Markdown cells that start with one or more #. They translate into `<h1>`, `<h2>`... HTML structure."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "# First level heading"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "#### Fourth level heading"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "##### Fifth level heading"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "###### Last (sixth) level heading"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### Code cell (shortcut `y` in command mode)"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Code cells are where you actually write Python code. Each code cell is made of two parts:\n- input: the code to execute\n- output: the result (if any)"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "To execute a code cell:\n- click the play icon on the left of the cell\n- ctrl+enter executes the cell (goes command)\n- shift+enter executes the cell and move to the next one (goes command)\n- alt+enter executes the cell and create a new code cell (edit mode)"
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:02:31.265347Z",
	"start_time": "2021-10-12T20:02:31.254415Z"
	},
	"slideshow": {
	"slide_type": "subslide"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "1+1",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:02:27.612670Z",
	"start_time": "2021-10-12T20:02:27.596217Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "print(\"Hello World!\")",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Important feature: all cells in a given notebook share the same space (Kernel). This allows to build your code one step at a time"
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:04:12.112142Z",
	"start_time": "2021-10-12T20:04:12.100677Z"
	},
	"slideshow": {
	"slide_type": ""
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "x = 2",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:04:16.317605Z",
	"start_time": "2021-10-12T20:04:16.309616Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "y = 3",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:04:58.989804Z",
	"start_time": "2021-10-12T20:04:58.968143Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "x+y",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Notes :\n\n- The cell/kernel menu allow to execute multiple/all cells of a notebook, reset your notebook...\n- Each notebook has its own kernel"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### First steps in Python"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Python uses building blocks that are very similar to what you find in other languages: if, then, for, while... and print!"
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:23:13.091311Z",
	"start_time": "2021-10-12T21:23:13.067376Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "name = \"Python Workshop\"\nprint(\"Hello, \"+ name + \"!\")",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "In most languages, the keyword end is used to tell when a loop/condition/definition ends.\n\nOr some curly brackets.\n\nIn Python, there is no end. Instead, the ranges are defined according to indentation."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:30:06.599671Z",
	"start_time": "2021-10-12T21:30:06.578395Z"
	},
	"slideshow": {
	"slide_type": "subslide"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "# In most languages, indentation is optional and only for lisibility\nx = 0\nfor i in range(5):\n x = x + i\n print(\"x is \"+str(x))",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:30:16.318091Z",
	"start_time": "2021-10-12T21:30:16.291552Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "# In python, indentation (hence lisibility) is part of the language itself!\n# The price: indentation errors translate into code errors\nx = 0\nfor i in range(5):\n x = x + i\nprint(\"x is \"+str(x))",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "When you make a mistake, you get some error message that (may?) help you."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:30:19.290312Z",
	"start_time": "2021-10-12T21:30:19.239216Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "print(\"This line is correct\")\nfor i in range(-3, 3):\n inv_i = 1/i\n print(f\"The inverse of {i} is {inv_i}\")\nprint(\"Will I get there?\")",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Defining functions is performed with def. "
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:21:49.313073Z",
	"start_time": "2021-10-12T21:21:49.268373Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "def my_division(x, y):\n return x / y\nmy_division(10, 2)",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "You can assign default values, which allows highly flexible function calls."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:27:57.811460Z",
	"start_time": "2021-10-12T21:27:57.779934Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "def my_division(x=10, y=2):\n return x / y\nprint(my_division(10, 2))\nprint(my_division(7))\nprint(my_division(y=4))\nprint(my_division(y=3, x=21))",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Packages can offer additional features. Many useful packages are shipped with Anaconda."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T21:41:31.806035Z",
	"start_time": "2021-10-12T21:41:31.315333Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "import numpy as np # numerical computation package. Very useful so we shorten its name as np\nfrom matplotlib import pyplot as plt # pyplot is a submodule (part of a module) to easily display figures\nx = np.linspace(-4, 4)\nplt.plot(x, np.sin(x), label='sin(x)')\nplt.legend()\nplt.show()",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "If you need another package, follows the package instructions (prefer `pip` if available).\nhttps://balouf.github.io/smup/installation.html"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### Magics"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Magics are used at the start of a cell or a line to change the way the code cell is interpreted. There are a lot of them by default (and a lot that can be added)."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Time it... to make stats about the execution time."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:53:56.502416Z",
	"start_time": "2021-10-12T20:53:54.196177Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "%timeit 6*7",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Time it is useful for fast benchmarking."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:21:45.502247Z",
	"start_time": "2021-10-12T20:21:39.028130Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "import numpy as np\n%timeit x = np.sum([np.random.rand()**2 for _ in range(10000)])",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:22:28.663652Z",
	"start_time": "2021-10-12T20:22:20.470206Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "%timeit x = np.sum(np.random.rand(10000)**2)",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Another one: write some javascript."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:18:37.444700Z",
	"start_time": "2021-10-12T20:18:37.425570Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "%%js\nif (true) alert(\"This is a javascript cell\")",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "### Extensions"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Jupyter extensions are not mandatory but they can facilitate your life. \n- Once installed, they are managed in the Nbextensions tab\n- You may have to close/reopen a notebook to see changes"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "Installation:\n- `conda install -c conda-forge jupyter_contrib_nbextensions`\n- ? (conda install -c conda-forge jupyter_nbextensions_configurator)\n- ? jupyter contrib nbextension install --user\n- `conda install -c conda-forge rise` (bonus)"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "#### Execution time"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "Automatic benchmark (not as precise as timeit) AND date of last execution."
	},
	{
	"metadata": {
	"ExecuteTime": {
	"end_time": "2021-10-12T20:27:00.746534Z",
	"start_time": "2021-10-12T20:27:00.731790Z"
	},
	"trusted": false
	},
	"cell_type": "code",
	"source": "x = np.sum(np.random.rand(10000)**2)",
	"execution_count": null,
	"outputs": []
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "#### Scratchpad (ctr+b)"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "When you want to test something in a hurry."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "#### Table of contents"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "#### Spellchecker"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "I don't neede that."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "subslide"
	}
	},
	"cell_type": "markdown",
	"source": "#### Others"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "- Gist-it\n- Rise\n- Split Cells\n- Initialization cells\n- ..."
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "## Jupyter Lab"
	},
	{
	"metadata": {},
	"cell_type": "markdown",
	"source": "A notebook file (`.ipynb`) has a simple structure that can be opened/used in many contexts:\n- Jupyter Lab (the replacement for jupyter notebook).\n- Colaboratory (online notebooks)\n- Github (display only)\n- PyCharm (although PC main focus is `.py` files)"
	},
	{
	"metadata": {
	"slideshow": {
	"slide_type": "slide"
	}
	},
	"cell_type": "markdown",
	"source": "## PyCharm"
	}
	],
	"metadata": {
	"celltoolbar": "Slideshow",
	"hide_input": false,
	"kernelspec": {
	"name": "python3",
	"display_name": "Python 3",
	"language": "python"
	},
	"language_info": {
	"name": "python",
	"version": "3.7.7",
	"mimetype": "text/x-python",
	"codemirror_mode": {
	"name": "ipython",
	"version": 3
	},
	"pygments_lexer": "ipython3",
	"nbconvert_exporter": "python",
	"file_extension": ".py"
	},
	"toc": {
	"nav_menu": {},
	"number_sections": true,
	"sideBar": true,
	"skip_h1_title": true,
	"base_numbering": 1,
	"title_cell": "Table of Contents",
	"title_sidebar": "Contents",
	"toc_cell": false,
	"toc_position": {},
	"toc_section_display": true,
	"toc_window_display": false
	},
	"gist": {
	"id": "",
	"data": {
	"description": "first_notebook.ipynb",
	"public": true
	}
	}
	},
	"nbformat": 4,
	"nbformat_minor": 5
	}