# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "slide"}
from plan import plan_extras
plan_extras("packaging")
# %% [markdown] slideshow={"slide_type": "slide"}
# ### *packaging* (software distribution)
# %% [markdown]
# ### portée
# on regroupe ici les outils pour
#
# * **packager** un composant logiciel
# * **exposer** différentes versions à une infrastructure
# * **installer** à partir de l'infrastructure
# * **installer** un module localement à partir des sources
# * **gérer** les dépendances entre composants
# * **maintenir** plusieurs combinaisons de composants
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "-"}
from plan import plan_extras
plan_extras("pandas")
# %% [markdown] slideshow={"slide_type": "slide"}
# # pandas
# %% [markdown]
# from `Python for data analysis`, Wes McKinney
# %% [markdown]
# # objectives
#
# * data structures with labeled axes
# * automatic or explicit data alignment
# * integrated time series functionality
# * same data structure for time series or non-time series
# * arithemtic operations and reductions (on a whole column)
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": ""}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "slide"}
from plan import plan_extras
plan_extras("notebooks")
# %% [markdown] slideshow={"slide_type": "slide"}
# # introduction aux notebooks
# %% [markdown]
# ## c'est quoi un notebook ?
# %% [markdown]
# un document hybride qui mélange:
#
# * du texte formatté,
# * des équations et illustrations,
# * avec des fragments de code exécutable,
# * possiblement interactifs.
# jupyter: # jupytext: # cell_metadata_filter: all,-hidden,-heading_collapsed,-run_control,-trusted # notebook_metadata_filter: all,-language_info,-toc,-jupytext.text_representation.jupytext_version,-jupytext.text_representation.format_version # text_representation: # extension: .py # format_name: percent # kernelspec: # display_name: Python 3 # language: python # name: python3 # notebookname: plan extras # rise: # autolaunch: true # slideNumber: c/t # start_slideshow_at: selected # theme: sky # transition: cube # version: '1.0' # --- # %% [markdown] # <div class="licence"> # <span>Licence CC BY-NC-ND</span> # <span>Thierry Parmentelat</span> # </div> # %% from plan import plan_extras plan_extras()
# autolaunch: true
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "-"}
from plan import plan_extras; plan_extras("numpy", "dimension 1")
# %% [markdown] slideshow={"slide_type": "slide"}
# # intro à numpy
# %% [markdown]
# ## pourquoi `numpy`
# %% [markdown]
# la librairie numpy est **très** utilisée pour les traitements gourmands en calculs, car
#
# * elle est **beaucoup plus efficace** que Python "*de base* " pour cet usage,
# * même si elle est un tout petit peu moins générale
# %% [markdown] slideshow={"slide_type": "slide"}
# la limitation concerne le fait que **les tableaux sont homogènes**
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "slide"}
from plan import plan_extras
plan_extras("numpy", "linéaire")
# %% [markdown] slideshow={"slide_type": "slide"}
# # algèbre linéaire
# %%
import numpy as np
import matplotlib.pyplot as plt
# %matplotlib inline
plt.ion()
# %% [markdown]
# * Un aspect important de l'utilisation de numpy
# * consiste à manipuler des matrices et vecteurs
# * [voir doc complète](https://docs.scipy.org/doc/numpy/reference/routines.linalg.html)
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "slide"}
from plan import plan_extras
plan_extras("doc", "readthedocs")
# %% [markdown] slideshow={"slide_type": "slide"}
# # `readthedocs.io`
# %% [markdown]
# * système communautaire
# * pour héberger la documentation
# * mode 'on installe et on oublie'
# * connecté aux *webhooks* de gitlab/github
# * i.e. regénère la documentation à chaque commit
# * possiblement pour plusieurs branches / tags
# %% [markdown] slideshow={"slide_type": "slide"}
# ## exemple
# 
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %%
from plan import plan_extras
plan_extras("doc", "outils")
# %% [markdown] slideshow={"slide_type": "slide"}
# # génération de documentation
# %% [markdown] slideshow={"slide_type": "-"}
# * PEPs
# * outils de génération de documents
# * [doxygen](http://www.stack.nl/~dimitri/doxygen/) (non détaillé)
# * [sphinx](http://www.sphinx-doc.org/en/stable/)
# * [mkdocs](http://www.mkdocs.org/): plus orienté markdown (non détaillé)
# * langages de markup : reST, markdown et doctest
# * styles de docstrings : reST, google et numpy
# * readthedocs.io (infra / présentation séparée)
# %% [markdown] slideshow={"slide_type": "slide"}
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "slide"}
from plan import plan_extras
plan_extras("numpy", "supérieures")
# %% [markdown] slideshow={"slide_type": "slide"}
# # numpy - dimensions > 1
# %% slideshow={"slide_type": "-"}
import numpy as np
import matplotlib.pyplot as plt
# %matplotlib inline
plt.ion()
# %% [markdown]
# # conventions sur les indices des dimension
# %% [markdown]
# ## dimension 2: **lignes** x **colonnes**
# slideNumber: c/t
# start_slideshow_at: selected
# theme: sky
# transition: cube
# version: '1.0'
# ---
# %% [markdown] slideshow={"slide_type": "slide"}
# <div class="licence">
# <span>Licence CC BY-NC-ND</span>
# <span>Thierry Parmentelat</span>
# </div>
# %% slideshow={"slide_type": "slide"}
from plan import plan_extras
plan_extras("numpy", "struct")
# %% [markdown] slideshow={"slide_type": "slide"}
# # structured arrays
# %%
import numpy as np
import matplotlib.pyplot as plt
# %matplotlib inline
plt.ion()
# %% [markdown]
# * jusqu'ici on a vu des tableaux *homogènes*
# * tous les éléments ont le même type
# * on peut aussi se définir des types structurés
# * comme un 'struct' en c - ou encore un 'record'