~~stoggle_buttons~~
===== Instalación de todos los backends =====
[[https://matplotlib.org/faq/usage_faq.html#what-is-a-backend]]
<code bash>
pip install tk PyQt5 # Estos funcionan con pip
</code>


===== Trucos =====

<code python>
import matplotlib.pyplot as plt
plt.ion() # Interactive mode ON
plt.rc('text', usetex=True) # Latex rendering
plt.minorticks_on() # Needs to be set before calling plot.grid()
plt.grid(True, which='both')
plt.tick_params(axis='both', which='major', labelsize=20) # Plot tick size
plt.tight_layout() # If xlabel gets cut off
plt.xlabel('$\\lambda \\mathrm{ (nm)}$')
plt.xlabel('$\\lambda \\text{ (nm)}$') # PETA fuertemente
plt.xlabel('$\\lambda $\\text{ (nm)}') # Probar
plt.title(r'Entrop\'ia') # renders to Entropía
plt.title(r"$Regresi\'on$") # Renders to Regresión
# Linear regression
slope, intercept, r, prob2, see = scipy.stats.linregress(x, y)
mx = x.mean()
sx2 = ((x-mx)**2).sum()                                                                                                                        
sd_intercept = see * np.sqrt(1./len(x) + mx*mx/sx2)
sd_slope = see * np.sqrt(1./sx2)
</code>

=== Cambiar el preámbulo de matplotlib ===
<code python>
import matplotlib.pyplot as plt

plt.rc('text', usetex=True)
plt.rc('text.latex', preamble=r'\usepackage{amsmath}
          \usepackage{foo-name} `...')
matplotlib.verbose.level = 'debug-annoying'
</code>

=== Estilos de plots ===
[[https://medium.com/analytics-vidhya/drastically-beautifying-visualizations-with-one-line-styling-plots-35a5712c4f54|estilos de plots]]
<code python>
plt.style.use('fivethirtyeight')
plt.style.use('ggplot')
</code>


=== Setup mínimo de tex para matplotlib ===
<code bash>

apt-get install texlive-base texlive texlive-fonts*
apt-get install dvipng
</code>

===== animaciones =====

<code python>
from matplotlib.animation import FuncAnimation
from matplotlib.animation import writers
Writer = writers['ffmpeg']
Writer = writers['ffmpeg_file'] # Para exportar muchos frames y no comerse la ram
duration = 120 # En segundos
fps = t_gridnum/duration if t_gridnum/duration > 25.0 else 25.0 # Mínimo 25fps
writer = Writer(fps=fps)
ani = FuncAnimation(fig, animate, init_func=init, frames=np.arange(len(times)), repeat=False, cache_frame_data=False)
# Importante el cache_frame_data para mayor velocidad con vídeos de muchos frames
</code>

[[https://brushingupscience.com/2016/06/21/matplotlib-animations-the-easy-way/]]