Tutorial (français)

Exemples

  1. Mixture of SIRD simulation and plotting

  2. SIRD simulation and plotting

  3. SIRDc simulation and plotting

  4. Stochastic Gradient Descent applied to linear regression

Mixture of SIRD simulation and plotting

(Source code, png, hires.png, pdf)

_images/model_covidsir_mixture-1.png

(original entry : covid_sird_mixture.py:docstring of aftercovid.models.covid_sird_mixture.CovidSIRDMixture, line 16)

SIRD simulation and plotting

(Source code, png, hires.png, pdf)

_images/model_covidsir-1.png

(original entry : covid_sird.py:docstring of aftercovid.models.covid_sird.CovidSIRD, line 13)

SIRDc simulation and plotting

(Source code, png, hires.png, pdf)

_images/model_covidsir_cst-1.png

(original entry : covid_sird_cst.py:docstring of aftercovid.models.covid_sird_cst.CovidSIRDc, line 80)

Stochastic Gradient Descent applied to linear regression

The following example how to optimize a simple linear regression.

<<<

import numpy
from aftercovid.optim import SGDOptimizer


def fct_loss(c, X, y):
    return numpy.linalg.norm(X @ c - y) ** 2


def fct_grad(c, x, y, i=0):
    return x * (x @ c - y) * 0.1


coef = numpy.array([0.5, 0.6, -0.7])
X = numpy.random.randn(10, 3)
y = X @ coef

sgd = SGDOptimizer(numpy.random.randn(3))
sgd.train(X, y, fct_loss, fct_grad, max_iter=15, verbose=True)
print('optimized coefficients:', sgd.coef)

>>>

    0/15: loss: 36.6 lr=0.1
    1/15: loss: 3.455 lr=0.1
    2/15: loss: 3.378 lr=0.1
    3/15: loss: 1.418 lr=0.1
    4/15: loss: 0.4894 lr=0.1
    5/15: loss: 0.09692 lr=0.1
    6/15: loss: 0.01206 lr=0.1
    7/15: loss: 0.000912 lr=0.1
    8/15: loss: 0.001238 lr=0.1
    9/15: loss: 0.001878 lr=0.1
    10/15: loss: 0.0008745 lr=0.1
    11/15: loss: 0.0006754 lr=0.1
    12/15: loss: 0.0002468 lr=0.1
    13/15: loss: 0.0001276 lr=0.1
    14/15: loss: 4.281e-05 lr=0.1
    15/15: loss: 3.307e-05 lr=0.1
    optimized coefficients: [ 0.5    0.6   -0.698]

(original entry : sgd.py:docstring of aftercovid.optim.sgd.SGDOptimizer, line 32)

Ligne de commande

Commande check

check

Vérifie que le module fonctionne comme prévu.

<<<

python -m aftercovid check --help

>>>

--SCRIPT---m aftercovid check --help
--OUT--

--ERR--
INFO: Showing help with the command '__main__.py check -- --help'.

NAME
    __main__.py check - Runs a couple of functions to check the module is working.

SYNOPSIS
    __main__.py check <flags>

DESCRIPTION
    Runs a couple of functions to check the module is working.

FLAGS
    --verbose=VERBOSE
        Default: 1
        0 to hide the standout output
--PATH--
None