标签: lmfit

我想使用lmfit模块将函数拟合到可变数量的数据集,包括一些共享和一些单独的参数.

以下是生成高斯数据并分别拟合每个数据集的示例:

import numpy as np
import matplotlib.pyplot as plt
from lmfit import minimize, Parameters, report_fit

def func_gauss(params, x, data=[]):
    A = params['A'].value
    mu = params['mu'].value
    sigma = params['sigma'].value
    model = A*np.exp(-(x-mu)**2/(2.*sigma**2))

    if data == []:
        return model
    return data-model

x  = np.linspace( -1, 2, 100 )
data = []
for i in np.arange(5):
    params = Parameters()
    params.add( 'A'    , value=np.random.rand() )
    params.add( 'mu'   , value=np.random.rand()+0.1 )
    params.add( 'sigma', value=0.2+np.random.rand()*0.1 )
    data.append(func_gauss(params,x))

plt.figure()
for y in data:
    fit_params …

我正在尝试使用lmfit (链接到docs)构建模型,我似乎无法找出为什么我ValueError: The input contains nan values在尝试适应模型时不断获得.

from lmfit import minimize, Minimizer, Parameters, Parameter, report_fit, Model
import numpy as np

def cde(t, Qi, at, vw, R, rhob_cb, al, d, r):
    # t (time), is the independent variable
    return Qi / (8 * np.pi * ((at * vw)/R) * t * rhob_cb * (np.sqrt(np.pi * ((al * vw)/R * t))))  * \
        np.exp(- (R * (d - (t * vw)/ R)**2) / (4 * (al * vw) * …

我想使用lmfit 将椭圆测量数据拟合到复杂模型中.两个测量参数,psi和delta是复杂函数中的变量rho.

我可以尝试用共享参数或picewise方法将问题分离到实部和虚部,但是有没有办法直接用复杂函数来做？仅适合函数的实部工作,但是当我定义复杂的残差函数时,我得到:

TypeError:没有为复数定义排序关系.

下面是我的实际功能拟合代码和我尝试解决复杂的拟合问题:

    from __future__ import division
    from __future__ import print_function
    import numpy as np
    from pylab import *
    from lmfit import minimize, Parameters, Parameter, report_errors


    #=================================================================
    #             MODEL

    def r01_p(eps2, th):
        c=cos(th)
        s=(sin(th))**2

        stev= sqrt(eps2) * c - sqrt(1-(s / eps2))
        imen= sqrt(eps2) * c + sqrt(1-(s / eps2))
        return stev/imen

    def r01_s(eps2, th):
        c=cos(th)
        s=(sin(th))**2

        stev= c - sqrt(eps2) * sqrt(1-(s/eps2))
        imen= c + …

我有实验数据:

xdata = [85,86,87,88,89,90,91,91.75,93,96,100,101,102,103,104,105,106,107.25,108.25,109,109.75,111,112,112.75,114,115.25,116,116.75,118,119.25,120,121,122,122.5,123.5,125.25,126,126.75,127.75,129.25,130.25,131,131.75,133,134.25,135,136,137,138,139,140,141,142,143,144,144.75,146,146.75,148,149.25,150,150.5,152,153.25,154,155,156.75,158,159,159.75,161,162,162.5,164,165,166]

ydata = [0.2,0.21,0.18,0.21,0.19,0.2,0.21,0.2,0.18,0.204,0.208,0.2,0.21,0.25,0.2,0.19,0.216,0.22,0.224,0.26,0.229,0.237,0.22,0.246,0.25,0.264,0.29,0.274,0.29,0.3,0.27,0.32,0.38,0.348,0.372,0.398,0.35,0.42,0.444,0.48,0.496,0.55,0.51,0.54,0.57,0.51,0.605,0.57,0.65,0.642,0.6,0.66,0.7,0.688,0.69,0.705,0.67,0.717,0.69,0.728,0.75,0.736,0.73,0.744,0.72,0.76,0.752,0.74,0.76,0.7546,0.77,0.74,0.758,0.74,0.78,0.76]

和公式f(x) = m1 + m2 / (1 + e ^ (-m3*(x - m4))).我需要找到m1, m2, m3, m4最小二乘法,其中0.05 <m1 <0.3 0.3 <m2 <0.8 0.05 <m3 <0.5 100 <m4 <200.

我用curve_fit,我的功能是:

def f(xdata, m1, m2, m3, m4):
    if m1 > 0.05 and m1 < 0.3 and \
       m2 > 0.3 and m2 < 0.8 and \
       m3 > 0.05 and m3 < 0.5 and \
       m4 > 100 and m4 …

我正在尝试将此函数拟合到一些数据中：

但是当我使用我的代码时

import numpy as np
from scipy.optimize import curve_fit
import matplotlib.pyplot as plt

def f(x, start, end):
    res = np.empty_like(x)
    res[x < start] =-1
    res[x > end] = 1
    linear = np.all([[start <= x], [x <= end]], axis=0)[0]
    res[linear] = np.linspace(-1., 1., num=np.sum(linear))
    return res

if __name__ == '__main__':

    xdata = np.linspace(0., 1000., 1000)
    ydata = -np.ones(1000)
    ydata[500:1000] = 1.
    ydata = ydata + np.random.normal(0., 0.25, len(ydata))

    popt, pcov = curve_fit(f, xdata, ydata, p0=[495., 505.])
    print(popt, pcov)
    plt.figure()
    plt.plot(xdata, …

I am trying to deconvolve complex gas chromatogram signals into individual gaussian signals. Here is an example, where the dotted line represents the signal I am trying to deconvolve.

I was able to write the code to do this using scipy.optimize.curve_fit; however, once applied to real data the results were unreliable. I believe being able to set bounds to my parameters will improve my results, so I am attempting to use lmfit, which allows this. I am having a problem …

我正在 Python 2.7 中运行拟合，并lmfit使用一些测试数据和以下代码。我需要重量为 的加权拟合1/y（使用 Leven-Marq. 例程）。我已经定义了权重并在这里使用它们：

from __future__ import division
from numpy import array, var
from lmfit import Model
from lmfit.models import GaussianModel, LinearModel

import matplotlib.pyplot as plt
import seaborn as sns

xd = array([1267, 1268, 1269, 1270, 1271, 1272, 1273, 1274, 1275, 1276,
    1277, 1278, 1279, 1280, 1281, 1282, 1283, 1284, 1285, 1286, 1287, 1288,
     1289, 1290, 1291, 1292, 1293, 1294, 1295, 1296, 1297, 1298, 1299, 1300,
     1301, 1302, 1303, 1304, 1305, 1306, 1307, …

我编写了一个程序来拟合一些拉曼光谱峰值。\n我需要返回拟合的参数（位置、幅度、HWHM）。

我使用 modul lmfit 创建一个带有约束的洛伦兹峰。

根据我的图形，拟合峰值和原始数据之间有很好的一致性。\n但是当拟合后提取参数时，我遇到了一个问题，程序仅返回初始值。

我绑定了“report_fit 模块”并更改了初始参数，但没有成功。参数值不会变化。

令我困扰的是，这个程序在我同事的电脑上运行，但在我的电脑上运行不了。\n所以问题可能出在我的 python 版本上。

我正在使用spyder 2.3.9，并在Windows 10下使用anaconda安装了python 3.4。\nlmfit模块0.9.3似乎可以部分工作，因为我可以获得一个很好的拟合协议（从图plt.plot中）。\n但是拟合后无法返回参数值。

这是我的代码：

#!/usr/bin/python3\n# -*- coding:utf-8 -*-\n\nimport os\nimport numpy as np\nimport matplotlib.pyplot as plt\nfrom math import factorial\nfrom scipy.interpolate import interp1d\nfrom lmfit import minimize, Parameters  #,report_fit\n\n##############################################################################\n# Fit of Raman peaks\n\ndef fmin150(pars,x,y):  \n    amp= pars[\'Amp_150\'].value\n    cent=pars[\'Cent_150\'].value\n    hwhm=pars[\'Wid_150\'].value\n    a=pars[\'a_150\'].value\n    b=pars[\'b_150\'].value\n    peak = (amp*hwhm)/(((x-cent)**2)+(hwhm**2)) + ((a*x)+b)\n    return peak - y    \n\ndef fmin220(pars,x,y):  \n    amp= pars[\'Amp_220\'].value\n    cent=pars[\'Cent_220\'].value\n    hwhm=pars[\'Wid_220\'].value\n    a=pars[\'a_220\'].value\n    b=pars[\'b_220\'].value\n    peak = …

如您所知，python 中的 lmfit 模块可以方便地扩展 scipy.optimize 函数的功能。

然而，我没有发现在我看来必要的东西：选择步长的可能性（用于偏导、参数空间中 chi2 的计算等......）。我曾经在 IDL 下安装时玩过这些步骤，我很惊讶我在 python 下没有找到这个。

很明显，默认的步长非常小，在拟合粗略模型时可能会导致恒定的 chi2……因此很尴尬。

所以我的问题是：在 python 下拟合时如何选择步骤？

这是关于从lmfit fit_report()（1）对象提取拟合统计信息的问题

在此 lmfit示例中，返回以下部分输出：

[[Model]]
    Model(gaussian)
[[Fit Statistics]]
    # function evals   = 31
    # data points      = 101
    # variables        = 3
    chi-square         = 3.409
    reduced chi-square = 0.035
    Akaike info crit   = -336.264
    Bayesian info crit = -328.418
.
.
.
.
.
.

我试图提取该Fit Statistics部分中的所有数量作为单独的变量。

例如。提取模型参数，我们可以使用（每1，2）：

for key in fit.params:
    print(key, "=", fit.params[key].value, "+/-", fit.params[key].stderr)

但是，这仅给出了模型参数。它没有提供拟合统计参数，这也很有用。我似乎在文档中找不到此内容。

有没有类似的方式来提取拟合统计参数（chi-square，reduced chi-square， …

我正在尝试使该数据适合高斯

x = [4170.177259096838, 4170.377258006199, 4170.577256915561, 4170.777255824922, 4170.977254734283, 4171.177253643645, 4171.377252553006, 4171.577251462368, 4171.777250371729, 4171.977249281091, 4172.177248190453, 4172.377247099814, 4172.577246009175, 4172.777244918537, 4172.977243827898, 4173.17724273726, 4173.377241646621, 4173.577240555983, 4173.777239465344, 4173.977238374706, 4174.177237284067, 4174.377236193429, 4174.57723510279, 4174.777234012152, 4174.977232921513, 4175.177231830875, 4175.377230740236, 4175.577229649598, 4175.777228558959, 4175.977227468321, 4176.177226377682, 4176.377225287044, 4176.577224196405, 4176.777223105767, 4176.977222015128, 4177.17722092449, 4177.377219833851, 4177.577218743213, 4177.777217652574, 4177.977216561936, 4178.177215471297, 4178.377214380659, 4178.57721329002, 4178.777212199382, 4178.977211108743, 4179.177210018105, 4179.377208927466, 4179.577207836828, 4179.777206746189, 4179.977205655551, 4180.177204564912, 4180.377203474274, 4180.577202383635, 4180.777201292997, 4180.977200202357, 4181.17719911172, 4181.377198021081, 4181.577196930443, 4181.777195839804, 4181.977194749166, 4182.177193658527, 4182.377192567888, 4182.5771914772495, 4182.777190386612, 4182.9771892959725, 4183.177188205335, 4183.377187114696, 4183.577186024058, 4183.777184933419, 4183.9771838427805, 4184.177182752143, 4184.3771816615035, 4184.5771805708655, 4184.777179480228, 4184.977178389589, 4185.1771772989505, 4185.3771762083115, …

我正在使用 lmfit 将我的数据拟合为高斯分布。我试图完成三件事：1）了解如何在 lmfit 中计算误差 2）如何在 lmfit 中包含我自己计算的误差 3）如何在拟合中绘制误差

def gaussian(x, amp, cen, fwhm):
    return + amp * np.exp(-(x - cen) ** 2 / (2 * (fwhm / 2.35482) ** 2))    

def gaussian_fit(x,y,guess=[1,0,0,5],varies=[True,True,True,True]):

c = 299792458 #m/s
gmod = Model(gaussian)
gmod.nan_policy = 'omit'
#x,y - your dataset to fit, with x and y values
print (np.max(y))
gmod.set_param_hint('amp', value=guess[0],vary=varies[0])  
gmod.set_param_hint('cen', value=guess[1],vary=varies[1])
gmod.set_param_hint('fwhm', value=guess[2],vary=varies[2])  
gmod.make_params()

result = gmod.fit(y,x=x,amp=guess[0], cen=guess[1], fwhm=guess[2])

amp = result.best_values['amp']
cen = result.best_values['cen']
fwhm = result.best_values['fwhm']
#level = result.best_values['level'] …