def poly_lsq(x, y, n, verbose=False, itmax=200):
''' Performs a polynomial least squares fit to the data,
with errors! Uses scipy odrpack, but for least squares.
IN:
x,y (arrays) - data to fit
n (int) - polinomial order
verbose - can be 0,1,2 for different levels of output
(False or True are the same as 0 or 1)
itmax (int) - optional maximum number of iterations
OUT:
coeff - polynomial coefficients, lowest order first
err - standard error (1-sigma) on the coefficients
--Tiago, 20071114
'''
func = models.polynomial(n)
mydata = odr.Data(x, y)
myodr = odr.ODR(mydata, func, maxit=itmax)
# Set type of fit to least-squares:
myodr.set_job(fit_type=2)
if verbose == 2:
myodr.set_iprint(final=2)
fit = myodr.run()
# Display results:
if verbose:
fit.pprint()
if fit.stopreason[0] == 'Iteration limit reached':
print('(WWW) poly_lsq: Iteration limit reached, result not reliable!')
# Results and errors
coeff = fit.beta
err = fit.sd_beta
return coeff, err
def poly_lsq(x,y,n,verbose=False,itmax=200):
''' Performs a polynomial least squares fit to the data,
with errors! Uses scipy odrpack, but for least squares.
IN:
x,y (arrays) - data to fit
n (int) - polinomial order
verbose - can be 0,1,2 for different levels of output
(False or True are the same as 0 or 1)
itmax (int) - optional maximum number of iterations
OUT:
coeff - polynomial coefficients, lowest order first
err - standard error (1-sigma) on the coefficients
--Tiago, 20071114
'''
# http://www.scipy.org/doc/api_docs/SciPy.odr.odrpack.html
# see models.py and use ready made models!!!!
func = models.polynomial(n)
mydata = odr.Data(x, y)
myodr = odr.ODR(mydata, func,maxit=itmax)
# Set type of fit to least-squares:
myodr.set_job(fit_type=2)
if verbose == 2: myodr.set_iprint(final=2)
fit = myodr.run()
# Display results:
if verbose: fit.pprint()
if fit.stopreason[0] == 'Iteration limit reached':
print '(WWW) poly_lsq: Iteration limit reached, result not reliable!'
# Results and errors
coeff = fit.beta[::-1]
err = fit.sd_beta[::-1]
return coeff, err
def poly_lsq(x, y, n, verbose=False, itmax=200):
"""
Performs a polynomial least squares fit to the data,
with errors! Uses scipy odrpack, but for least squares.
Parameters
----------
x, y : 1-D arrays
Data to fit.
n : int
Polynomial order
verbose : bool or int, optional
Can be 0,1,2 for different levels of output (False or True
are the same as 0 or 1)
itmax : int, optional
Maximum number of iterations.
Returns
-------
coeff : 1-D array
Polynomial coefficients, lowest order first.
err : 1-D array
Standard error (1-sigma) on the coefficients.
"""
func = models.polynomial(n)
mydata = odr.Data(x, y)
myodr = odr.ODR(mydata, func, maxit=itmax)
# Set type of fit to least-squares:
myodr.set_job(fit_type=2)
if verbose == 2:
myodr.set_iprint(final=2)
fit = myodr.run()
# Display results:
if verbose:
fit.pprint()
if fit.stopreason[0] == 'Iteration limit reached':
print('(WWW) poly_lsq: Iteration limit reached, result not reliable!')
# Results and errors
coeff = fit.beta
err = fit.sd_beta
return coeff, err
lwc = float(lines[1].strip())
lwhmag = float(lines[2].strip())
'''
################################################################################
'''
# def funtion
def linear_reg_fix_slope(c, x):
from numpy import log10, zeros_like
grd = grad # set gradient
ans = c[0] + x * grd
return ans
linreg = models.polynomial(1)
im = [7.5, 8.0]
isi = [0.2, 0.1]
'''
################################################################################
'''
# set data
ftypes = ['Intraslab', 'Outer Rise', 'Strike-Slip']
grads = [lengrd1, widgrd1, areagrd1, mxsgrd, avsgrd, lwgrd]
fmag = [intra_fmag, outer_fmag, trans_fmag]
flen = [intra_flen, outer_flen, trans_flen]
fwid = [intra_fwid, outer_fwid, trans_fwid]
farea = [intra_farea, outer_farea, trans_farea]
fmxs = [intra_fmxs, outer_fmxs, trans_fmxs]
favs = [intra_favs, outer_favs, trans_favs]
def fit_offset(pha, offset, sigma=1, prange=None, orange=None, method='ols', flip=False):
"""
Fit a pha and an offset (DC level)
Parameters (and their default values):
pha: pha data (array-like)
offset: offset data (array-like)
sigma: sigmas allowed for median filter (Default: 1)
prange: a tuple of range for pha to fit if not None (Default: None)
orange: a tuple of range for offset to fit if not None (Default: None)
method: fitting method from ols (ordinal least squares)
or odr (orthogonal distance regression) (Default: ols)
flip: flip pha and offset when fitting if True (Default: False)
Return ((a, b), coef):
a, b: fitting results to pha = a*(1+b*offset)
coef: correlation coefficient
"""
# Sanity check
if len(pha) != len(offset):
raise ValueError("data length of pha and offset does not match")
pha = np.asarray(pha)
offset = np.asarray(offset)
# Reduction
if prange is not None:
pmask = (pha >= prange[0]) & (pha <= prange[1])
else:
pmask = median_filter(pha, sigma=sigma)
if orange is not None:
omask = (offset >= orange[0]) & (offset <= orange[1])
else:
omask = median_filter(offset, sigma=sigma)
mask = pmask & omask
# Correlation coefficient
coef = np.corrcoef(pha[mask], offset[mask])[0,1]
# Fitting to a*x+b
if method.lower() == 'ols':
if flip:
_a, _b = np.polyfit(pha[mask], offset[mask], 1)
else:
_a, _b = np.polyfit(offset[mask], pha[mask], 1)
elif method.lower() == 'odr':
f = models.polynomial(1)
if flip:
data = odrpack.Data(pha[mask], offset[mask])
else:
data = odrpack.Data(offset[mask], pha[mask])
odr = odrpack.ODR(data, f, maxit=100)
fit = odr.run()
_a, _b = fit.beta[::-1]
else:
raise ValueError('Unknown method: %s' % method)
if flip:
a = -_b/_a
b = -1/_b
else:
a = _b
b = _a/_b
return (a, b), coef
coeftxt = 'Function,a,b,SEa,SEb,sig,Condition\n' # make uncertainty arrays for weighting for ODR reg_logMo = (mw2m0(array(reg_fmag))) #sx = ones(len(reg_fmag))*0.1 # uniform uncertainty of 0.2 mu #sx = ones(len(reg_fmag))*0.2 im = [7.5, 8.0] isi = [0.2, 0.1] sx = interp(reg_fmag, im, isi) #idx = reg_fmag >= 8.0 #sx[idx] = 0.1 sy = sx # do ODR on raw data func = models.polynomial(1) #data = Data(array(inter_fmag),log10(array(inter_flen))) data = odrpack.RealData(array(reg_fmag), log10(array(reg_flen)), sx=sx, sy=sy) # set data mrng = arange(7., 9.61, 0.01) ''' ######################################################################################## test linear vs ODR linear vs ODR bi-linear ''' # regress M vs L - ODR beta0 = [-2.7, 0.62] odr = ODR(data, func, beta0) odr.set_job(fit_type=0) #if set fit_type=2, returns the same as leastsq out = odr.run()