| 1 | from sas.sascalc.data_util.calcthread import CalcThread |
|---|
| 2 | from sas.sascalc.dataloader.data_info import Data1D |
|---|
| 3 | from scipy.fftpack import dct |
|---|
| 4 | from scipy.integrate import trapz |
|---|
| 5 | import numpy as np |
|---|
| 6 | from time import sleep |
|---|
| 7 | |
|---|
| 8 | class FourierThread(CalcThread): |
|---|
| 9 | def __init__(self, raw_data, extrapolated_data, bg, extrap_fn=None, |
|---|
| 10 | updatefn=None, completefn=None): |
|---|
| 11 | CalcThread.__init__(self, updatefn=updatefn, completefn=completefn) |
|---|
| 12 | self.data = raw_data |
|---|
| 13 | self.background = bg |
|---|
| 14 | self.extrapolation = extrapolated_data |
|---|
| 15 | self.extrap_fn = extrap_fn |
|---|
| 16 | |
|---|
| 17 | def check_if_cancelled(self): |
|---|
| 18 | if self.isquit(): |
|---|
| 19 | self.update("Fourier transform cancelled.") |
|---|
| 20 | self.complete(transforms=None) |
|---|
| 21 | return True |
|---|
| 22 | return False |
|---|
| 23 | |
|---|
| 24 | def compute(self): |
|---|
| 25 | qs = self.extrapolation.x |
|---|
| 26 | iqs = self.extrapolation.y |
|---|
| 27 | q = self.data.x |
|---|
| 28 | background = self.background |
|---|
| 29 | |
|---|
| 30 | xs = np.pi*np.arange(len(qs),dtype=np.float32)/(q[1]-q[0])/len(qs) |
|---|
| 31 | |
|---|
| 32 | self.ready(delay=0.0) |
|---|
| 33 | self.update(msg="Fourier transform in progress.") |
|---|
| 34 | self.ready(delay=0.0) |
|---|
| 35 | |
|---|
| 36 | if self.check_if_cancelled(): return |
|---|
| 37 | try: |
|---|
| 38 | # ----- 1D Correlation Function ----- |
|---|
| 39 | gamma1 = dct((iqs-background)*qs**2) |
|---|
| 40 | gamma1 = gamma1 / gamma1.max() |
|---|
| 41 | |
|---|
| 42 | if self.check_if_cancelled(): return |
|---|
| 43 | |
|---|
| 44 | # ----- 3D Correlation Function ----- |
|---|
| 45 | # gamma3(R) = 1/R int_{0}^{R} gamma1(x) dx |
|---|
| 46 | # trapz uses the trapezium rule to calculate the integral |
|---|
| 47 | mask = xs <= 200.0 # Only calculate gamma3 up to x=200 (as this is all that's plotted) |
|---|
| 48 | gamma3 = [trapz(gamma1[:n], xs[:n])/xs[n-1] for n in range(2, len(xs[mask]) + 1)] |
|---|
| 49 | gamma3.insert(0, 1.0) # Gamma_3(0) is defined as 1 |
|---|
| 50 | gamma3 = np.array(gamma3) |
|---|
| 51 | |
|---|
| 52 | if self.check_if_cancelled(): return |
|---|
| 53 | |
|---|
| 54 | # ----- Interface Distribution function ----- |
|---|
| 55 | dmax = 200.0 # Max real space value to calculate IDF up to |
|---|
| 56 | dstep = 0.5 |
|---|
| 57 | qmax = 1.0 # Max q value to integrate up to when calculating IDF |
|---|
| 58 | |
|---|
| 59 | # Units of x axis depend on qmax (for some reason?). This scales |
|---|
| 60 | # the xgamma array appropriately, since qmax was set to 0.6 in |
|---|
| 61 | # the original fortran code. |
|---|
| 62 | x_scale = qmax / 0.6 |
|---|
| 63 | |
|---|
| 64 | xgamma = np.arange(0, dmax/x_scale, step=dstep/x_scale) |
|---|
| 65 | idf = np.zeros(len(xgamma)) |
|---|
| 66 | |
|---|
| 67 | # nth moment = integral(q^n * I(q), q=0, q=inf) |
|---|
| 68 | moment = np.zeros(5) |
|---|
| 69 | for n in range(5): |
|---|
| 70 | integrand = qs**n * (iqs-background) |
|---|
| 71 | moment[n] = trapz(integrand[qs < qmax], qs[qs < qmax]) |
|---|
| 72 | if self.check_if_cancelled(): return |
|---|
| 73 | |
|---|
| 74 | # idf(x) = -integral(q^4 * I(q)*cos(qx), q=0, q=inf) / 2nd moment |
|---|
| 75 | # => idf(0) = -integral(q^4 * I(q), 0, inf) / (2nd moment) |
|---|
| 76 | # = -(4th moment)/(2nd moment) |
|---|
| 77 | idf[0] = -moment[4] / moment[2] |
|---|
| 78 | for i in range(1, len(xgamma)): |
|---|
| 79 | d = xgamma[i] |
|---|
| 80 | |
|---|
| 81 | integrand = -qs**4 * (iqs-background) * np.cos(d*qs) |
|---|
| 82 | idf[i] = trapz(integrand[qs < qmax], qs[qs < qmax]) |
|---|
| 83 | idf[i] /= moment[2] |
|---|
| 84 | if self.check_if_cancelled(): return |
|---|
| 85 | |
|---|
| 86 | xgamma *= x_scale |
|---|
| 87 | |
|---|
| 88 | except Exception as e: |
|---|
| 89 | import logging |
|---|
| 90 | logger = logging.getLogger(__name__) |
|---|
| 91 | logger.error(e) |
|---|
| 92 | |
|---|
| 93 | self.update(msg="Fourier transform failed.") |
|---|
| 94 | self.complete(transforms=None) |
|---|
| 95 | return |
|---|
| 96 | if self.isquit(): |
|---|
| 97 | return |
|---|
| 98 | self.update(msg="Fourier transform completed.") |
|---|
| 99 | |
|---|
| 100 | transform1 = Data1D(xs, gamma1) |
|---|
| 101 | transform3 = Data1D(xs[xs <= 200], gamma3) |
|---|
| 102 | idf = Data1D(xgamma, idf) |
|---|
| 103 | |
|---|
| 104 | transforms = (transform1, transform3, idf) |
|---|
| 105 | |
|---|
| 106 | self.complete(transforms=transforms) |
|---|
| 107 | |
|---|
| 108 | class HilbertThread(CalcThread): |
|---|
| 109 | def __init__(self, raw_data, extrapolated_data, bg, updatefn=None, |
|---|
| 110 | completefn=None): |
|---|
| 111 | CalcThread.__init__(self, updatefn=updatefn, completefn=completefn) |
|---|
| 112 | self.data = raw_data |
|---|
| 113 | self.background = bg |
|---|
| 114 | self.extrapolation = extrapolated_data |
|---|
| 115 | |
|---|
| 116 | def compute(self): |
|---|
| 117 | qs = self.extrapolation.x |
|---|
| 118 | iqs = self.extrapolation.y |
|---|
| 119 | q = self.data.x |
|---|
| 120 | background = self.background |
|---|
| 121 | |
|---|
| 122 | self.ready(delay=0.0) |
|---|
| 123 | self.update(msg="Starting Hilbert transform.") |
|---|
| 124 | self.ready(delay=0.0) |
|---|
| 125 | if self.isquit(): |
|---|
| 126 | return |
|---|
| 127 | |
|---|
| 128 | # TODO: Implement hilbert transform |
|---|
| 129 | |
|---|
| 130 | self.update(msg="Hilbert transform completed.") |
|---|
| 131 | |
|---|
| 132 | self.complete(transforms=None) |
|---|
