[570a58f9] | 1 | """ |
---|
| 2 | Adds a linear fit plot to the chart |
---|
| 3 | """ |
---|
[fed94a2] | 4 | import re |
---|
[570a58f9] | 5 | import numpy |
---|
| 6 | from PyQt4 import QtGui |
---|
| 7 | from PyQt4 import QtCore |
---|
| 8 | |
---|
| 9 | #from sas.sasgui.transform import |
---|
| 10 | from sas.qtgui.GuiUtils import formatNumber |
---|
| 11 | from sas.sasgui.plottools import fittings |
---|
| 12 | from sas.sasgui.plottools import transform |
---|
| 13 | |
---|
| 14 | from sas.sasgui.plottools.LineModel import LineModel |
---|
| 15 | |
---|
| 16 | # Local UI |
---|
| 17 | from sas.qtgui.UI.LinearFitUI import Ui_LinearFitUI |
---|
| 18 | |
---|
| 19 | class LinearFit(QtGui.QDialog, Ui_LinearFitUI): |
---|
| 20 | def __init__(self, parent=None, |
---|
| 21 | data=None, |
---|
| 22 | max_range=(0.0, 0.0), |
---|
| 23 | fit_range=(0.0, 0.0), |
---|
| 24 | xlabel="", |
---|
| 25 | ylabel=""): |
---|
| 26 | super(LinearFit, self).__init__() |
---|
| 27 | |
---|
| 28 | self.setupUi(self) |
---|
| 29 | assert(isinstance(max_range, tuple)) |
---|
| 30 | assert(isinstance(fit_range, tuple)) |
---|
| 31 | |
---|
| 32 | self.data = data |
---|
| 33 | self.parent = parent |
---|
| 34 | |
---|
| 35 | self.max_range = max_range |
---|
| 36 | self.fit_range = fit_range |
---|
| 37 | self.xLabel = xlabel |
---|
| 38 | self.yLabel = ylabel |
---|
| 39 | |
---|
| 40 | self.txtFitRangeMin.setValidator(QtGui.QDoubleValidator()) |
---|
| 41 | self.txtFitRangeMax.setValidator(QtGui.QDoubleValidator()) |
---|
| 42 | |
---|
| 43 | # Default values in the line edits |
---|
| 44 | self.txtA.setText("1") |
---|
| 45 | self.txtB.setText("1") |
---|
| 46 | self.txtAerr.setText("0") |
---|
| 47 | self.txtBerr.setText("0") |
---|
| 48 | self.txtChi2.setText("0") |
---|
[fed94a2] | 49 | |
---|
[570a58f9] | 50 | # Initial ranges |
---|
| 51 | self.txtRangeMin.setText(str(max_range[0])) |
---|
| 52 | self.txtRangeMax.setText(str(max_range[1])) |
---|
| 53 | self.txtFitRangeMin.setText(str(fit_range[0])) |
---|
| 54 | self.txtFitRangeMax.setText(str(fit_range[1])) |
---|
| 55 | |
---|
[fed94a2] | 56 | # cast xLabel into html |
---|
| 57 | label = re.sub(r'\^\((.)\)(.*)', r'<span style=" vertical-align:super;">\1</span>\2', |
---|
| 58 | str(self.xLabel).rstrip()) |
---|
| 59 | self.lblRange.setText('Fit range of ' + label) |
---|
[570a58f9] | 60 | |
---|
| 61 | self.model = LineModel() |
---|
| 62 | # Display the fittings values |
---|
| 63 | self.default_A = self.model.getParam('A') |
---|
| 64 | self.default_B = self.model.getParam('B') |
---|
| 65 | self.cstA = fittings.Parameter(self.model, 'A', self.default_A) |
---|
| 66 | self.cstB = fittings.Parameter(self.model, 'B', self.default_B) |
---|
| 67 | self.transform = transform |
---|
| 68 | |
---|
[fed94a2] | 69 | # connect Fit button |
---|
[570a58f9] | 70 | self.cmdFit.clicked.connect(self.fit) |
---|
| 71 | |
---|
| 72 | def setRangeLabel(self, label=""): |
---|
| 73 | """ |
---|
| 74 | Overwrite default fit range label to correspond to actual unit |
---|
| 75 | """ |
---|
| 76 | assert(isinstance(label, basestring)) |
---|
| 77 | self.lblRange.setText(label) |
---|
| 78 | |
---|
| 79 | def a(self): |
---|
| 80 | return (float(self.txtA.text()), float(self.txtAerr.text())) |
---|
| 81 | |
---|
| 82 | def b(self): |
---|
| 83 | return (float(self.txtB.text()), float(self.txtBerr.text())) |
---|
| 84 | |
---|
| 85 | def chi(self): |
---|
| 86 | return float(self.txtChi2.text()) |
---|
| 87 | |
---|
| 88 | def range(self): |
---|
| 89 | return (float(self.txtFitRangeMin.text()), float(self.txtFitRangeMax.text())) |
---|
| 90 | |
---|
| 91 | def fit(self, event): |
---|
| 92 | """ |
---|
| 93 | Performs the fit. Receive an event when clicking on |
---|
| 94 | the button Fit.Computes chisqr , |
---|
| 95 | A and B parameters of the best linear fit y=Ax +B |
---|
| 96 | Push a plottable to the caller |
---|
| 97 | """ |
---|
| 98 | tempx = [] |
---|
| 99 | tempy = [] |
---|
| 100 | tempdy = [] |
---|
| 101 | |
---|
| 102 | # Checks to assure data correctness |
---|
| 103 | if len(self.data.view.x) < 2: |
---|
| 104 | return |
---|
| 105 | if not self.checkFitValues(self.txtFitRangeMin): |
---|
| 106 | return |
---|
| 107 | |
---|
| 108 | self.xminFit, self.xmaxFit = self.range() |
---|
| 109 | |
---|
| 110 | xminView = self.xminFit |
---|
| 111 | xmaxView = self.xmaxFit |
---|
| 112 | xmin = xminView |
---|
| 113 | xmax = xmaxView |
---|
| 114 | # Set the qmin and qmax in the panel that matches the |
---|
| 115 | # transformed min and max |
---|
[fed94a2] | 116 | #value_xmin = X_VAL_DICT[self.xLabel].floatTransform(xmin) |
---|
| 117 | #value_xmax = X_VAL_DICT[self.xLabel].floatTransform(xmax) |
---|
| 118 | value_xmin = self.floatInvTransform(xmin) |
---|
| 119 | value_xmax = self.floatInvTransform(xmax) |
---|
| 120 | self.txtRangeMin.setText(formatNumber(value_xmin)) |
---|
| 121 | self.txtRangeMax.setText(formatNumber(value_xmax)) |
---|
[570a58f9] | 122 | |
---|
| 123 | # Store the transformed values of view x, y and dy before the fit |
---|
| 124 | xmin_check = numpy.log10(xmin) |
---|
| 125 | x = self.data.view.x |
---|
| 126 | y = self.data.view.y |
---|
| 127 | dy = self.data.view.dy |
---|
| 128 | |
---|
| 129 | if self.yLabel == "log10(y)": |
---|
| 130 | if self.xLabel == "log10(x)": |
---|
| 131 | tempy = [numpy.log10(y[i]) |
---|
| 132 | for i in range(len(x)) if x[i] >= xmin_check] |
---|
| 133 | tempdy = [transform.errToLogX(y[i], 0, dy[i], 0) |
---|
| 134 | for i in range(len(x)) if x[i] >= xmin_check] |
---|
| 135 | else: |
---|
| 136 | tempy = map(numpy.log10, y) |
---|
| 137 | tempdy = map(lambda t1,t2:transform.errToLogX(t1,0,t2,0),y,dy) |
---|
| 138 | else: |
---|
| 139 | tempy = y |
---|
| 140 | tempdy = dy |
---|
| 141 | |
---|
| 142 | if self.xLabel == "log10(x)": |
---|
| 143 | tempx = [numpy.log10(x) for x in self.data.view.x if x > xmin_check] |
---|
| 144 | else: |
---|
| 145 | tempx = self.data.view.x |
---|
| 146 | |
---|
| 147 | # Find the fitting parameters |
---|
| 148 | # Always use the same defaults, so that fit history |
---|
| 149 | # doesn't play a role! |
---|
| 150 | self.cstA = fittings.Parameter(self.model, 'A', self.default_A) |
---|
| 151 | self.cstB = fittings.Parameter(self.model, 'B', self.default_B) |
---|
| 152 | tempdy = numpy.asarray(tempdy) |
---|
| 153 | tempdy[tempdy == 0] = 1 |
---|
| 154 | |
---|
| 155 | if self.xLabel == "log10(x)": |
---|
| 156 | chisqr, out, cov = fittings.sasfit(self.model, |
---|
| 157 | [self.cstA, self.cstB], |
---|
| 158 | tempx, tempy, |
---|
| 159 | tempdy, |
---|
| 160 | numpy.log10(xmin), |
---|
| 161 | numpy.log10(xmax)) |
---|
| 162 | else: |
---|
| 163 | chisqr, out, cov = fittings.sasfit(self.model, |
---|
| 164 | [self.cstA, self.cstB], |
---|
| 165 | tempx, tempy, tempdy, |
---|
| 166 | xminView, xmaxView) |
---|
| 167 | # Use chi2/dof |
---|
| 168 | if len(tempx) > 0: |
---|
| 169 | chisqr = chisqr / len(tempx) |
---|
| 170 | |
---|
| 171 | # Check that cov and out are iterable before displaying them |
---|
| 172 | errA = numpy.sqrt(cov[0][0]) if cov is not None else 0 |
---|
| 173 | errB = numpy.sqrt(cov[1][1]) if cov is not None else 0 |
---|
| 174 | cstA = out[0] if out is not None else 0.0 |
---|
| 175 | cstB = out[1] if out is not None else 0.0 |
---|
| 176 | |
---|
| 177 | # Reset model with the right values of A and B |
---|
| 178 | self.model.setParam('A', float(cstA)) |
---|
| 179 | self.model.setParam('B', float(cstB)) |
---|
| 180 | |
---|
| 181 | tempx = [] |
---|
| 182 | tempy = [] |
---|
| 183 | y_model = 0.0 |
---|
| 184 | # load tempy with the minimum transformation |
---|
| 185 | if self.xLabel == "log10(x)": |
---|
| 186 | y_model = self.model.run(numpy.log10(xmin)) |
---|
| 187 | tempx.append(xmin) |
---|
| 188 | else: |
---|
| 189 | y_model = self.model.run(xminView) |
---|
| 190 | tempx.append(xminView) |
---|
| 191 | |
---|
| 192 | if self.yLabel == "log10(y)": |
---|
| 193 | tempy.append(numpy.power(10, y_model)) |
---|
| 194 | else: |
---|
| 195 | tempy.append(y_model) |
---|
| 196 | |
---|
| 197 | # load tempy with the maximum transformation |
---|
| 198 | if self.xLabel == "log10(x)": |
---|
| 199 | y_model = self.model.run(numpy.log10(xmax)) |
---|
| 200 | tempx.append(xmax) |
---|
| 201 | else: |
---|
| 202 | y_model = self.model.run(xmaxView) |
---|
| 203 | tempx.append(xmaxView) |
---|
| 204 | |
---|
| 205 | if self.yLabel == "log10(y)": |
---|
| 206 | tempy.append(numpy.power(10, y_model)) |
---|
| 207 | else: |
---|
| 208 | tempy.append(y_model) |
---|
| 209 | # Set the fit parameter display when FitDialog is opened again |
---|
| 210 | self.Avalue = cstA |
---|
| 211 | self.Bvalue = cstB |
---|
| 212 | self.ErrAvalue = errA |
---|
| 213 | self.ErrBvalue = errB |
---|
| 214 | self.Chivalue = chisqr |
---|
| 215 | |
---|
| 216 | # Update the widget |
---|
| 217 | self.txtA.setText(formatNumber(self.Avalue)) |
---|
| 218 | self.txtAerr.setText(formatNumber(self.ErrAvalue)) |
---|
| 219 | self.txtB.setText(formatNumber(self.Bvalue)) |
---|
| 220 | self.txtBerr.setText(formatNumber(self.ErrBvalue)) |
---|
| 221 | self.txtChi2.setText(formatNumber(self.Chivalue)) |
---|
| 222 | |
---|
| 223 | #self.parent.updatePlot.emit((tempx, tempy)) |
---|
| 224 | self.parent.emit(QtCore.SIGNAL('updatePlot'), (tempx, tempy)) |
---|
| 225 | |
---|
| 226 | def checkFitValues(self, item): |
---|
| 227 | """ |
---|
| 228 | Check the validity of input values |
---|
| 229 | """ |
---|
| 230 | flag = True |
---|
| 231 | value = item.text() |
---|
| 232 | p_white = item.palette() |
---|
| 233 | p_white.setColor(item.backgroundRole(), QtCore.Qt.white) |
---|
| 234 | p_pink = item.palette() |
---|
| 235 | p_pink.setColor(item.backgroundRole(), QtGui.QColor(255, 128, 128)) |
---|
| 236 | # Check for possible values entered |
---|
| 237 | if self.xLabel == "log10(x)": |
---|
| 238 | if float(value) > 0: |
---|
| 239 | item.setPalette(p_white) |
---|
| 240 | else: |
---|
| 241 | flag = False |
---|
| 242 | item.setPalette(p_pink) |
---|
| 243 | return flag |
---|
| 244 | |
---|
| 245 | def floatInvTransform(self, x): |
---|
| 246 | """ |
---|
| 247 | transform a float.It is used to determine the x.View min and x.View |
---|
| 248 | max for values not in x. Also used to properly calculate RgQmin, |
---|
| 249 | RgQmax and to update qmin and qmax in the linear range boxes on the |
---|
| 250 | panel. |
---|
| 251 | |
---|
| 252 | """ |
---|
| 253 | # TODO: refactor this. This is just a hack to make the |
---|
| 254 | # functionality work without rewritting the whole code |
---|
| 255 | # with good design (which really should be done...). |
---|
| 256 | if self.xLabel == "x": |
---|
| 257 | return x |
---|
| 258 | elif self.xLabel == "x^(2)": |
---|
| 259 | return numpy.sqrt(x) |
---|
| 260 | elif self.xLabel == "x^(4)": |
---|
| 261 | return numpy.sqrt(math.sqrt(x)) |
---|
| 262 | elif self.xLabel == "log10(x)": |
---|
| 263 | return numpy.power(10, x) |
---|
| 264 | elif self.xLabel == "ln(x)": |
---|
| 265 | return numpy.exp(x) |
---|
| 266 | elif self.xLabel == "log10(x^(4))": |
---|
| 267 | return numpy.sqrt(numpy.sqrt(numpy.power(10, x))) |
---|
| 268 | return x |
---|