Improvements of figures styles

docs/conf.py

@@ -66,9 +66,9 @@
 # the built documents.
 #
 # The short X.Y version.
-version = '0.1.2'
+version = '0.1.3'
 # The full version, including alpha/beta/rc tags.
-release = '0.1.2'
+release = '0.1.3'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

docs/history.rst

@@ -15,4 +15,9 @@ History
 0.1.2 (2020-10-11)
 ------------------
 
-* Minors updates to html documentation.
+* Minors updates to html documentation.
+
+0.1.3 (2020-11-30)
+------------------
+
+* Improvements of figures styles.

docs/structure.rst

@@ -1,7 +1,7 @@
 Structure
 =========
 
-**pyBIMstab** was written under the object-oriented paradigm and is divided into six modules,
+**pySigmaP** was written under the object-oriented paradigm and is divided into six modules,
 one of them contains the class to load and manage the data of the compressibility curve, the other
 five contain the classes that solve the nine methods for getting the preconsolidation pressure
 or yield stress.
@@ -18,4 +18,3 @@ Each module is well documented and contains basic usage examples.
    pachecosilva
    boone
 
-

pysigmap/__init__.py

@@ -12,4 +12,4 @@
 __all__ = ['data', 'casagrande', 'energy', 'bilog', 'pachecosilva', 'boone']
 __author__ = """Exneyder A. Montoya-Araque et al."""
 __email__ = '[email protected]'
-__version__ = '0.1.2'
+__version__ = '0.1.3'

pysigmap/bilog.py

@@ -30,6 +30,10 @@
 plt.rcParams['font.family'] = 'Serif'
 plt.rcParams['font.size'] = 12
 plt.rcParams['text.usetex'] = True
+# High-contrast qualitative colour scheme
+colors = ('#DDAA33',  # yellow
+          '#BB5566',  # red
+          '#004488')  # blue
 
 
 class Bilog():
@@ -49,24 +53,26 @@ class Bilog():
 
     Examples
     --------
-    >>> data = Data(pd.read_csv('testData/testData.csv'), sigmaV=75)
+    >>> urlCSV = ''.join(['https://raw.githubusercontent.com/eamontoyaa/',
+    >>>                   'data4testing/main/pysigmap/testData.csv'])
+    >>> data = Data(pd.read_csv(urlCSV), sigmaV=75)
     >>> method = Bilog(data)
     >>> method.getSigmaP(opt=1)  # Butterfield (1979)
     >>> method.sigmaP, method.ocr
     (433.29446692547555, 5.777259559006341)
-    >>> method.getSigmaP(range2fitLCR=[1000, 9000], opt=1) # Butterfield
+    >>> method.getSigmaP(range2fitCR=[1000, 9000], opt=1) # Butterfield
     >>> method.sigmaP, method.ocr
     (399.3361458736937, 5.324481944982582)
     >>> method.getSigmaP(opt=2)  # Oikawa (1987)
     >>> method.sigmaP, method.ocr
     (433.2944669254731, 5.777259559006308)
-    >>> method.getSigmaP(range2fitLCR=[1000, 9000], opt=2)  # Oikawa
+    >>> method.getSigmaP(range2fitCR=[1000, 9000], opt=2)  # Oikawa
     >>> method.sigmaP, method.ocr
     (399.3361458736935, 5.32448194498258)
     >>> method.getSigmaP(opt=3)  # Onitsuka et al. (1995)
     >>> method.sigmaP, method.ocr
     (433.2944669254753, 5.777259559006338)
-    >>> method.getSigmaP(range2fitLCR=[1000, 9000], opt=3)  # Onitsuka et al.
+    >>> method.getSigmaP(range2fitCR=[1000, 9000], opt=3)  # Onitsuka et al.
     >>> method.sigmaP, method.ocr
     (399.3361458736939, 5.324481944982585)
     """
@@ -76,24 +82,28 @@ def __init__(self, data):
         self.data = data
         return
 
-    def getSigmaP(self, range2fitSCR=None, range2fitLCR=None, opt=1):
+    def getSigmaP(self, range2fitRR=None, range2fitCR=None, opt=1):
         """
         Return the value of the preconsolidation pressure or yield stress.
 
         Parameters
         ----------
-        range2fitSCR : list, tuple or array (length=2), optional
-            Initial and final pressures between which the first-order
-            polynomial will be fit to the compressibility curve on the small
-            compressibility range (SCR). If None, the SCR will be fit from the
-            first point of the compresibility curve to the point before the
-            in-situ vertical effective stress. The default is None.
-        range2fitLCR : list, tuple or array (length=2), optional
-            Initial and final pressures between which the first-order
-            polynomial will be fit to the compressibility curve on the normally
-            consolidated line (NCL), also knonw as the "large compressibility
-            range (LCR)". If None, the NCL will be automatically fit to the
-            last three points of the data.
+        range2fitRR : list, tuple or array (length=2), optional
+            Initial and final pressures between which the first order
+            polynomial will be fit to the data on the recompression range (RR)
+            (range before the preconsolidation pressure). If None, the first
+            order polynomial will be fit from the first point of the curve to
+            the point before the in-situ vertical effective stress. The default
+            is None.
+        range2fitCR : list, tuple or array (length=2), optional
+            Initial and final pressures between which the first order
+            polynomial will be fit to the data on the compression range (CR)
+            (range beyond the preconsolidation pressure). If None, the CR will
+            be automatically fit to the same points used for calculating the
+            compression index with the ``Data`` class only if it was calculated
+            with a linear fit, otherwise, the steepest slope of the cubic
+            spline that passes through the data will be used. The default is
+            None.
         opt : int, optional
             Integer value to indicate which bilogarithmic method will be used.
             Butterfield (opt=1), Oikawa (opt=2) and Onitsuka et al. (opt=3).
@@ -116,33 +126,24 @@ def transformY(y, opt=1):
 
         # def ticks(x, pos): return f'$e^{np.log(x):.0f}$'
 
-        if range2fitSCR is None:  # Indices for fitting the SCR line
-            idxInitSCR = 1
-            idxEndSCR = self.data.findStressIdx(
+        if range2fitRR is None:  # Indices for fitting the RR line
+            idxInitRR = 1
+            idxEndRR = self.data.findStressIdx(
                 stress2find=self.data.sigmaV, cleanedData=True) - 1
         else:
-            idxInitSCR = self.data.findStressIdx(
-                stress2find=range2fitSCR[0], cleanedData=True)
-            idxEndSCR = self.data.findStressIdx(
-                stress2find=range2fitSCR[1], cleanedData=True)
-
-        # -- Linear regresion of points on the small compressibility range(SCR)
-        sigmaSCR = self.data.cleaned['stress'][idxInitSCR: idxEndSCR+1]
-        volSCR = self.data.cleaned['vol'][idxInitSCR: idxEndSCR+1]
-        sigmaSCRlog = transformX(sigmaSCR, opt)
-        volSCRlog = transformY(volSCR, opt)
-        p1_0, p1_1 = polyfit(sigmaSCRlog, volSCRlog, deg=1)
-        r2SCR = r2_score(
-            y_true=volSCRlog, y_pred=polyval(sigmaSCRlog, [p1_0, p1_1]))
-
-        if range2fitLCR is None:  # Indices for fitting the NCL line
-            idxInitNCL = -3
-            idxEndNCL = None
-        else:
-            idxInitNCL = self.data.findStressIdx(
-                stress2find=range2fitLCR[0], cleanedData=True)
-            idxEndNCL = self.data.findStressIdx(
-                stress2find=range2fitLCR[1], cleanedData=True)
+            idxInitRR = self.data.findStressIdx(
+                stress2find=range2fitRR[0], cleanedData=True)
+            idxEndRR = self.data.findStressIdx(
+                stress2find=range2fitRR[1], cleanedData=True) - 1
+
+        # -- Linear regresion of points on the recompression range(RR)
+        sigmaRR = self.data.cleaned['stress'][idxInitRR: idxEndRR+1]
+        volRR = self.data.cleaned['vol'][idxInitRR: idxEndRR+1]
+        sigmaRRlog = transformX(sigmaRR, opt)
+        volRRlog = transformY(volRR, opt)
+        p1_0, p1_1 = polyfit(sigmaRRlog, volRRlog, deg=1)
+        r2RR = r2_score(
+            y_true=volRRlog, y_pred=polyval(sigmaRRlog, [p1_0, p1_1]))
 
         # -- Cubic spline that passes through the data
         sigmaLog = transformX(self.data.cleaned['stress'][1:], opt)
@@ -151,25 +152,25 @@ def transformY(y, opt=1):
         # Specific volume at sigma V
         volSigmaV = cs(transformX(self.data.sigmaV, opt))
 
-        # -- Post yield range or large compressibility range
-        self.maskLCR = np.full(len(self.data.cleaned), False)
-        if range2fitLCR is not None or self.data.fitCc:
-            if range2fitLCR is not None:
-                idxInitLCR = self.data.findStressIdx(
-                    stress2find=range2fitLCR[0], cleanedData=True)
-                idxEndLCR = self.data.findStressIdx(
-                    stress2find=range2fitLCR[1], cleanedData=True)
-                self.maskLCR[idxInitLCR: idxEndLCR] = True
+        # -- Compression range (CR)
+        self.maskCR = np.full(len(self.data.cleaned), False)
+        if range2fitCR is not None or self.data.fitCc:  # Using a linear fit
+            if range2fitCR is not None:
+                idxInitCR = self.data.findStressIdx(
+                    stress2find=range2fitCR[0], cleanedData=True)
+                idxEndCR = self.data.findStressIdx(
+                    stress2find=range2fitCR[1], cleanedData=True)
+                self.maskCR[idxInitCR: idxEndCR] = True
             elif self.data.fitCc:
-                self.maskLCR = self.data.maskCc
+                self.maskCR = self.data.maskCc
             # -- Linear regresion of points on post yield line
-            sigmaLCR = self.data.cleaned['stress'][self.maskLCR]
-            sigmaLCRlog = transformX(sigmaLCR, opt)
-            volLCR = self.data.cleaned['vol'][self.maskLCR]
-            volLCRlog = transformY(volLCR, opt)
-            lcrInt, lcrSlope = polyfit(sigmaLCRlog, volLCRlog, deg=1)
-            r2LCR = r2_score(y_true=volLCRlog,
-                             y_pred=polyval(volLCRlog, [lcrInt, lcrSlope]))
+            sigmaCR = self.data.cleaned['stress'][self.maskCR]
+            sigmaCRlog = transformX(sigmaCR, opt)
+            volCR = self.data.cleaned['vol'][self.maskCR]
+            volCRlog = transformY(volCR, opt)
+            lcrInt, lcrSlope = polyfit(sigmaCRlog, volCRlog, deg=1)
+            r2CR = r2_score(y_true=volCRlog,
+                            y_pred=polyval(sigmaCRlog, [lcrInt, lcrSlope]))
 
         else:  # Using the steepest point of a cubic spline
             sigmaCS = np.linspace(sigmaLog.iloc[0], sigmaLog.iloc[-1], 500)
@@ -187,48 +188,54 @@ def transformY(y, opt=1):
         self.ocr = self.sigmaP / self.data.sigmaV
 
         # -- Lines of the bilogarithmic methods
-        xLCR = np.linspace(self.sigmaP, self.data.cleaned['stress'].iloc[-1])
-        yLCR = polyval(transformX(xLCR, opt), [lcrInt, lcrSlope])
+        xCR = np.linspace(self.sigmaP, self.data.cleaned['stress'].iloc[-1])
+        yCR = polyval(transformX(xCR, opt), [lcrInt, lcrSlope])
 
-        # Small compressibility range
-        xSCRFit = np.linspace(sigmaSCR.iloc[0], self.sigmaP)
-        ySCRFit = polyval(transformX(xSCRFit, opt), [p1_0, p1_1])
+        # Recompression range
+        xRRFit = np.linspace(sigmaRR.iloc[0], self.sigmaP)
+        yRRFit = polyval(transformX(xRRFit, opt), [p1_0, p1_1])
 
         # -- plot compresibility curve
         fig = plt.figure(figsize=[9, 4.8])
         ax = fig.add_axes([0.08, 0.12, 0.55, 0.85])
         ax.semilogx(self.data.raw['stress'][1:],
-                    transformY(self.data.raw['vol'][1:], opt),
-                    basex=xScl, ls='--', marker='o', lw=1, c='k', mfc='w',
-                    label='Data')
+                    transformY(self.data.raw['vol'][1:], opt), basex=xScl,
+                    ls=(0, (1, 1)), marker='o', lw=1.5, c='k', mfc='w',
+                    label='Compressibility curve')
         methods = ['Butterfield', 'Oikawa', r'Onitsuka \textit{et al.}']
-        # Small compressibility range
-        ax.plot(xSCRFit, ySCRFit, ls='--', c='darkred', lw=0.8,
-                label='Small compressibility range')
-        ax.plot(sigmaSCR, volSCRlog, ls='', marker='+', c='darkred',
-                label=f'Data for linear fit\n(R$^2={r2SCR:.3f}$)')
-        # Large compressibility range
-        ax.plot(xLCR, yLCR, ls='--', c='darkgreen', lw=0.8,
-                label='Large compressibility range')
-        if range2fitLCR is not None or self.data.fitCc:
-            ax.plot(sigmaLCR, volLCRlog, ls='', marker='x', c='darkgreen',
-                    label=f'Data for linear fit\n(R$^2={r2LCR:.3f}$)')
+        # Recompression range
+        ax.plot(xRRFit, yRRFit, ls='-', c=colors[2], lw=1.125,
+                label='Recompression range')
+        ax.plot(sigmaRR, volRRlog, ls='', marker='+', c=colors[2],
+                label=f'Data for linear fit\n(R$^2={r2RR:.3f}$)')
+        # Compression range
+        ax.plot(xCR, yCR, ls='-', c=colors[1], lw=1.125,
+                label='Compression range')
+        if range2fitCR is not None or self.data.fitCc:
+            ax.plot(sigmaCR, volCRlog, ls='', marker='x', c=colors[1],
+                    label=f'Data for linear fit\n(R$^2={r2CR:.3f}$)')
         # Other plots
         ax.plot(self.data.sigmaV, volSigmaV, ls='', marker='|', c='r', ms=15,
-                mfc='w', label=str().join([r'$\sigma^\prime_\mathrm{v0}=$ ',
-                                           f'{self.data.sigmaV:.0f} kPa']))
-        ax.plot(self.sigmaP, self.vSigmaP, ls='', marker='D', c='r', ms=5,
-                mfc='w', label=str().join([r'$\sigma^\prime_\mathrm{p}=$ ',
-                                           f'{self.sigmaP:.0f} kPa\n',
-                                           f'OCR= {self.ocr:.1f}']))
+                mfc='w', mew=1.5,
+                label=str().join([r'$\sigma^\prime_\mathrm{v0}=$ ',
+                                  f'{self.data.sigmaV:.0f} kPa']))
+        ax.plot(self.sigmaP, self.vSigmaP, ls='', marker='o', c=colors[0],
+                ms=7, mfc='w', mew=1.5, label=str().join([
+                    r'$\sigma^\prime_\mathrm{p}=$ ',
+                    f'{self.sigmaP:.0f} kPa\n',
+                    f'OCR= {self.ocr:.1f}']))
         # Other details
+        ax.spines['top'].set_visible(False)
+        ax.spines['right'].set_visible(False)
         ax.set(ylabel=f'Specific volume, {yLabel}',
-               xlabel=str().join(['Vertical effective stress ',
-                                  r'$(\sigma^\prime_\mathrm{v})$ [kPa]']))
+               xlabel=str().join(['Vertical effective stress, ',
+                                  r'$\sigma^\prime_\mathrm{v}$ [kPa]']))
         ax.xaxis.set_major_formatter(mtick.ScalarFormatter())
-        ax.grid(True, which="both", ls='--', lw=0.5)
-        ax.legend(bbox_to_anchor=(1.04, 0.5), loc=6, title=str().join([
-            r"\textbf{", f"{methods[opt-1]}", "'s method}"]))
+        # ax.xaxis.set_minor_locator(mtick.AutoMinorLocator())
+        ax.yaxis.set_minor_locator(mtick.AutoMinorLocator())
+        ax.grid(False)
+        ax.legend(bbox_to_anchor=(1.125, 0.5), loc=6, title=str().join([
+            r"\textbf{", f"{methods[opt-1]}", " method}"]))
         return fig