def showBarChart(title, xLabel, yLabel, labels, dataList, color, isBarH):
plt.title(title)
plt.xlabel(xLabel)
plt.yLabel(yLabel)
if isBarH:
plt.barh(labels, dataList, color=color)
else:
plt.bar(labels, dataList, color=color)
plt.show()
def plot_hist_comp(data, data2, dataLabels, title='', xLabel=''):
"""!
@ingroup OptiPlot
Histograms and plots the comparison of two sets of function evaluation data.
@param data: \e list \n
Contains the number of function evals for each optimization run. \n
@param data2: \e list \n
Contains the number of function evals for each optimization run for a
second set of runs. \n
@param dataLabels: \e list \n
Contains the legend label names for each data set. \n
@param title: \e string \n
Title for plot. \n
@param xLabel: \e string \n
Label for independent variable. \n
"""
# Allow use of Tex sybols and set formats
plt.rc('text', usetex=True)
plt.rcParams['savefig.dpi'] = 900
# Establish labels for each data set and title for the plot
if xLabel == '':
xLabel = ('\\textbf{# Function Evals}')
yLabel = ('\\textbf{Probability}')
if title == '':
title = "Histogram of Function Evaluations for Optimization"
# Plot Histogram
num = len(data)
w = np.ones_like(data)/float(num)
maxVal = max([max(data), max(data2)])
bins = np.arange(0, maxVal, ceil(maxVal/100))
plt.hist(data, bins=bins, weights=w, facecolor='black',
histtype='stepfilled', alpha=1.0, label=dataLabels[0])
plt.hist(data2, bins=bins, weights=w, facecolor='grey',
histtype='stepfilled', alpha=0.85, label=dataLabels[1])
# Plot Labels
plt.xLabel('\\textbf{%s}' %xLabel, fontsize=15, y=-01.04)
plt.yLabel('\\textbf{%s}' %yLabel, fontsize=15, x=-0.04)
plt.title('\\textbf{%s}' %title, fontsize=17, y=1.04)
plt.yticks(fontsize=14)
plt.xticks(fontsize=14)
# Tweak spacing to prevent clipping of yLabel
plt.subplots_adjust(left=0.15)
# Add legend
plt.legend(borderaxespad=0.75, loc=1, fontsize=14, handlelength=5,
borderpad=0.5, labelspacing=0.75, fancybox=True, framealpha=0.5)
plt.show()
def plot_tlf(alpha=1.5, gamma=1., numSamp=1E7, cutPoint=10.):
"""!
@ingroup OptiPlot
Plots a comparison of the TLF to the Levy distribution.
@param alpha: \e float \n
Levy exponent - defines the index of the distribution and controls
scale properties of the stochastic process.
@param gamma: \e float \n
Gamma - Scale unit of process for Levy flights. \n
@param numSamp: \e integer \n
Number of Levy flights to sample. \n
@param cutPoint: \e float \n
Point at which to cut sampled Levy values and resample. \n
"""
# Initialize variables
l = [] #List to store Levy distribution values
bins = np.array(range(0, int(cutPoint + 1), 1)) / cutPoint
# Calculate the Levy distribution
for i in range(0, len(bins)):
l.append(1/pi*integrate.quad(lambda x: exp(-gamma*x**(alpha)) \
*cos(x*bins[i]*cutPoint),
0, float("inf"), epsabs=0, epsrel=1.e-5,
limit=150)[0]*2)
# Draw numSamp samples from the Levy distribution
levySamp = abs(levy(1, numSamp) / cutPoint).reshape(numSamp)
# Resample values above the range (0,1)
for i in range(len(levySamp)):
while levySamp[i] > 1:
levySamp[i] = abs(levySamp(1, 1) / cutPoint).reshape(1)
#Plot the TLF and Levy distribution on the interval (0,1)
w = np.ones_like(levySamp) / float(
numSamp) #Weights to normalize histogram
plt.rc('text', usetex=True)
ax = plt.subplot(111)
ax.hist(levy, bins=bins, facecolor='grey', weights=w)
ax.plot(bins, l, color='k', linewidth=3.0)
ax.set_yscale("log")
plt.xLabel('\\textbf{z}', fontsize=15, y=-0.04)
plt.yLabel('\\textbf{P(z)}', fontsize=15, x=-0.04)
plt.title('\\textbf{Comparison of TLF and Levy Function}', fontsize=17)
plt.yticks(fontsize=14)
plt.xticks(fontsize=14)
plt.show()
# In[ ]:
td = data[data.TotalDeath == 'Total Deaths']
# In[ ]:
td
# In[ ]:
plt.plot(tc.Dates_epicrv, tc.Dates_epicrv / 10**6)
plt.plot(td.Dates_epicrv, td.Dates_epicrv / 10**6)
plt.xLabel('time')
plt.yLabel('Those infected')
plt.legend(['Total Cases', 'Total Deaths'])
plt.show()
# In[ ]:
