def plot_zipf_law_on_corpus(corpus):
words = getWordsFromCorpus(corpus)
words = remove_stopwords_from_corpus_words(words)
fdist = FreqDist(words)
words = fdist.most_common()
x = [math.log(i[1]) for i in words]
y = [math.log(i) for i in range(1, len(x) + 1)]
(m, b) = pylab.polyfit(x, y, 1)
yp = pylab.polyval([m, b], x)
slope, intercept, r_value, p_value, std_err = stats.linregress(x, y)
pylab.plot(x, y, 'r')
pylab.plot(x, yp, 'b')
pylab.ylim([min(y), max(y)])
pylab.xlim([min(x), max(x)])
pylab.text(x=1,
y=1,
s="Best Fit Line (Blue) \nslope = {slope}".format(
slope=np.round(slope, 2)))
pylab.grid(True)
pylab.ylabel('Counts of words (log)')
pylab.xlabel('Ranks of words (log)')
pylab.title(
'ZIPF LAW TEST ON CORPUS. IDEALLY SLOPE OF THE LINE MUST BE = -1 for IDEAL ZIPF CASE'
)
pylab.show()
def plotScatter(pearsonStats,data,args,color='b'):
""""""
fig = pl.figure()
ax = fig.add_subplot(111)
if args.log:
ax.set_xscale('log')
ax.set_yscale('log')
ax.scatter(data[0],data[1], s=15, c=color, marker='o', alpha=1)
if not args.log:
ax.set_autoscale_on(False)
ax.set_xlabel(args.label1)
ax.set_ylabel(args.label2)
upperLim = max(data[0]+data[1])
m,b = pl.polyfit(data[0],data[1],1)
bfYs = pl.polyval([m,b], [1,max(data[0])])
ax.plot([1,max(data[0])],bfYs,'r-')
pl.text(0.02,0.95,'Pearson: %.4f, %s\nBest Fit: y=%.3f*x+%.3f' % (pearsonStats[0],pearsonStats[1],m,b),
bbox=dict(facecolor='#87AACD', alpha=1),
horizontalalignment='left',
verticalalignment='top',
transform = ax.transAxes)
if args.pdf:
pdf_or_png = 'pdf'
else:
pdf_or_png = 'png'
# construct outfile name
if not args.log:
outName = '%s_%s_vs_%s.%s' % (args.out,args.label1.replace(' ','_'),args.label2.replace(' ','_'),pdf_or_png)
else:
outName = '%s_%s_vs_%s.log.%s' % (args.out,args.label1.replace(' ','_'),args.label2.replace(' ','_'),pdf_or_png)
pl.savefig(outName)
if args.galaxy:
os.rename(outName,args.out)
print 'Show? %s' % (args.show)
if args.show:
pl.show()
def evaluate_models_on_training(x, y, models):
"""
For each regression model, compute the R-squared value for this model with the
standard error over slope of a linear regression line (only if the model is
linear), and plot the data along with the best fit curve.
For the plots, you should plot data points (x,y) as blue dots and your best
fit curve (aka model) as a red solid line. You should also label the axes
of this figure appropriately and have a title reporting the following
information:
degree of your regression model,
R-square of your model evaluated on the given data points,
and SE/slope (if degree of this model is 1 -- see se_over_slope).
Args:
x: an 1-d pylab array with length N, representing the x-coordinates of
the N sample points
y: an 1-d pylab array with length N, representing the y-coordinates of
the N sample points
models: a list containing the regression models you want to apply to
your data. Each model is a pylab array storing the coefficients of
a polynomial.
Returns:
None
"""
for model in models:
predict_y = pylab.polyval(model, x)
#Get type of model
if len(model) <= 4:
types_of_model = ["linear", "quadratic", "cubic"]
model_type = types_of_model[len(model) - 2]
else:
model_type = f"{len(model)-1} degree"
#make the title
title = f"Years against degrees C with {model_type} model \n R2 = {round(r_squared(y,predict_y),5)}"
#If model is linear get se_over_slope and add to title
if len(model) == 2:
title += f"\nStandard error to slope ratio = {round(se_over_slope(x,y,predict_y,model),5)}"
#Draw two pairs of values
pylab.figure()
pylab.plot(x, y, "bo", x, predict_y, "-r")
pylab.title(title)
pylab.xlabel("Years")
pylab.ylabel("Temperature in degrees C")
pylab.show()
def evaluate_models_on_training(x, y, models):
"""
For each regression model, compute the R-square for this model with the
standard error over slope of a linear regression line (only if the model is
linear), and plot the data along with the best fit curve.
For the plots, you should plot data points (x,y) as blue dots and your best
fit curve (aka model) as a red solid line. You should also label the axes
of this figure appropriately and have a title reporting the following
information:
degree of your regression model,
R-square of your model evaluated on the given data points
Args:
x: a list of length N, representing the x-coords of N sample points
y: a list of length N, representing the y-coords of N sample points
models: a list containing the regression models you want to apply to
your data. Each model is a numpy array storing the coefficients of
a polynomial.
Returns:
None
"""
for model in models:
est_y = pylab.polyval(model, x)
error = r_squared(y, est_y)
pylab.figure()
pylab.plot(x, y, 'b o')
pylab.plot(x,
est_y,
'r',
label="degree of your regression model" + "R - Squared is" +
str(round(error, 5)))
pylab.xlabel("x-coords of N sample points")
pylab.ylabel("y-coords of N sample points")
pylab.legend(loc="best")
pylab.title("best fit")
pylab.show()
def plotScatter(pearsonStats,normedTxCntsList,opts):
""""""
fig = pl.figure()
ax = fig.add_subplot(111)
if opts.log:
ax.set_xscale('log')
ax.set_yscale('log')
ax.scatter(normedTxCntsList[0],normedTxCntsList[1], s=15, c='b', marker='o', alpha=1)
if not opts.log:
ax.set_autoscale_on(False)
ax.set_xlabel(opts.name_a)
ax.set_ylabel(opts.name_b)
upperLim = max(normedTxCntsList[0]+normedTxCntsList[1])
m,b = pl.polyfit(normedTxCntsList[0],normedTxCntsList[1],1)
bfYs = pl.polyval([m,b], [1,max(normedTxCntsList[0])])
ax.plot([1,max(normedTxCntsList[0])],bfYs,'r-')
pl.text(0.01,0.99,'Pearson: %.4f, %s\nBest Fit: y=%.3f*x+%.3f' % (pearsonStats[0],pearsonStats[1],m,b),
bbox=dict(facecolor='#87AACD', alpha=1),
horizontalalignment='left',
verticalalignment='top',
transform = ax.transAxes)
mkdirp(opts.dir)
if not opts.log:
pl.savefig('%s%s_vs_%s.png' % (opts.dir,opts.name_a,opts.name_b))
else:
pl.savefig('%s%s_vs_%s.log.png' % (opts.dir,opts.name_a,opts.name_b))
print 'Show? %s' % (opts.show)
if opts.show:
pl.show()
def evaluate_models_on_training(x, y, models):
"""
For each regression model, compute the R-square for this model with the
standard error over slope of a linear regression line (only if the model is
linear), and plot the data along with the best fit curve.
For the plots, you should plot data points (x,y) as blue dots and your best
fit curve (aka model) as a red solid line. You should also label the axes
of this figure appropriately and have a title reporting the following
information:
degree of your regression model,
R-square of your model evaluated on the given data points
Args:
x: a list of length N, representing the x-coords of N sample points
y: a list of length N, representing the y-coords of N sample points
models: a list containing the regression models you want to apply to
your data. Each model is a numpy array storing the coefficients of
a polynomial.
Returns:
None
"""
r2 = []
for model in models:
estimated = np.polyval(model, x)
r2.append(r_squared(y, list(estimated)))
xVals = pylab.array(x)
yVals = pylab.array(y)
#xVals = xVals * 9.81 # get force
pylab.plot(xVals, yVals, 'bo', label='Measured points')
estYVals = pylab.polyval(model, xVals)
pylab.plot(xVals,
estYVals,
'r',
label='Linear fit, k = ' + str(round(1 / model[0], 5)))
pylab.legend(loc='best')
pylab.show()
print(r2)
END of each time step, and fox_populations is a record of the fox population
at the END of each time step.
Both lists should be `numSteps` items long.
"""
ret_rabbit = [CURRENTRABBITPOP]
ret_fox = [CURRENTFOXPOP]
for i in range(numSteps):
if CURRENTFOXPOP >= 10 and CURRENTRABBITPOP >= 10:
rabbitGrowth()
foxGrowth()
ret_rabbit.append(CURRENTRABBITPOP)
ret_fox.append(CURRENTFOXPOP)
return ret_rabbit, ret_fox
rp, fp = runSimulation(200)
rl, = pyplot.plot(rp, label="Rabbit population")
fl, = pyplot.plot(fp, label="Fox population")
pyplot.legend(handles=[rl, fl])
pyplot.show()
rcoeff = pylab.polyfit(range(len(rp)), rp, 2)
rcl, = pyplot.plot(pylab.polyval(rcoeff, range(len(rp))), label="Rabbit Coefficients")
fcoeff = pylab.polyfit(range(len(fp)), fp, 2)
fcl, = pyplot.plot(pylab.polyval(fcoeff, range(len(fp))), label="Fox Coefficients")
pyplot.legend(handles=[rcl, fcl])
pyplot.show()
from geobricks_raster_correlation.core.raster_correlation_core import get_correlation
from matplotlib import pyplot as plt
from matplotlib.pylab import polyfit, polyval
# input to your raster files
raster_path1 = "../../tests/data/geoserver_data_dir/data/workspace/wheat_actual_biomprod_201010_doukkala/wheat_actual_biomprod_201010_doukkala.geotiff"
raster_path2 = "../../tests/data/geoserver_data_dir/data/workspace/wheat_potential_biomprod_201010_doukkala/wheat_potential_biomprod_201010_doukkala.geotiff"
# Number of sampling bins
bins = 150
corr = get_correlation(raster_path1, raster_path2, bins)
x = []
y = []
colors = []
#print corr['series']
for serie in corr['series']:
colors.append(serie['color'])
for data in serie['data']:
x.append(data[0])
y.append(data[1])
# Adding regression line
(m, b) = polyfit(x, y, 1)
yp = polyval([m, b], x)
plt.plot(x, yp)
# plotting scatter
plt.scatter(x, y, c=colors)
plt.show()
if opts.log:
ax.set_xscale('log')
ax.set_yscale('log')
ax.scatter(vecFile[1],vecFile[2], s=15, c='b', marker='o', alpha=1)
if not opts.log:
ax.set_autoscale_on(False)
ax.set_xlabel(vecFile[0][0])
ax.set_ylabel(vecFile[0][1])
upperLim = max(vecFile[1]+vecFile[2])
m,b = pl.polyfit(vecFile[1],vecFile[2],1)
bfYs = pl.polyval([m,b], [1,max(vecFile[1])])
ax.plot([1,max(vecFile[1])],bfYs,'r-')
pl.text(0.01,0.99,'Pearson: %.4f, %s\nBest Fit: y=%.3f*x+%.3f' % (pearson[0],pearson[1],m,b),
bbox=dict(facecolor='#87AACD', alpha=1),
horizontalalignment='left',
verticalalignment='top',
transform = ax.transAxes)
pl.savefig('%s_vs_%s.png' % (vecFile[0][0],vecFile[0][1]))
print 'Show? %s' % (opts.show)
if opts.show:
pl.show()
from geobricks_raster_correlation.core.raster_correlation_core import get_correlation
from matplotlib import pyplot as plt
from matplotlib.pylab import polyfit, polyval
# input to your raster files
raster_path1 = "../../tests/data/geoserver_data_dir/data/workspace/wheat_actual_biomprod_201010_doukkala/wheat_actual_biomprod_201010_doukkala.geotiff"
raster_path2 = "../../tests/data/geoserver_data_dir/data/workspace/wheat_potential_biomprod_201010_doukkala/wheat_potential_biomprod_201010_doukkala.geotiff"
# Number of sampling bins
bins = 150
corr = get_correlation(raster_path1, raster_path2, bins)
x = []
y = []
colors = []
#print corr['series']
for serie in corr['series']:
colors.append(serie['color'])
for data in serie['data']:
x.append(data[0])
y.append(data[1])
# Adding regression line
(m, b) = polyfit(x, y, 1)
yp = polyval([m, b], x)
plt.plot(x, yp)
# plotting scatter
plt.scatter(x, y, c=colors)
plt.show()
#ax.set_xscale('log')
#ax.set_yscale('log')
ax.scatter(vector0,vector1, s=10, c='b', marker='o', alpha=0.6)
# -- set axis labels --
xLab = DEGseqParser._file.name.split('_')[0]
yLab = DEGseqParser._file.name.split('_')[2]
ax.set_xlabel(xLab)
ax.set_ylabel(yLab)
m,b = pl.polyfit(vector0,vector1,1)
min_xVec = min(vector0)
max_xVec = max(vector0)
bfYs = pl.polyval([m,b], [min_xVec,max_xVec])
ax.plot([min_xVec,max_xVec],bfYs,'r-')
pl.text(0.01,0.99,'Pearson: %.4f, %s\nBest Fit: y=%.3f*x+%.3f' % (pearson[0],pearson[1],m,b),
bbox=dict(facecolor='#87AACD', alpha=1),
horizontalalignment='left',
verticalalignment='top',
transform = ax.transAxes)
pl.savefig('%s_vs_%s.indvDEG.png' % (xLab,yLab))
print 'Show? %s' % (opts.show)
if opts.show:
pl.show()
Both lists should be `numSteps` items long.
"""
ret_rabbit = [CURRENTRABBITPOP]
ret_fox = [CURRENTFOXPOP]
for i in range(numSteps):
if CURRENTFOXPOP >= 10 and CURRENTRABBITPOP >= 10:
rabbitGrowth()
foxGrowth()
ret_rabbit.append(CURRENTRABBITPOP)
ret_fox.append(CURRENTFOXPOP)
return ret_rabbit, ret_fox
rp, fp = runSimulation(200)
rl, = pyplot.plot(rp, label="Rabbit population")
fl, = pyplot.plot(fp, label="Fox population")
pyplot.legend(handles=[rl, fl])
pyplot.show()
rcoeff = pylab.polyfit(range(len(rp)), rp, 2)
rcl, = pyplot.plot(pylab.polyval(rcoeff, range(len(rp))),
label="Rabbit Coefficients")
fcoeff = pylab.polyfit(range(len(fp)), fp, 2)
fcl, = pyplot.plot(pylab.polyval(fcoeff, range(len(fp))),
label="Fox Coefficients")
pyplot.legend(handles=[rcl, fcl])
pyplot.show()