def _spearman_r(X, Y):
"""
Calculates a Spearman rank-order correlation coefficient
and the p-value to test for non-correlation.
"""
rho, p_value = spearman(X, Y)
return rho
def _spearman_r(X, Y):
"""
Calculates a Spearman rank-order correlation coefficient
and the p-value to test for non-correlation.
"""
rho, p_value = spearman(X, Y)
return rho
def __get_speaker_accommodation(self, speaker: str):
accommodationPath = self.cfg.paths[
'accommPath'] + speaker + self.fileSuffix
for task in self.__get_tasks(speaker):
with open(accommodationPath, 'r') as f:
reader = csv.reader(f)
times, distances = self.__extract_task_ratios(reader, task)
correlation = stats.spearman(times, distances)
spearCoefficient = correlation[0]
pValue = correlation[1]
taskLength = max(
times
) # Approximate the task end by the start of the last word.
row = [
speaker + " " + task, spearCoefficient,
abs(spearCoefficient), pValue, taskLength
]
print(",".join(row))
pos2 = np.stack(([values_LesothoN,values_SAN]),axis = 0)
print(np.cov(pos1))
print(np.cov(pos2))
rv = mvn([mean_Lesotho,mean_SA],np.cov(pos1))
rv1 = mvn([mean_LesothoN,mean_SAN],np.cov(pos2))
x1,y1 = np.mgrid[300:450:0.5, 300:400:0.5]
posnew = np.dstack((x1,y1))
levels = [0.01,0.05, 0.10, 0.50, 0.90,0.95,0.99]
levels1 = [0.88,0.90,0.91,0.92,0.93, 0.94,0.95, 0.96,0.965,0.97, 0.975,0.976,0.977,0.978,0.979,0.98,0.981,0.982,0.983,0.985,0.987, 0.99,0.991, 0.992, 0.993, 0.994,0.995, 0.996, 0.997, 0.998, 0.999]
kendal = kdtau(values_Lesotho,values_SA)
spearman = spearman(values_Lesotho,values_SA)
print(kendal)
print(spearman[0])
####
data_normalized = pd.DataFrame({'val-L':values_Lesotho-np.mean(values_Lesotho),'val-SA': values_SA-np.mean(values_SA)})
data_normalized_N = pd.DataFrame({'val-L':values_LesothoN-np.mean(values_LesothoN),'val-SA': values_SAN-np.mean(values_SAN)})
data_small = data_normalized[data_normalized['val-L']<0 & (data_normalized['val-SA'] < 0)]
data_small1 = data_normalized_N[data_normalized_N['val-L']<0 & (data_normalized_N['val-SA'] < 0)]
print(np.corrcoef(data_small['val-L'],data_small['val-SA']))
print(np.corrcoef(data_small1['val-L'],data_small1['val-SA']))
fig, ax = plt.subplots()
ax.plot(values_Lesotho,values_SA,'ro',markersize='2.0')
ax.plot(values_LesothoN,values_SAN,'bo', markersize='2.0')
plt.title('Receiver Operating Characteristic')
plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)
plt.legend(loc = 'lower right')
plt.plot([0, 1], [0, 1],'r--')
plt.xlim([0, 1])
plt.ylim([0, 1])
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show()
plot_roc_curve()
### SCC and PCC
scc = stats.spearman(pred, targets)
def plot_scc():
plt.title('Spearmans Correlation Coefficient')
plt.scatter(y_test.flatten().detach().numpy(), pred.flatten().detach().numpy(), label = 'SCC = %0.2f' % scc)
plt.legend(loc = 'lower right')
plt.ylabel('Predicted')
plt.xlabel('Validation targets')
plt.show()
pcc = stats.pearsonr(pred, targets)
def plot_scc():
plt.title('Pearsons Correlation Coefficient')
plt.scatter(y_test.flatten().detach().numpy(), pred.flatten().detach().numpy(), label = 'PCC = %0.2f' % pcc)
plt.legend(loc = 'lower right')
plt.ylabel('Predicted')