pos_track5.append(pos)
# can6 = [pool2[i] for i in sort_order_certain6[start:start + step]]
# train6.extend(can6)
# pos = Counter(labels[train6])["yes"]
# pos_track6.append(pos)
start = start + step
print("Round #{id} passed\r".format(id=round), end="")
result["begin"] = begin
result["stable"] = stable
result["x"] = steps
result["linear_review"] = pos_track
result["simple_active"] = pos_track2
result["semi_continuous_aggressive"] = pos_track3
result["continuous_active"] = pos_track4
result["aggressive_undersampling"] = pos_track5
# result["smote"] = pos_track6
result["continuous_aggressive"] = pos_track7
result["semi_continuous"] = pos_track8
result["new_continuous_aggressive"] = pos_track9
return result
if __name__ == "__main__":
eval(cmd())
datalist=['beck-s','farmer-d','kaminski-v','kitchen-l','lokay-m','sanders-r','williams-w3']
plt.figure()
result={}
for classifier in c_name:
Y_100 = list()
for filename in datalist:
with open('./dump/'+classifier+'_'+filename+'.pickle', 'rb') as handle:
result[filename] = pickle.load(handle)
tmp = result[filename]['F_M']
tmp100 = np.median(tmp['100'])
t = Y_100 + [tmp100]
Y_100 = t
line,=plt.plot(range( len(datalist) ),Y_100,label=classifier)
plt.xticks(range( len(datalist) ), (datalist))
plt.yticks(np.arange(0,1.0,0.2))
plt.ylabel("F_M score")
plt.xlabel("Training Folder")
plt.legend(bbox_to_anchor=(0.35, 1), loc=1, ncol = 1, borderaxespad=0.)
plt.savefig("../Results/comp_classifier.eps")
plt.savefig("../Results/comp_classifier.png")
if __name__ == '__main__':
eval(cmd())
