def test_result(self,data):
mat=self.get_ConfusionMat(data)
print(" ")
print("-------- Confusion_matrix ---------_")
print(" ")
print(np.matrix(mat))
hp.get_result(mat)
def predict(id):
if request.method == 'POST':
return redirect('/')
else:
try:
result, pred = get_result(id)
return render_template('predict.html', result=result, pred=pred)
except:
return redirect('/invalid_id')
"""
TP: correct prediction in the positive classes
TN: correct prediction in the negative classes
FP: the number of over-predictions in the signal peptide class
FN: the number of under-predictions in the signal peptide class
"""
X_p = readdata('./dataset/test_SP.fasta_out', LENGTH, 'test')
X_n_1 = readdata('./dataset/test_TM.fasta_out', LENGTH, 'test')
X_n_2 = readdata('./dataset/test_NC.fasta_out', LENGTH, 'test')
p_pred = model.predict(X_p)
n_1_pred = model.predict(X_n_1)
n_2_pred = model.predict(X_n_2)
S_S, S_N = get_result(p_pred, 0)
N_S_1, N_N_1 = get_result(n_1_pred, 1)
N_S_2, N_N_2 = get_result(n_2_pred, 2)
TP = S_S
TN = N_N_1 + N_N_2
FP = N_S_1 + N_S_2
FN = S_N
print()
print('TP:', TP, 'TN:', TN, 'FP:', FP, 'FN:', FN)
print('S: ', S_S, S_N)
print('T: ', N_S_1, N_N_1)
print('N: ', N_S_2, N_N_2)
divv = TP * TN - FP * FN