def knn(p_id):
# get_session_data(p_id)
(exists, test) = get_session_data(p_id)
if not exists:
print("Running KNN Model")
s_id = crud.get_latest_session_table_by_id(db, p_id).split('_')[3]
model = TimeSeriesKMeans.from_json('res/knn_model.txt')
pred = model.predict(test)
leng = len(test)
more = int(0.8 * leng)
less = int(0.2 * leng)
if more + less < leng:
more += 1
elif more + less > leng:
more -= 1
a = np.zeros((more,), dtype=int)
b = np.ones((less,), dtype=int)
true = np.concatenate([a, b])
res = 0
if (pred == 0).sum() < (pred == 1).sum():
res = 1
print("-----------------------------")
print(exists)
print("-----------------------------")
res = models.Result(session_id=int(s_id), patient_id=int(p_id), result=int(res), model_id=0)
return crud.create_patient_result(db, res)
else:
return crud.get_last_result(db, p_id, m_id)
# return crud.get_last_result_by_patient_id(db, p_id)
print(pred)
print(confusion_matrix(true, pred))
def assess(self, stk, stki):
type = ['km.json', 'dba_km.json', 'sdtw_km.json']
asmnt = ['before the pump.', 'near the dump.', 'after the dump.']
gbw = list('gbw')
rating = [0, 0, 0]
srating = [0, 0, 0]
data1 = stki['Close']
data1 = data1.to_numpy()
data1 = data1.reshape(-1, 1)
data1 = data1[np.newaxis, ...]
if data1.shape[1] != 15:
need = 15 - data1.shape[1]
stki = data.DataReader(stk,
start=datetime.today() -
timedelta(days=21 + need, hours=14),
end=self.now,
data_source='yahoo')
data1 = stki['Close']
data1 = data1.to_numpy()
data1 = data1.reshape(-1, 1)
data1 = data1[np.newaxis, ...]
for num in range(3): # no scaler data processing
for num1 in range(3):
model = tsm.from_json('MLModels/' + gbw[num] + type[num1])
bits = model.transform(data1)
bits = np.sum(bits)
rating[num] = rating[num] + bits
data1 = TimeSeriesScalerMeanVariance(1, .5).fit_transform(data1)
for num in range(3): # scaler data processing
for num1 in range(3):
model = tsm.from_json('MLModels/s' + gbw[num] + type[num1])
bits = model.transform(data1)
bits = np.sum(bits)
srating[num] = srating[num] + bits
print("This stock is possibly " + asmnt[srating.index(min(srating))])
return "This stock is possibly " + asmnt[srating.index(min(srating))]
pred
a = np.zeros((320,), dtype=int)
b = np.ones((80,), dtype=int)
true = np.concatenate([a, b])
confusion_matrix(true, pred)
centers = model.cluster_centers_
centers = np.array([centers[0].flatten(), centers[1].flatten()])
centers
plt.plot(centers[0], color = 'red')
for dataset in [Z, O]:
for i in range(1):
plt.plot(dataset.iloc[:, i][0 : 1000], color = 'c')
plt.plot(centers[1], color = 'r')
for dataset in [N, F]:
for i in range(1):
plt.plot(dataset.iloc[:, i][0: 1000], color = 'c')
model.to_json("Model")
model = TimeSeriesKMeans.from_json('Model.txt')
model
pred = model.predict(test)
a = np.zeros((320,), dtype=int)
b = np.ones((80,), dtype=int)
true = np.concatenate([a, b])
confusion_matrix(true, pred)
