def setAnalysisRightGraph(self, term):
raw_data = getRawData()
if term == 0:
x = raw_data.iloc[:, [0, 1, 5]]
else:
x = raw_data.iloc[len(raw_data) - term * 7 : len(raw_data), [0, 1, 5]]
model = PdTable(x)
self.analysis_table_1.setModel(model)
def setMonitorPredicted(self, input_name):
raw_data = getRawData()
if len(raw_data) == 0:
self.warnRawData()
return
PredictWindow(self.ax, input_name, raw_data)
self.canvas.draw()
print(f"{LOG_EXCUTE[9]} 파라미터 : {self.getMonitorInput()}")
def setAllAnalysisGraph(self):
raw_data = getRawData()
if len(raw_data) == 0:
self.warnRawData()
return
term = 0
self.setAnalysisLeftGraph(term)
self.setAnalysisRightGraph(term)
print(LOG_EXCUTE[7])
def setAnalysisGraph(self):
raw_data = getRawData()
if len(raw_data) == 0:
self.warnRawData()
return
term = self.analysis_spin_1.value()
if term * 7 > len(raw_data):
self.warnExceedData()
return
else:
self.setAnalysisLeftGraph(term)
self.setAnalysisRightGraph(term)
print(f"{LOG_EXCUTE[6]} 기간 = {self.analysis_spin_1.value()}")
def getValues(playerName, stat):
rawData = data.getRawData()
values = []
for i, game_id in enumerate(rawData):
playedInGame = False
for player in rawData[game_id][0]:
if player[idx('name')] == playerName:
values.append((player[idx(stat)], i+1))
playedInGame = True
if playedInGame == False:
values.append((None, i+1))
return values
def getValues(playerName, stat):
rawData = data.getRawData()
values = []
for i, game_id in enumerate(rawData):
playedInGame = False
for player in rawData[game_id][0]:
if player[idx('name')] == playerName:
values.append((player[idx(stat)], i + 1))
playedInGame = True
if playedInGame == False:
values.append((None, i + 1))
return values
def setMonitorEstimated(self, input_name):
raw_data = getRawData()
if len(raw_data) == 0:
self.warnRawData()
return
self.ax.cla()
x = raw_data["Time"]
y = raw_data[input_name]
f = raw_data[input_name].rolling(window=5, center=True).mean() #결함31
self.ax.plot(x, y, "-b", x, f, "--r", linewidth=0.8)
self.ax.plot(x, f, "--r", linewidth=1.5)
self.ax.set_title(f"Raw data versus estimated trend, {input_name}")
self.ax.legend(["raw data", "estimated trend"], loc="upper left")
self.ax.set_xlabel("Time(days)")
self.ax.set_ylabel(input_name)
self.ax.grid(True)
self.canvas.draw()
print(f"{LOG_EXCUTE[8]} 파라미터 : {self.getMonitorInput()}")
def main():
sig, bg = getRawData()
sig[sig == 1.] = 10.
bg[bg == 1.] = 10.
'''
newWindow()
plt.hist(bg[:, 0, 0, 2])
newWindow()
plt.hist(bg[:, 0, 0, 2], range=(200, 600))
'''
sig = cutData(sig, 200, 400)
bg = cutData(bg, 200, 400)
sig, bg = evenData(sig, bg)
lundPlot(sig, SIG)
lundPlot(bg, BG)
plt.show()
saveProData(sig, bg)
'''
pt2_f = sig[:, :, :, 4].flatten()
pt2_f = pt2_f[pt2_f != 10]
newWindow()
plt.hist(pt2_f)
newWindow()
binwidth = 1
plt.hist(pt2_f, bins=range(int(np.min(pt2_f)), int(np.max(pt2_f)) + binwidth, binwidth), range=(0, 20))
'''
print('Signal Shape: ' + str(sig.shape))
print('Background Shape: ' + str(bg.shape))
'''
def setAnalysisLeftGraph(self, term):
raw_data = getRawData()
if term == 0:
x = raw_data
else:
x = raw_data.iloc[len(raw_data) - term * 7 : len(raw_data), :]
self.a1.cla()
x1 = x["Time"]
y1 = x["OLR"]
self.a1.plot(x1, y1, "-b", linewidth=0.8)
self.a1.set_xlabel("Time(days)")
self.a1.set_ylabel("OLR")
plt.tight_layout(pad=0.5)
self.a1.grid(True)
self.a2.cla()
x2 = x["Time"]
y2 = x["Biogas Production"]
self.a2.plot(x2, y2, "-r", linewidth=0.8)
self.a2.set_xlabel("Time(days)")
self.a2.set_ylabel("Biogas Production")
plt.tight_layout(pad=0.5)
self.a2.grid(True)
self.a3.cla()
x3 = x["Time"]
x_shift = x.shift(1)
x_shift=x_shift.fillna(0)
y3 = (x["OLR"] - x_shift["OLR"]) / x["OLR"] * 100
y4 = (
(x["Biogas Production"] - x_shift["Biogas Production"])
/ x["Biogas Production"]
* 100
)
self.a3.fill_between(x3, 0, y3, color="b", alpha=0.4)
self.a3.fill_between(x3, 0, y4, color="r", alpha=0.4)
self.a3.set_xlabel("Time(days)")
self.a3.set_ylabel("Variation of OLR and BP")
self.a3.legend(["OLR", "Biogas Production"], loc="lower left")
self.a3.grid(True)
self.canvas.draw()
error = abs(y3 - y4).mean()
total_input_mean = raw_data["OLR"].mean()
total_output_mean = raw_data["Biogas Production"].mean()
input_ratio = x["OLR"].mean() / total_input_mean * 100
output_ratio = x["Biogas Production"].mean() / total_output_mean * 100
error = round(error, 2)
total_input_mean = round(total_input_mean, 2)
total_output_mean = round(total_output_mean, 2)
input_ratio = round(input_ratio, 2)
output_ratio = round(output_ratio, 2)
info_text = f"입력 대비 출력 상관 예측 정확도 : {100-error}%" \
f"\n입력 대비 출력 오차율 (일별) : {error}%" \
f"\n전체 데이터 평균 OLR 부하량 : {total_input_mean} kg VS/m3" \
f"\n전체 데이터 평균 Biogas 출하량 : {total_output_mean} Nm3/kg Vsadd" \
f"\n평균 대비 최근 OLR 부하량 비율 : {input_ratio}%" \
f"\n평균 대비 최근 BP 출하량 비율 : {output_ratio}%"
self.analysis_text_1.setText(info_text)
