def __init__(self, master):
os.chdir(
'D:/Confidential/Projects/Steel/LD2 BDS/prelim_analysis/data/constructed data/'
)
temp = pd.read_csv('data_dump_24_3_17.csv')
print(temp.shape)
temp = temp.dropna()
print(temp.shape)
X_all_left = pd.DataFrame(temp.iloc[:, 0:20])
X_all_trigger = pd.DataFrame(temp.iloc[:, 20:40])
X_all_right = pd.DataFrame(temp.iloc[:, 40:60])
Y_all_clean = temp.iloc[:, 60]
index_file_clean = temp.iloc[:, 61]
X_all_trigger.columns = [
"X1", "X2", "X3", "X4", "X5", "X6", "X7", "X8", "X9", "X10", "X11",
"X12", "X13", "X14", "X15", "X16", "X17", "X18", "X19", "X20"
]
X_all_left.columns = [
"X1", "X2", "X3", "X4", "X5", "X6", "X7", "X8", "X9", "X10", "X11",
"X12", "X13", "X14", "X15", "X16", "X17", "X18", "X19", "X20"
]
X_all_right.columns = [
"X1", "X2", "X3", "X4", "X5", "X6", "X7", "X8", "X9", "X10", "X11",
"X12", "X13", "X14", "X15", "X16", "X17", "X18", "X19", "X20"
]
print(X_all_trigger.shape, ' : Size of all X trigger')
print(X_all_left.shape, ' : Size of all X left')
print(X_all_right.shape, ' : Size of all X right')
print(Y_all_clean.shape, ' : Size of all Y')
print(index_file_clean.shape, ' : Shape of index file')
temp = X_all_trigger.apply(include_tree, axis=1)
data_logit = pd.concat([X_all_trigger, temp], axis=1)
print(X_all_trigger.shape)
print(data_logit.shape)
global woe_table
woe_table = mf.calc_WOE(data_logit.iloc[:, -1], Y_all_clean)
temp2 = data_logit.apply(subs_WOE, axis=1)
# just to make sure everything is going okay
temp3 = pd.concat([temp2, data_logit.iloc[:, -1]], axis=1)
data_logit = pd.concat([X_all_trigger, temp2], axis=1)
write_data = pd.concat([data_logit, Y_all_clean], axis=1)
print(write_data.shape)
write_data.to_csv('logit_data_6_4.csv', index=False)
clf_logit_trigger = LogisticRegression()
clf_logit_trigger = clf_logit_trigger.fit(data_logit, Y_all_clean)
beta_0 = clf_logit_trigger.intercept_
global beta
beta = clf_logit_trigger.coef_[0].tolist()
beta_0_back = clf_logit_trigger.intercept_
global beta_back
beta_back = clf_logit_trigger.coef_[0].tolist()
print('Initial Values')
print(beta_0, beta)
file_name = "D:/Confidential/Projects/Steel/LD2 BDS/prelim_analysis/data/breakout files/d5122018.csv"
title_name = file_name.split('/')[-1]
file = pd.read_csv(file_name)
file = file.iloc[2500:3000, :].reset_index()
TC_layer = 12
tt1 = 'TC' + str(TC_layer)
tt2 = 'TC' + str(TC_layer + 20)
tt3 = 'TC' + str(TC_layer + 40)
L1 = file.loc[:, tt1]
L2 = file.loc[:, tt2]
L3 = file.loc[:, tt3]
ML = file.loc[:, 'M.level']
CS = file.loc[:, 'C.speed']
CP = file.loc[:, 'C.percent']
MW = file.loc[:, 'M.width']
TC_layer_left = mf.find_left(MW[0], TC_layer)
tt1 = 'TC' + str(TC_layer_left)
tt2 = 'TC' + str(TC_layer_left + 20)
LL1 = file.loc[:, tt1]
LL2 = file.loc[:, tt2]
TC_layer_right = mf.find_right(MW[0], TC_layer)
tt1 = 'TC' + str(TC_layer_right)
tt2 = 'TC' + str(TC_layer_right + 20)
LR1 = file.loc[:, tt1]
LR2 = file.loc[:, tt2]
TC_layer_opp = mf.find_opposite(TC_layer)
tt1 = 'TC' + str(TC_layer_opp)
tt2 = 'TC' + str(TC_layer_opp + 20)
tt3 = 'TC' + str(TC_layer_opp + 40)
LO1 = file.loc[:, tt1]
LO2 = file.loc[:, tt2]
LO3 = file.loc[:, tt3]
# making the continuous x's
temp_x_trigger = mf.make_cont_x(L1, L2, ML, CP, CS, LO1, LO2, MW)
# including the tree based information
temp = temp_x_trigger.apply(include_tree, axis=1)
temp_data = pd.concat([temp_x_trigger, temp], axis=1)
temp2 = temp_data.apply(subs_WOE, axis=1)
temp_x_trigger = pd.concat([temp_x_trigger, temp2], axis=1)
logit_trigger = pd.DataFrame(
clf_logit_trigger.predict_proba(temp_x_trigger))
logit_trigger = list(logit_trigger.iloc[:, -1])
b = list(np.zeros(60))
b.extend(logit_trigger)
b = pd.DataFrame(b)
# Create a container
frame = tkinter.Frame(master)
var = tkinter.StringVar(frame)
var.set("Which feature do you want") # initial value
choices = [
'1 : slope_n(l2,3)', '2 : slope_n(l2,5)',
'3 : sign_present(l1,l2,2)', '4 : sign_present(l1,l2,5)',
'5 : sign_present(l1,l2,7)', '6 : last_nm(l1,5,32)',
'7 : last_nm(l2,5,32)', '8 : np.mean(cp)',
'9 : np.std(l1.iloc[-12:])', '10 : np.std(ml.iloc[-12:])',
'11 : find_l1_peak_slope(l1,lo1,cs)',
'12 : find_l1_peak_ratio(l1,lo1,cs)', '13 : find_drop(l1)',
'14 : find_drop(l2)', '15 : cs_change(cs)', '16 : get_kinks(l2)',
'17 : first_derivative(l2)', '18 : second_derivative(l2)',
'19 : bin_crossover(l1,l2)', '20 : peak_diff(l1,l2)'
]
option = tkinter.OptionMenu(frame, var, *choices)
option.pack(side=tkinter.LEFT)
slider = tkinter.Scale(frame,
from_=-10,
to=10,
resolution=0.01,
orient=tkinter.HORIZONTAL,
length=750)
slider.pack(side=tkinter.LEFT)
self.button_select = tkinter.Button(
frame,
text="Select the value ?",
command=lambda: self.select(var.get(), slider.get(), beta_0, beta,
temp_x_trigger, choices))
self.button_select.pack(side=tkinter.LEFT)
self.button_reset = tkinter.Button(
frame,
text="Reset",
command=lambda: self.reset_func(clf_logit_trigger, beta_0,
temp_x_trigger)).pack(
side=tkinter.LEFT)
self.button_print_beta_all = tkinter.Button(
frame, text="Print All Beta",
command=self.print_beta_all).pack(side=tkinter.LEFT)
self.button_print_beta_selected = tkinter.Button(
frame,
text="Print Selected Beta",
command=lambda: self.print_beta_selected(choices, var.get())).pack(
side=tkinter.LEFT)
## self.textBox = tkinter.Text(frame,height=10, width = 25)
## self.textBox.insert(tkinter.END,beta)
## self.textBox.config(state=tkinter.DISABLED)
## self.textBox.pack(side = tkinter.BOTTOM)
fig = Figure()
fig.suptitle(title_name)
ax = fig.add_subplot(311)
ax.set_ylim(0, 1)
self.line, = ax.plot(range(len(b)), b)
ax.grid(b=True, which='both')
ax2 = fig.add_subplot(312)
ax2.plot(range(len(L1)), L1, range(len(L2)), L2)
ax2.legend(["L1", "L2"])
ax2.grid(b=True, which='both')
ax3 = fig.add_subplot(313)
ax3.plot(range(len(ML)), ML, range(len(CS)), 100 * CS)
ax3.legend(["Mold Level", "100x Casting Speed"])
ax3.grid(b=True, which='both')
self.canvas = FigureCanvasTkAgg(fig, master=master)
self.canvas.show()
self.canvas.get_tk_widget().pack(side='top', fill='both', expand=1)
frame.pack()
count = count + 1
print(count,"/",ttt)
file = pd.read_csv(file_name)
if file.loc[0,'layer'] == 2 :
TC_layer = int(np.mean(file.loc[:,'TC_layer']))
tt1 = 'TC' + str(TC_layer)
tt2 = 'TC' + str(TC_layer + 20)
L1 = file.loc[:,tt1]
L2 = file.loc[:,tt2]
ML = file.loc[:,'M.level']
CS = file.loc[:,'C.speed']
CP = file.loc[:,'C.percent']
MW = file.loc[:,'M.width']
TC_layer_left = mf.find_left(TC_layer,file)
tt1 = 'TC' + str(TC_layer_left)
tt2 = 'TC' + str(TC_layer_left + 20)
LL1 = file.loc[:,tt1]
LL2 = file.loc[:,tt2]
TC_layer_right = mf.find_right(TC_layer,file)
tt1 = 'TC' + str(TC_layer_right)
tt2 = 'TC' + str(TC_layer_right + 20)
LR1 = file.loc[:,tt1]
LR2 = file.loc[:,tt2]
count = count + 1
print(count, "/", ttt, file_name)
file = pd.read_csv(file_name)
if file.loc[0, 'layer'] == 2:
TC_layer = int(np.mean(file.loc[:, 'TC_layer']))
tt1 = 'TC' + str(TC_layer)
tt2 = 'TC' + str(TC_layer + 20)
L1 = file.loc[:, tt1]
L2 = file.loc[:, tt2]
ML = file.loc[:, 'M.level']
CS = file.loc[:, 'C.speed']
CP = file.loc[:, 'C.percent']
MW = file.loc[:, 'M.width']
TC_layer_left = mf.find_left(MW[0], TC_layer)
tt1 = 'TC' + str(TC_layer_left)
tt2 = 'TC' + str(TC_layer_left + 20)
LL1 = file.loc[:, tt1]
LL2 = file.loc[:, tt2]
TC_layer_right = mf.find_right(MW[0], TC_layer)
tt1 = 'TC' + str(TC_layer_right)
tt2 = 'TC' + str(TC_layer_right + 20)
LR1 = file.loc[:, tt1]
LR2 = file.loc[:, tt2]