-
Notifications
You must be signed in to change notification settings - Fork 2
/
code.py
127 lines (98 loc) · 4.26 KB
/
code.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
import csv
import pandas as pd
import numpy as np
from collections import defaultdict
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.externals import joblib
df = pd.read_excel('raw_data.xlsx')
data = df.fillna(method='ffill')
data.head()
# Process Disease and Symptom Names
def process_data(data):
data_list = []
data_name = data.replace('^','_').split('_')
n = 1
for names in data_name:
if (n % 2 == 0):
data_list.append(names)
n += 1
return data_list
disease_list = []
disease_symptom_dict = defaultdict(list)
disease_symptom_count = {}
count = 0
for idx, row in data.iterrows():
# Get the Disease Names
if (row['Disease'] !="\xc2\xa0") and (row['Disease'] != ""):
disease = row['Disease']
disease_list = process_data(data=disease)
count = row['Count of Disease Occurrence']
# Get the Symptoms Corresponding to Diseases
if (row['Symptom'] !="\xc2\xa0") and (row['Symptom'] != ""):
symptom = row['Symptom']
symptom_list = process_data(data=symptom)
for d in disease_list:
for s in symptom_list:
disease_symptom_dict[d].append(s)
disease_symptom_count[d] = count
# See that the data is Processed Correctly
disease_symptom_dict
# Count of Disease Occurence w.r.t each Disease
disease_symptom_count
df1 = pd.DataFrame(list(disease_symptom_dict.items()), columns=['Disease', 'Symptom'])
diseases = df1['Disease']
############################################ GENERATING DATASET ################################
import random
l=[]
for i in range(10000):
l.append([])
maxi = 0
mini = 1000000000000
for i in range(10000):
l[i].append(diseases[random.randint(0,148)])
l[i].append(random.randint(0,23))
l[i].append(random.randint(0,59))
l[i].append(random.randint(0,59))
l[i].append(random.randint(0,20))
l[i].append(random.randint(0,100))
l[i].append((l[i][1]-11)*(l[i][1]-19) + 20 + l[i][2]**2 + l[i][3] + 160000 - l[i][4] + l[i][5]**3/1000.0 + ((l[i][1] - 11)*(l[i][1] - 19) + 20 + l[i][2]**2 + l[i][3])*(160000 - l[i][4]**4)*(l[i][5]**3) + random.randint(0,148)**4/10000.0 + (random.randint(0,148)**4/10000.0)*((l[i][1] - 11)*(l[i][1] - 19) + 20 + l[i][2]**2 + l[i][3])*(160000 - l[i][4]**4)*(l[i][5]**3))
maxi = max(l[i][6], maxi)
mini = min(l[i][6], mini)
for i in range(10000):
l[i][6] = (l[i][6] - mini)/(maxi - mini)*5.0
train_data = pd.DataFrame(l, columns = ['Disease', 'hrs', 'mins', 'secs', 'num_docs', 'num_pen_req', 'time_taken'])
################################## DL MODEL ################################################
y = train_data['time_taken']
train_data.drop(['time_taken'], axis = 1, inplace = True)
from sklearn.preprocessing import StandardScaler,OneHotEncoder,LabelEncoder
from keras.models import Sequential
from keras.layers import Dense, Dropout
le=LabelEncoder()
train_data['Disease']=le.fit_transform(train_data['Disease'])
oe=OneHotEncoder(categorical_features=[0])
train_data=oe.fit_transform(train_data).toarray()
sc = StandardScaler()
train_data = sc.fit_transform(train_data)
joblib.dump(sc,"scaler.save")
joblib.dump(le,"label.save")
joblib.dump(oe,"onehot.save")
model = Sequential()
model.add(Dense(70, kernel_initializer = 'uniform', activation = 'relu', input_shape=(154,)))
model.add(Dropout(0.5))
model.add(Dense(30, kernel_initializer = 'uniform', activation = 'relu'))
model.add(Dropout(0.4))
model.add(Dense(10, kernel_initializer = 'uniform', activation = 'relu'))
model.add(Dropout(0.3))
model.add(Dense(1, activation = 'relu'))
model.compile(optimizer = "adam", loss = "mean_squared_error", metrics = ['mse'])
model.fit(train_data, y, epochs = 100, batch_size = 64)
model.save_weights("weights.h5")
model.save("model.h5")
############################# TEST ##############################################
import pandas as pd
test_data = pd.DataFrame([['anemia', '17', 1, 1, 50, 500]], columns = ['Disease', 'hrs', 'mins', 'secs', 'num_docs', 'num_pen_req'])
test_data['Disease'] = le.transform(test_data['Disease'])
test_data = oe.transform(test_data).toarray()
test_data = sc.transform(test_data)
x=model.predict(test_data)