forked from connor-create/wsb-ml-trades
-
Notifications
You must be signed in to change notification settings - Fork 0
/
model.py
186 lines (163 loc) · 7.44 KB
/
model.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
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
from keras.metrics import Accuracy
from nltk.corpus.reader.chasen import test
from sklearn.linear_model import SGDClassifier
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.naive_bayes import MultinomialNB
from sklearn.pipeline import Pipeline
from nltk.corpus import stopwords
import sklearn.utils
import pandas as pd
import re
import database
s = set(stopwords.words('english'))
def preprocess_text(sen, symbol):
# Remove punctuations and numbers
sentence = re.sub('[^a-zA-Z]', ' ', sen)
# Single character removal
sentence = re.sub(r"\s+[a-zA-Z]\s+", ' ', sentence)
# Removing multiple spaces
sentence = re.sub(r'\s+', ' ', sentence)
# Remove stopwords
sentence = sentence.lower()
sentence = ' '.join(word for word in sentence.split() if word not in s)
sentence = sentence.replace(' ' + symbol.lower() + ' ', '')
sentence = sentence.replace('deleted', '')
sentence = sentence.replace('removed', '')
return sentence
def build_hour_model():
db = database.Database()
# Collect data to be used and preprocess
allList = db.get_all_post_records()
data = {"1hourWinner": [], "not1Hour": []}
for p in allList:
if p.oneHourWinner:
data['1hourWinner'].append(preprocess_text(p.title + " " + p.description, p.ticker))
else:
data['not1Hour'].append(preprocess_text(p.title + " " + p.description, p.ticker))
# Adjust the dataset
while len(data['1hourWinner']) > len(data['not1Hour']):
data['1hourWinner'].pop()
while len(data['not1Hour']) > len(data['1hourWinner']):
data['not1Hour'].pop()
# Do this calculation 100 times to check average
win_sum = 0.0
lose_sum = 0.0
unclassified_sum = 0.0
num_points = 0
for x in range(100):
# Build dataframes
train_df = pd.DataFrame(columns = ['classification', 'text'])
for t in data['1hourWinner']:
train_df = train_df.append({'classification': '1hourWinner', 'text': t}, ignore_index=True)
for t in data['not1Hour']:
train_df = train_df.append({'classification': 'not1Hour', 'text': t}, ignore_index=True)
train_df = sklearn.utils.shuffle(train_df, random_state = x)
rows_to_keep = round(len(train_df.index) * .80)
test_df = train_df.iloc[:rows_to_keep, :]
train_df = train_df.iloc[rows_to_keep + 1:, :]
# Build a pipeline
text_clf = Pipeline([('vect', CountVectorizer()),
('tfidf', TfidfTransformer()),
('clf', MultinomialNB()),
])
text_clf = text_clf.fit(train_df.text, train_df.classification)
# Check predicted
win_num_correct = 0
lose_num_correct = 0
local_win_total = 0
local_lose_total = 0
unclassified_percent = 0
predicted = text_clf.predict_proba(test_df.text)
i = 0
for index, row in test_df.iterrows():
if predicted[i][1] > .53:
local_lose_total += 1
if row.classification == 'not1Hour':
lose_num_correct += 1
elif predicted[i][0] > .53:
local_win_total += 1
if row.classification == '1hourWinner':
win_num_correct += 1
else:
unclassified_percent += 1
i += 1
print('Win Accuracy:', float(win_num_correct) / float(local_win_total), 'Lose Accuracy:', float(lose_num_correct) / float(local_lose_total))
win_sum += float(win_num_correct) / float(local_win_total)
lose_sum += float(lose_num_correct) / float(local_lose_total)
unclassified_sum += float(unclassified_percent) / float(i)
num_points += 1
if float(win_num_correct) / float(local_win_total) > .58 and float(lose_num_correct) / float(local_lose_total) > .55:
return text_clf
print('Total Long Accuracy:', float(win_sum / num_points))
print('Total Short Accuracy:', float(lose_sum / num_points))
print('Amount unclassified:', float(unclassified_sum / num_points))
return None
def build_30min_model():
db = database.Database()
# Collect data to be used and preprocess
allList = db.get_all_post_records()
data = {"30minuteWinner": [], "not30Minute": []}
for p in allList:
if p.thirtyMinuteWinner:
data['30minuteWinner'].append(preprocess_text(p.title + " " + p.description, p.ticker))
else:
data['not30Minute'].append(preprocess_text(p.title + " " + p.description, p.ticker))
# Adjust the dataset
while len(data['30minuteWinner']) > len(data['not30Minute']):
data['30minuteWinner'].pop()
while len(data['not30Minute']) > len(data['30minuteWinner']):
data['not30Minute'].pop()
# Do this calculation 100 times to check average
win_sum = 0.0
lose_sum = 0.0
unclassified_sum = 0.0
num_points = 0
for x in range(100):
# Build dataframes
train_df = pd.DataFrame(columns = ['classification', 'text'])
for t in data['30minuteWinner']:
train_df = train_df.append({'classification': '30minuteWinner', 'text': t}, ignore_index=True)
for t in data['not30Minute']:
train_df = train_df.append({'classification': 'not30Minute', 'text': t}, ignore_index=True)
train_df = sklearn.utils.shuffle(train_df, random_state = x)
rows_to_keep = round(len(train_df.index) * .80)
test_df = train_df.iloc[:rows_to_keep, :]
train_df = train_df.iloc[rows_to_keep + 1:, :]
# Build a pipeline
text_clf = Pipeline([('vect', CountVectorizer()),
('tfidf', TfidfTransformer()),
('clf', MultinomialNB()),
])
text_clf = text_clf.fit(train_df.text, train_df.classification)
# Check predicted
win_num_correct = 0
lose_num_correct = 0
local_win_total = 0
local_lose_total = 0
unclassified_percent = 0
predicted = text_clf.predict_proba(test_df.text)
i = 0
for index, row in test_df.iterrows():
if predicted[i][1] > .53:
local_lose_total += 1
if row.classification == 'not30Minute':
lose_num_correct += 1
elif predicted[i][0] > .53:
local_win_total += 1
if row.classification == '30minuteWinner':
win_num_correct += 1
else:
unclassified_percent += 1
i += 1
print('Win Accuracy:', float(win_num_correct) / float(local_win_total), 'Lose Accuracy:', float(lose_num_correct) / float(local_lose_total))
win_sum += float(win_num_correct) / float(local_win_total)
lose_sum += float(lose_num_correct) / float(local_lose_total)
unclassified_sum += float(unclassified_percent) / float(i)
num_points += 1
if float(win_num_correct) / float(local_win_total) > .60:
return text_clf
print('Total Long Accuracy:', float(win_sum / num_points))
print('Total Short Accuracy:', float(lose_sum / num_points))
print('Amount unclassified:', float(unclassified_sum / num_points))
return None