-
Notifications
You must be signed in to change notification settings - Fork 2
/
beat_the_benchmark.py
128 lines (96 loc) · 3.77 KB
/
beat_the_benchmark.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
import pandas as pd
from sklearn.feature_extraction.text import ENGLISH_STOP_WORDS
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn import preprocessing
from sklearn.preprocessing import Normalizer
from sklearn.svm import SVC
from sklearn.decomposition import TruncatedSVD
from sklearn.model_selection import cross_validate
from nltk import PorterStemmer
from sklearn.model_selection import StratifiedKFold
pd.set_option('precision', 6)
# Use porter stemmer #
stemmer = PorterStemmer()
# 10-fold #
kf = StratifiedKFold(n_splits=10, random_state=123)
# Read data #
train_data = pd.read_csv('dataSets/train_set.csv', encoding="utf-8", sep="\t")
test_data = pd.read_csv('dataSets/test_set.csv', encoding="utf-8", sep="\t")
# Drop useless columns #
train_data = train_data.drop(['RowNum', 'Id'], axis=1)
y_train = train_data["Category"]
X_train = train_data["Content"]
X_test = test_data["Content"]
X_title = train_data["Title"]
Y_title = test_data["Title"]
# Perform stemming #
lst = []
for i in range(X_train.shape[0]):
s = X_train.iloc[i]
x = []
for t in s.split(" "):
x.append(stemmer.stem(t))
lst.append(" ".join(x))
tmp = pd.DataFrame(lst, columns=["Content"])
X_train = tmp["Content"]
# Perform stemming in test set #
lst = []
for i in range(X_test.shape[0]):
s = X_test.iloc[i]
x = []
for t in s.split(" "):
x.append(stemmer.stem(t))
lst.append(" ".join(x))
tmp = pd.DataFrame(lst, columns=["Content"])
X_test = tmp["Content"]
# Add labels #
le_train = preprocessing.LabelEncoder()
X_train_le = le_train.fit_transform(y_train)
X_train_cat = le_train.inverse_transform(X_train_le)
# Create matrix of TF-IDF features #
# Use title efficiently #
tfidf_vectorizer = TfidfVectorizer(stop_words=ENGLISH_STOP_WORDS)
X_train_tfidf = tfidf_vectorizer.fit_transform(X_train + X_title)
X_test_tfidf = tfidf_vectorizer.transform(X_test + Y_title)
# Normalize data #
norm = Normalizer()
X_train_tfidf = norm.fit_transform(X_train_tfidf)
X_test_tfidf = norm.transform(X_test_tfidf)
# Classifier #
clf = SVC(C=1, kernel="rbf", gamma=10)
# Use LSA for dimensionality reduction #
svd = TruncatedSVD(n_components=100, random_state=123)
# Perform dimensionality reduction #
X_train_reduced = svd.fit_transform(X_train_tfidf)
X_test_tfidf = svd.transform(X_test_tfidf)
# Metrics #
scoring = ['precision_macro', 'recall_macro', 'f1_macro', 'accuracy']
# Evaluate my method #
scores = cross_validate(clf, X_train_reduced, X_train_le, scoring=scoring, cv=kf)
# Print results to csv #
Evaluation_metric_df = pd.read_csv('EvaluationMetric_10fold.csv', sep="\t")
Evaluation_metric_df['My Method'] = [str(round(scores['test_accuracy'].mean(), 6)),
str(round(scores['test_precision_macro'].mean(), 6)),
str(round(scores['test_recall_macro'].mean(), 6)),
str(round(scores['test_f1_macro'].mean(), 6))]
# Create csv #
Evaluation_metric_df.to_csv("EvaluationMetric_10fold.csv", encoding='utf-8', index=False, sep="\t")
# Predict test set #
# Train model #
clf.fit(X_train_reduced, X_train_le)
# Predict categories #
y_test = clf.predict(X_test_tfidf)
y_cat = le_train.inverse_transform(y_test)
# Create csv of predicted categories #
cols = ['Id', 'Category']
lst = []
# Lst: list of lists #
# Every single list #
# contains id and #
# predicted category #
for i in range(test_data.shape[0]):
curr_id = test_data.iloc[i]['Id']
lst.append([curr_id, y_cat[i]])
# Create a dataframe and convert it into csv #
pf = pd.DataFrame(lst, columns=cols)
pf.to_csv("testSet_categories.csv", encoding="utf-8", sep="\t", index=False)