-
Notifications
You must be signed in to change notification settings - Fork 0
/
menu_inference.py
251 lines (220 loc) · 9.8 KB
/
menu_inference.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
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
import streamlit as st
import pandas as pd
import plotly.express as px
import plotly.graph_objs as go
from plotly.subplots import make_subplots
from wordcloud import WordCloud
import matplotlib.pyplot as plt
import re
from collections import Counter
def display_inference():
# read data
df_comments = pd.read_csv('data/data_complete_visualization.csv')
df_coronas = pd.read_csv('data/data_corona_semarang_14112021.csv')
df_sentiment_counter = pd.read_csv('data/sentiment_counter.csv')
# convert data types
df_comments['datetime'] = pd.to_datetime(df_comments['datetime'])
df_coronas['Tanggal'] = pd.to_datetime(df_coronas['Tanggal'])
df_sentiment_counter['date'] = pd.to_datetime(df_sentiment_counter['date'])
df_sentiment_counter['neutral'] = pd.to_numeric(df_sentiment_counter['neutral'])
df_sentiment_counter['positive'] = pd.to_numeric(df_sentiment_counter['positive'])
df_sentiment_counter['negative'] = pd.to_numeric(df_sentiment_counter['negative'])
# filter only for data for inference
df_comments = df_comments[df_comments['data_type'] == 'inference'].reset_index(drop=True)
# get start date and end date
start_date = df_comments['datetime'].min()
end_date = df_comments['datetime'].max()
df_coronas = df_coronas[(df_coronas['Tanggal'] >= start_date) & (df_coronas['Tanggal'] <= end_date)]
df_sentiment_counter = df_sentiment_counter[(df_sentiment_counter['date'] >= start_date) & (
df_sentiment_counter['date'] <= end_date)]
text = """
# Model Inference
Model machine learning yang sudah dilatih digunakan untuk memprediksi data baru (*inference*). Berdasarkan
tahap modeling, model machine learning terbaik adalah `TF-IDF SVM` dengan akurasi skor `74.60%` pada data test.
### Data inference
Data baru yang digunakan untuk proses *inference* memiliki deskripsi sebagai berikut.
"""
st.markdown(text, unsafe_allow_html=True)
st.markdown(f"Shape dari data inference: {df_comments.shape[0]} baris, {df_comments.shape[1]} kolom")
st.markdown(
f'Rentang waktu: `{df_comments["datetime"].min().strftime("%d %b %Y")}` sampai `{df_comments["datetime"].max().strftime("%d %b %Y")}`')
st.dataframe(df_comments.head(10))
st.info('Data yang ditampilkan sudah dilakukan *preprocessing*, *masking*, dan diprediksi labelnya.')
text = """
### Hasil inference
Berikut ini adalah hasil inference model pada data baru.
| Sentiment | Jumlah |
|------------|--------|
| `neutral` | 8718 |
| `positive` | 3050 |
| `negative` | 3491 |
<br>
"""
st.markdown(text, unsafe_allow_html=True)
temp = df_comments['label'].value_counts().reset_index()
temp = temp.rename(columns={'index': 'label', 'label': 'count'})
# funnel chart
st.markdown("Untuk visualisasi lebih baik, kita buat Funnel-Chart untuk mengetahui persentase dari tiap sentimen.")
fig_funnel = px.funnel_area(names=temp['label'],
values=temp['count'],
title='Funnel-Chart of Sentiment Distribution',
color_discrete_map={'neutral': '#636EFA', 'positive': '#00CC96', 'negative': '#EF553B'})
st.plotly_chart(fig_funnel)
# make plot
fig = make_subplots(specs=[[{"secondary_y": True}]])
fig.add_trace(go.Bar(x=df_coronas['Tanggal'], y=df_coronas['POSITIVE ACTIVE'],
name='Positive Active',
marker_color='rgba(255, 161, 90, 0.8)',
marker_line_width=0),
secondary_y=False)
fig.add_trace(go.Scatter(x=df_sentiment_counter['date'], y=df_sentiment_counter['neutral'],
mode='lines+markers', marker_color='#636EFA', name='Sentiment Neutral'),
secondary_y=True)
fig.add_trace(go.Scatter(x=df_sentiment_counter['date'], y=df_sentiment_counter['positive'],
mode='lines+markers', marker_color='#00CC96', name='Sentiment Positive'),
secondary_y=True)
fig.add_trace(go.Scatter(x=df_sentiment_counter['date'], y=df_sentiment_counter['negative'],
mode='lines+markers', marker_color='#EF553B', name='Sentiment Negative'),
secondary_y=True)
# Add figure title
fig.update_layout(
title_text="COVID-19 Positive Active Cases vs Sentiment (Neutral, Positive, Negative)",
template='presentation',
plot_bgcolor='rgb(275, 270, 273)'
)
# Set y-axes titles
fig.update_yaxes(title_text="Number of Sentiments",
secondary_y=True, rangemode='tozero')
fig.update_yaxes(title_text="Number of Cases",
secondary_y=False, rangemode='tozero')
st.plotly_chart(fig, use_container_width=True)
# WORD CLOUD
# filter data for each sentiment
neutral = df_comments[df_comments.label == "neutral"]
positive = df_comments[df_comments.label == "positive"]
negative = df_comments[df_comments.label == "negative"]
# create slider for the number of words
n_words = st.slider("Set the number of words in Word Cloud",
min_value=50, max_value=200, step=10)
# create 3 columns
col1, col2, col3 = st.columns(3)
# neutral
with col1:
texts = ''
for val in neutral['text']:
val = str(val)
tokens = val.split()
for i in range(len(tokens)):
tokens[i] = re.sub(r'[^\x00-\x7F]+', ' ', tokens[i])
tokens[i] = tokens[i].strip()
texts += " ".join(tokens)+" "
wc = WordCloud(
colormap="Blues",
mode="RGBA",
max_words=n_words,
background_color="white",
collocations=False,
width=400, height=400,
)
wc.generate(texts)
st.markdown("#### Sentiment Neutral", unsafe_allow_html=False)
# plot
fig = plt.figure(figsize=(20, 8), dpi=80)
plt.imshow(wc, interpolation='bilinear')
plt.axis("off")
plt.tight_layout()
st.pyplot(fig)
# top 10 words
top10 = Counter(texts.split()).most_common(10)
df_top10 = pd.DataFrame(top10, columns=['word', 'freq'])
df_top10 = df_top10.sort_values(by='freq', ascending=True)
# plot
fig_bar = px.bar(df_top10, x="freq", y="word",
orientation='h', height=400, width=350)
fig_bar.update_layout(title_text='Top 10 words sentiment neutral',
plot_bgcolor='rgb(275, 270, 273)')
fig_bar.update_traces(marker_color='#636EFA')
# Set x-axes and y-axes titles
fig_bar.update_yaxes(title_text="")
fig_bar.update_xaxes(title_text="frequency")
st.plotly_chart(fig_bar)
# positive
with col2:
texts = ''
for val in positive['text']:
val = str(val)
tokens = val.split()
for i in range(len(tokens)):
tokens[i] = re.sub(r'[^\x00-\x7F]+', ' ', tokens[i])
tokens[i] = tokens[i].strip()
texts += " ".join(tokens)+" "
wc = WordCloud(
colormap="Greens",
mode="RGBA",
max_words=n_words,
background_color="white",
collocations=False,
width=400, height=400,
)
wc.generate(texts)
st.markdown("#### Sentiment Positive", unsafe_allow_html=False)
# plot
fig = plt.figure(figsize=(20, 8), dpi=80)
plt.imshow(wc, interpolation='bilinear')
plt.axis("off")
plt.tight_layout()
st.pyplot(fig)
# top 10 words
top10 = Counter(texts.split()).most_common(10)
df_top10 = pd.DataFrame(top10, columns=['word', 'freq'])
df_top10 = df_top10.sort_values(by='freq', ascending=True)
# plot
fig_bar = px.bar(df_top10, x="freq", y="word",
orientation='h', height=400, width=350)
fig_bar.update_layout(title_text='Top 10 words sentiment positive',
plot_bgcolor='rgb(275, 270, 273)')
fig_bar.update_traces(marker_color='#00CC96')
# Set x-axes and y-axes titles
fig_bar.update_yaxes(title_text="")
fig_bar.update_xaxes(title_text="frequency")
st.plotly_chart(fig_bar)
# negative
with col3:
texts = ''
for val in negative['text']:
val = str(val)
tokens = val.split()
for i in range(len(tokens)):
tokens[i] = re.sub(r'[^\x00-\x7F]+', ' ', tokens[i])
tokens[i] = tokens[i].strip()
texts += " ".join(tokens)+" "
wc = WordCloud(
colormap="Reds",
mode="RGBA",
max_words=n_words,
background_color="white",
collocations=False,
width=400, height=400,
)
wc.generate(texts)
st.markdown("#### Sentiment Negative", unsafe_allow_html=False)
# plot
fig = plt.figure(figsize=(20, 8), dpi=80)
plt.imshow(wc, interpolation='bilinear')
plt.axis("off")
plt.tight_layout()
st.pyplot(fig)
# top 10 words
top10 = Counter(texts.split()).most_common(10)
df_top10 = pd.DataFrame(top10, columns=['word', 'freq'])
df_top10 = df_top10.sort_values(by='freq', ascending=True)
# plot
fig_bar = px.bar(df_top10, x="freq", y="word",
orientation='h', height=400, width=350)
fig_bar.update_layout(title_text='Top 10 words sentiment negative',
plot_bgcolor='rgb(275, 270, 273)')
fig_bar.update_traces(marker_color='#EF553B')
# Set x-axes and y-axes titles
fig_bar.update_yaxes(title_text="")
fig_bar.update_xaxes(title_text="frequency")
st.plotly_chart(fig_bar)