Spaces:
Runtime error
Runtime error
add charts
Browse files- .ipynb_checkpoints/Untitled-checkpoint.ipynb +0 -0
- .ipynb_checkpoints/app-checkpoint.py +19 -3
- .ipynb_checkpoints/requirements-checkpoint.txt +2 -1
- .ipynb_checkpoints/utils-checkpoint.py +7 -14
- Untitled.ipynb +0 -0
- app.py +19 -3
- requirements.txt +2 -1
- utils.py +7 -14
.ipynb_checkpoints/Untitled-checkpoint.ipynb
ADDED
The diff for this file is too large to render.
See raw diff
|
|
.ipynb_checkpoints/app-checkpoint.py
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
import streamlit as st
|
|
|
2 |
from utils import *
|
3 |
|
4 |
########## Title for the Web App ##########
|
@@ -10,9 +11,24 @@ feedback = st.text_input('Type your text here', 'The website was user friendly a
|
|
10 |
if st.button('Click for predictions!'):
|
11 |
with st.spinner('Generating predictions...'):
|
12 |
|
13 |
-
|
14 |
-
|
15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
st.write("\n")
|
18 |
st.subheader('Or... Upload a csv file if you have a file instead.')
|
|
|
1 |
import streamlit as st
|
2 |
+
import plotly.express as px
|
3 |
from utils import *
|
4 |
|
5 |
########## Title for the Web App ##########
|
|
|
11 |
if st.button('Click for predictions!'):
|
12 |
with st.spinner('Generating predictions...'):
|
13 |
|
14 |
+
topics_prob, sentiment_prob = get_single_prediction(feedback)
|
15 |
+
|
16 |
+
bar = px.bar(topics_prob, x='probability', y='topic')
|
17 |
+
st.plotly_chart(bar, use_container_width=True)
|
18 |
+
|
19 |
+
pie = px.pie(sentiment_prob,
|
20 |
+
values='probability',
|
21 |
+
names='sentiment',
|
22 |
+
title='Sentiment Probability',
|
23 |
+
color_discrete_map={'positive':'rgb(0, 204, 0)',
|
24 |
+
'negative':'rgb(215, 11, 11)'
|
25 |
+
},
|
26 |
+
color='sentiment'
|
27 |
+
)
|
28 |
+
st.plotly_chart(pie, use_container_width=True)
|
29 |
+
|
30 |
+
|
31 |
+
#st.success(f'Your text has been predicted to fall under the following topics: {result[:-1]}. The sentiment of this text is {result[-1]}.')
|
32 |
|
33 |
st.write("\n")
|
34 |
st.subheader('Or... Upload a csv file if you have a file instead.')
|
.ipynb_checkpoints/requirements-checkpoint.txt
CHANGED
@@ -4,4 +4,5 @@ transformers==4.16.1
|
|
4 |
scikit-learn
|
5 |
pandas==1.2.4
|
6 |
torch==1.10.1
|
7 |
-
numpy==1.19.5
|
|
|
|
4 |
scikit-learn
|
5 |
pandas==1.2.4
|
6 |
torch==1.10.1
|
7 |
+
numpy==1.19.5
|
8 |
+
plotly==5.1.0
|
.ipynb_checkpoints/utils-checkpoint.py
CHANGED
@@ -39,23 +39,16 @@ def get_single_prediction(text):
|
|
39 |
text_vectors = np.mean([w2v[i] for i in text.split()], axis=0)
|
40 |
|
41 |
# Make predictions
|
42 |
-
results = model.
|
43 |
-
|
44 |
|
45 |
# Get sentiment
|
46 |
-
|
47 |
-
|
48 |
-
|
|
|
49 |
|
50 |
-
|
51 |
-
pred_labels = [labels[idx] for idx, tag in enumerate(results) if tag == 1]
|
52 |
-
|
53 |
-
if len(pred_labels) == 0:
|
54 |
-
pred_labels.append('others')
|
55 |
-
|
56 |
-
pred_labels.append(sentiment)
|
57 |
-
|
58 |
-
return pred_labels
|
59 |
|
60 |
def get_multiple_predictions(csv):
|
61 |
|
|
|
39 |
text_vectors = np.mean([w2v[i] for i in text.split()], axis=0)
|
40 |
|
41 |
# Make predictions
|
42 |
+
results = model.predict_proba(text_vectors.reshape(1,300)).squeeze().round(2)
|
43 |
+
pred_prob = pd.DataFrame({'topic': labels, 'probability': results}).sort_values('probability', ascending=True)
|
44 |
|
45 |
# Get sentiment
|
46 |
+
sentiment_results = classifier(text,
|
47 |
+
candidate_labels=['positive', 'negative'],
|
48 |
+
hypothesis_template='The sentiment of this is {}')
|
49 |
+
sentiment_prob = pd.DataFrame({'sentiment': sentiment_results['labels'], 'probability': sentiment_results['scores']})
|
50 |
|
51 |
+
return (pred_prob, sentiment_prob)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
def get_multiple_predictions(csv):
|
54 |
|
Untitled.ipynb
ADDED
The diff for this file is too large to render.
See raw diff
|
|
app.py
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
import streamlit as st
|
|
|
2 |
from utils import *
|
3 |
|
4 |
########## Title for the Web App ##########
|
@@ -10,9 +11,24 @@ feedback = st.text_input('Type your text here', 'The website was user friendly a
|
|
10 |
if st.button('Click for predictions!'):
|
11 |
with st.spinner('Generating predictions...'):
|
12 |
|
13 |
-
|
14 |
-
|
15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
st.write("\n")
|
18 |
st.subheader('Or... Upload a csv file if you have a file instead.')
|
|
|
1 |
import streamlit as st
|
2 |
+
import plotly.express as px
|
3 |
from utils import *
|
4 |
|
5 |
########## Title for the Web App ##########
|
|
|
11 |
if st.button('Click for predictions!'):
|
12 |
with st.spinner('Generating predictions...'):
|
13 |
|
14 |
+
topics_prob, sentiment_prob = get_single_prediction(feedback)
|
15 |
+
|
16 |
+
bar = px.bar(topics_prob, x='probability', y='topic')
|
17 |
+
st.plotly_chart(bar, use_container_width=True)
|
18 |
+
|
19 |
+
pie = px.pie(sentiment_prob,
|
20 |
+
values='probability',
|
21 |
+
names='sentiment',
|
22 |
+
title='Sentiment Probability',
|
23 |
+
color_discrete_map={'positive':'rgb(0, 204, 0)',
|
24 |
+
'negative':'rgb(215, 11, 11)'
|
25 |
+
},
|
26 |
+
color='sentiment'
|
27 |
+
)
|
28 |
+
st.plotly_chart(pie, use_container_width=True)
|
29 |
+
|
30 |
+
|
31 |
+
#st.success(f'Your text has been predicted to fall under the following topics: {result[:-1]}. The sentiment of this text is {result[-1]}.')
|
32 |
|
33 |
st.write("\n")
|
34 |
st.subheader('Or... Upload a csv file if you have a file instead.')
|
requirements.txt
CHANGED
@@ -4,4 +4,5 @@ transformers==4.16.1
|
|
4 |
scikit-learn
|
5 |
pandas==1.2.4
|
6 |
torch==1.10.1
|
7 |
-
numpy==1.19.5
|
|
|
|
4 |
scikit-learn
|
5 |
pandas==1.2.4
|
6 |
torch==1.10.1
|
7 |
+
numpy==1.19.5
|
8 |
+
plotly==5.1.0
|
utils.py
CHANGED
@@ -39,23 +39,16 @@ def get_single_prediction(text):
|
|
39 |
text_vectors = np.mean([w2v[i] for i in text.split()], axis=0)
|
40 |
|
41 |
# Make predictions
|
42 |
-
results = model.
|
43 |
-
|
44 |
|
45 |
# Get sentiment
|
46 |
-
|
47 |
-
|
48 |
-
|
|
|
49 |
|
50 |
-
|
51 |
-
pred_labels = [labels[idx] for idx, tag in enumerate(results) if tag == 1]
|
52 |
-
|
53 |
-
if len(pred_labels) == 0:
|
54 |
-
pred_labels.append('others')
|
55 |
-
|
56 |
-
pred_labels.append(sentiment)
|
57 |
-
|
58 |
-
return pred_labels
|
59 |
|
60 |
def get_multiple_predictions(csv):
|
61 |
|
|
|
39 |
text_vectors = np.mean([w2v[i] for i in text.split()], axis=0)
|
40 |
|
41 |
# Make predictions
|
42 |
+
results = model.predict_proba(text_vectors.reshape(1,300)).squeeze().round(2)
|
43 |
+
pred_prob = pd.DataFrame({'topic': labels, 'probability': results}).sort_values('probability', ascending=True)
|
44 |
|
45 |
# Get sentiment
|
46 |
+
sentiment_results = classifier(text,
|
47 |
+
candidate_labels=['positive', 'negative'],
|
48 |
+
hypothesis_template='The sentiment of this is {}')
|
49 |
+
sentiment_prob = pd.DataFrame({'sentiment': sentiment_results['labels'], 'probability': sentiment_results['scores']})
|
50 |
|
51 |
+
return (pred_prob, sentiment_prob)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
def get_multiple_predictions(csv):
|
54 |
|