add charts

.ipynb_checkpoints/app-checkpoint.py

@@ -1,4 +1,5 @@
 import streamlit as st
+import plotly.express as px
 from utils import *
 
 ########## Title for the Web App ##########
@@ -10,9 +11,24 @@ feedback = st.text_input('Type your text here', 'The website was user friendly a
 if st.button('Click for predictions!'):
     with st.spinner('Generating predictions...'):
         
-        result = get_single_prediction(feedback)
-    
-    st.success(f'Your text has been predicted to fall under the following topics: {result[:-1]}. The sentiment of this text is {result[-1]}.')
+        topics_prob, sentiment_prob = get_single_prediction(feedback)
+        
+        bar = px.bar(topics_prob, x='probability', y='topic')
+        st.plotly_chart(bar, use_container_width=True)
+        
+        pie = px.pie(sentiment_prob, 
+               values='probability', 
+               names='sentiment', 
+               title='Sentiment Probability', 
+               color_discrete_map={'positive':'rgb(0, 204, 0)', 
+                                 'negative':'rgb(215, 11, 11)'
+                                  },
+               color='sentiment'
+              )
+        st.plotly_chart(pie, use_container_width=True)
+        
+        
+    #st.success(f'Your text has been predicted to fall under the following topics: {result[:-1]}. The sentiment of this text is {result[-1]}.')
 
 st.write("\n")    
 st.subheader('Or... Upload a csv file if you have a file instead.')

.ipynb_checkpoints/requirements-checkpoint.txt

@@ -4,4 +4,5 @@ transformers==4.16.1
 scikit-learn
 pandas==1.2.4
 torch==1.10.1
-numpy==1.19.5
+numpy==1.19.5
+plotly==5.1.0

.ipynb_checkpoints/utils-checkpoint.py

@@ -39,23 +39,16 @@ def get_single_prediction(text):
     text_vectors = np.mean([w2v[i] for i in text.split()], axis=0)
 
     # Make predictions
-    results = model.predict(text_vectors.reshape(1,300)).squeeze()
-    print(results)
+    results = model.predict_proba(text_vectors.reshape(1,300)).squeeze().round(2)
+    pred_prob = pd.DataFrame({'topic': labels, 'probability': results}).sort_values('probability', ascending=True)
     
     # Get sentiment
-    sentiment = get_sentiment_label_facebook(classifier(text, 
-                                                    candidate_labels=['positive', 'negative'], 
-                                                    hypothesis_template='The sentiment of this is {}'))
+    sentiment_results = classifier(text, 
+                                candidate_labels=['positive', 'negative'], 
+                                hypothesis_template='The sentiment of this is {}')
+    sentiment_prob = pd.DataFrame({'sentiment': sentiment_results['labels'], 'probability': sentiment_results['scores']})
     
-    # Consolidate results
-    pred_labels = [labels[idx] for idx, tag in enumerate(results) if tag == 1]
-    
-    if len(pred_labels) == 0:
-        pred_labels.append('others')
-    
-    pred_labels.append(sentiment)
-    
-    return pred_labels
+    return (pred_prob, sentiment_prob)
 
 def get_multiple_predictions(csv):
     

app.py

@@ -1,4 +1,5 @@
 import streamlit as st
+import plotly.express as px
 from utils import *
 
 ########## Title for the Web App ##########
@@ -10,9 +11,24 @@ feedback = st.text_input('Type your text here', 'The website was user friendly a
 if st.button('Click for predictions!'):
     with st.spinner('Generating predictions...'):
         
-        result = get_single_prediction(feedback)
-    
-    st.success(f'Your text has been predicted to fall under the following topics: {result[:-1]}. The sentiment of this text is {result[-1]}.')
+        topics_prob, sentiment_prob = get_single_prediction(feedback)
+        
+        bar = px.bar(topics_prob, x='probability', y='topic')
+        st.plotly_chart(bar, use_container_width=True)
+        
+        pie = px.pie(sentiment_prob, 
+               values='probability', 
+               names='sentiment', 
+               title='Sentiment Probability', 
+               color_discrete_map={'positive':'rgb(0, 204, 0)', 
+                                 'negative':'rgb(215, 11, 11)'
+                                  },
+               color='sentiment'
+              )
+        st.plotly_chart(pie, use_container_width=True)
+        
+        
+    #st.success(f'Your text has been predicted to fall under the following topics: {result[:-1]}. The sentiment of this text is {result[-1]}.')
 
 st.write("\n")    
 st.subheader('Or... Upload a csv file if you have a file instead.')

requirements.txt

@@ -4,4 +4,5 @@ transformers==4.16.1
 scikit-learn
 pandas==1.2.4
 torch==1.10.1
-numpy==1.19.5
+numpy==1.19.5
+plotly==5.1.0

utils.py

@@ -39,23 +39,16 @@ def get_single_prediction(text):
     text_vectors = np.mean([w2v[i] for i in text.split()], axis=0)
 
     # Make predictions
-    results = model.predict(text_vectors.reshape(1,300)).squeeze()
-    print(results)
+    results = model.predict_proba(text_vectors.reshape(1,300)).squeeze().round(2)
+    pred_prob = pd.DataFrame({'topic': labels, 'probability': results}).sort_values('probability', ascending=True)
     
     # Get sentiment
-    sentiment = get_sentiment_label_facebook(classifier(text, 
-                                                    candidate_labels=['positive', 'negative'], 
-                                                    hypothesis_template='The sentiment of this is {}'))
+    sentiment_results = classifier(text, 
+                                candidate_labels=['positive', 'negative'], 
+                                hypothesis_template='The sentiment of this is {}')
+    sentiment_prob = pd.DataFrame({'sentiment': sentiment_results['labels'], 'probability': sentiment_results['scores']})
     
-    # Consolidate results
-    pred_labels = [labels[idx] for idx, tag in enumerate(results) if tag == 1]
-    
-    if len(pred_labels) == 0:
-        pred_labels.append('others')
-    
-    pred_labels.append(sentiment)
-    
-    return pred_labels
+    return (pred_prob, sentiment_prob)
 
 def get_multiple_predictions(csv):