Update app.py

app.py

@@ -5,16 +5,18 @@ from transformers import AutoModelForSequenceClassification,AutoTokenizer, AutoC
 import numpy as np
 #convert logits to probabilities
 from scipy.special import softmax
+from transformers import pipeline
 
 
 
 
 #import the model
-tokenizer = AutoTokenizer.from_pretrained('Junr-syl/sentiments_analysis_Roberta')
+pipe=pipeline(model="Junr-syl/sentiments_analysis_DISTILBERT")
+# tokenizer = AutoTokenizer.from_pretrained('Junr-syl/sentiments_analysis_DISTILBERT')
 
-model_path = f"Junr-syl/sentiments_analysis_Roberta"
-config = AutoConfig.from_pretrained(model_path)
-model = AutoModelForSequenceClassification.from_pretrained(model_path)
+# model_path = f"Junr-syl/sentiments_analysis_DISTILBERT"
+# config = AutoConfig.from_pretrained(model_path)
+# model = AutoModelForSequenceClassification.from_pretrained(model_path)
 
 #Set the page configs
 st.set_page_config(page_title='Sentiments Analysis',page_icon='😎',layout='centered')
@@ -34,58 +36,61 @@ col1.title('Sentiment Emoji')
 col2.title('How this user feels about the vaccine')
 col3.title('Confidence of this prediction')
 
-if submit:
-    print('submitted')
-    #pass text to preprocessor
-    def preprocess(text):
-    #initiate an empty list 
-        new_text = []
-        #split text by space
-        for t in text.split(" "):
-            #set username to @user
-            t = '@user' if t.startswith('@') and len(t) > 1 else t  
-            #set tweet source to http
-            t = 'http' if t.startswith('http') else t 
-            #store text in the list
-            new_text.append(t)
-            #change text from list back to string
-        return " ".join(new_text) 
+# if submit:
+#     print('submitted')
+#     #pass text to preprocessor
+#     def preprocess(text):
+#     #initiate an empty list 
+#         new_text = []
+#         #split text by space
+#         for t in text.split(" "):
+#             #set username to @user
+#             t = '@user' if t.startswith('@') and len(t) > 1 else t  
+#             #set tweet source to http
+#             t = 'http' if t.startswith('http') else t 
+#             #store text in the list
+#             new_text.append(t)
+#             #change text from list back to string
+#         return " ".join(new_text) 
     
 
-    #pass text to model
+#     #pass text to model
+output=pipe(text)
+output_dict=output[0]
+lable=output_dict['label']
+score=output_dict['score']
+#     #change label id 
+#     #config.id2label = {0: 'NEGATIVE', 1: 'NEUTRAL', 2: 'POSITIVE'}
 
-    #change label id 
-    config.id2label = {0: 'NEGATIVE', 1: 'NEUTRAL', 2: 'POSITIVE'}
+#     text = preprocess(text)
 
-    text = preprocess(text)
+#     # PyTorch-based models
+#     encoded_input = tokenizer(text, return_tensors='pt')
+#     output = model(**encoded_input)
+#     scores = output[0][0].detach().numpy()
+#     scores = softmax(scores)
 
-    # PyTorch-based models
-    encoded_input = tokenizer(text, return_tensors='pt')
-    output = model(**encoded_input)
-    scores = output[0][0].detach().numpy()
-    scores = softmax(scores)
-
-    #Process scores
-    ranking = np.argsort(scores)
-    ranking = ranking[::-1]  
-    l = config.id2label[ranking[0]]
-    s = scores[ranking[0]]
+#     #Process scores
+#     ranking = np.argsort(scores)
+#     ranking = ranking[::-1]  
+#     l = config.id2label[ranking[0]]
+#     s = scores[ranking[0]]
 
     #output
-    if l=='NEGATIVE':
+    if lable=='NEGATIVE':
         with col1:
             com.iframe("https://embed.lottiefiles.com/animation/125694")
         col2.write('NEGATIVE')
-        col3.write(f'{s*100:.2f}%')
-    elif l=='POSITIVE':
+        col3.write(f'{score*100:.2f}%')
+    elif lable=='POSITIVE':
         with col1:
             com.iframe("https://embed.lottiefiles.com/animation/148485")
         col2.write('POSITIVE')
-        col3.write(f'{s*100:.2f}%')
+        col3.write(f'{score*100:.2f}%')
     else:
         with col1:
             com.iframe("https://embed.lottiefiles.com/animation/136052")
         col2.write('NEUTRAL')
-        col3.write(f'{s*100:.2f}%')
+        col3.write(f'{score*100:.2f}%')