Update app.py

app.py

@@ -28,6 +28,14 @@ from spacy import displacy
 import streamlit as st
 import spacy_streamlit
 nlp = spacy.load('en_core_web_sm')
+import torch
+import tensorflow as tf
+from transformers import RobertaTokenizer, RobertaModel
+from transformers import AutoModelForSequenceClassification
+from transformers import TFAutoModelForSequenceClassification
+from transformers import AutoTokenizer
+tokenizer = AutoTokenizer.from_pretrained("cardiffnlp/twitter-roberta-base-sentiment")
+model = AutoModelForSequenceClassification.from_pretrained("paragon-analytics/bert_resil")
 
 kw_extractor = yake.KeywordExtractor()
 custom_kw_extractor = yake.KeywordExtractor(lan="en", n=2, dedupLim=0.2, top=10, features=None)
@@ -52,6 +60,11 @@ def process_final_text(text):
     lstm_prob = lmodel.predict(test_sequences_matrix.tolist()).flatten()
     lstm_pred = np.where(lstm_prob>=0.5,1,0)
     
+    encoded_input = tokenizer(X_test, return_tensors='pt')
+    output = model(**encoded_input)
+    scores = output[0][0].detach().numpy()
+    scores = tf.nn.softmax(scores)
+    
     # Get Keywords:
     keywords = custom_kw_extractor.extract_keywords(X_test)
     letter = []
@@ -80,7 +93,7 @@ def process_final_text(text):
         + sp_html
         + ""
     )
-    return {"Resilience": float(lstm_prob[0]), "Non-Resilience": 1-float(lstm_prob[0])},keywords,NER
+    return {"Resilience": float(scores.numpy()[1]), "Non-Resilience": 1-float(scores.numpy()[0])},keywords,NER
     
 def main(prob1):
     text = str(prob1)
@@ -117,6 +130,6 @@ with gr.Blocks(title=title) as demo:
     
     
     gr.Markdown("### Click on any of the examples below to see how it works:")
-    gr.Examples([["It is difficult to write persuasive product descriptions."],["Talking to your friends about their problems with drugs and alcohol might not be easy."],["When experiencing depression, I couldn't get out of bed or focus."],["Up to 6 million homeless animals enter shelters nationwide every year."]], [prob1], [label,impplot,NER], main, cache_examples=True)
+    gr.Examples([["Please stay at home and avoid unnecessary trips."],["Please stay at home and avoid unnecessary trips. We will survive this."],["We will survive this."]], [prob1], [label,impplot,NER], main, cache_examples=True)
     
 demo.launch()