Update app.py

app.py

@@ -24,6 +24,11 @@ pred = transformers.pipeline("text-classification", model=model,
 
 explainer = shap.Explainer(pred)
 
+
+##
+classifier = transformers.pipeline("text-classification", model = "cross-encoder/qnli-electra-base")
+##
+
 def adr_predict(x):
     encoded_input = tokenizer(x, return_tensors='pt')
     output = model(**encoded_input)
@@ -32,14 +37,16 @@ def adr_predict(x):
    
     shap_values = explainer([str(x).lower()])
     local_plot = shap.plots.text(shap_values[0], display=False)
+
+    med = classifier(x+str("There is a medication."))[0]
    
-    return {"Severe Reaction": float(scores.numpy()[1]), "Non-severe Reaction": float(scores.numpy()[0])}, local_plot
+    return {"Severe Reaction": float(scores.numpy()[1]), "Non-severe Reaction": float(scores.numpy()[0])}, local_plot, med
 
 
 def main(prob1):
     text = str(prob1).lower()
     obj = adr_predict(text)
-    return obj[0],obj[1]
+    return obj[0],obj[1],obj[2]
 
 title = "Welcome to **ADR Detector** 🪐"
 description1 = """This app takes text (up to a few sentences) and predicts to what extent the text describes severe (or non-severe) adverse reaction to medicaitons."""
@@ -60,12 +67,12 @@ with gr.Blocks(title=title) as demo:
         main,
         [prob1],
         [label
-         ,local_plot
+         ,local_plot, med
         ], api_name="adr"
     )
 
     gr.Markdown("### Click on any of the examples below to see to what extent they contain resilience messaging:")
-    gr.Examples([["I have severe pain."],["I have minor pain."]], [prob1], [label,local_plot
+    gr.Examples([["I have severe pain."],["I have minor pain."]], [prob1], [label,local_plot, med
                                                                           ], main, cache_examples=True)
     
 demo.launch()