Update app/main.py

app/main.py

@@ -15,6 +15,16 @@ models = {
     
 }
 
+model_descriptions = {
+    "fBert Convabuse": "This is the model fBert, trained on the conversational abuse public dataset. It is a binary classification model that predicts whether a given text is abusive or not. The model is based on the fBert architecture and was trained using the Sentence Transformers library.",
+    "fBert HTDM": "This is the model fBert, trained on the hate speech public dataset. It is a binary classification model that predicts whether a given text is hate speech or not. The model is based on the fBert architecture and was trained using the Sentence Transformers library.",
+    "hateBert Convabuse": "This is the model hateBert, trained on the conversational abuse public dataset. It is a binary classification model that predicts whether a given text is abusive or not. The model is based on the hateBert architecture and was trained using the Sentence Transformers library.",
+    "hateBert HTDM": "This is the model hateBert, trained on the hate speech public dataset. It is a binary classification model that predicts whether a given text is hate speech or not. The model is based on the hateBert architecture and was trained using the Sentence Transformers library.",
+    "berTweet Convabuse": "This is the model berTweet, trained on the conversational abuse public dataset. It is a binary classification model that predicts whether a given text is abusive or not. The model is based on the berTweet architecture and was trained using the Sentence Transformers library.",
+    "berTweet HTDM": "This is the model berTweet, trained on the hate speech public dataset. It is a binary classification model that predicts whether a given text is hate speech or not. The model is based on the berTweet architecture and was trained using the Sentence Transformers library.",
+    "roberta Convabuse": "This is the model roberta, trained on the conversational abuse public dataset. It is a binary classification model that predicts whether a given text is abusive or not. The model is based on the roberta architecture and was trained using the Sentence Transformers library.",
+    "roberta HTDM": "This is the model roberta, trained on the hate speech public dataset. It is a binary classification model that predicts whether a given text is hate speech or not. The model is based on the roberta architecture and was trained using the Sentence Transformers library.",
+}
 
 
 def classify_text(text, model):
@@ -33,6 +43,9 @@ user_input = st.text_input("")
 
 
 selected_model = st.selectbox("Select Model", options=list(models.keys()))
+selected_model_description = model_descriptions[selected_model]
+st.write("Model Description:")
+st.write(selected_model_description)
 
 label_map ={
     "LABEL_0": "Not Homotransphobic",
@@ -46,6 +59,7 @@ if st.button('Classify'):
     prediction_raw_text = st.empty()
     prediction_text = st.empty()
     
+    
     loading_text = st.text("Predicting... (if the model has not been used before, this may take a while)")
     # Classify the text
     prediction = classify_text(user_input, models[selected_model])