Update app.py

app.py

@@ -15,23 +15,28 @@ def load_and_analyze_csv(file, text_field, event_model):
     if text_field not in df.columns:
         raise gr.Error(f"Error: Enter text column'{text_field}' not in CSV file.")
 
-    floods, fires, nones, scores = [], [], [], []
+    # floods, fires, nones, scores = [], [], [], []
+    labels, scores = [], []
     for post in df[text_field].to_list():
         res = classify(post, event_model, HFTOKEN)
-        if res["event"] == 'flood':
-            floods.append(post)
-        elif res["event"] == 'fire':
-            fires.append(post)
-        else:
-            nones.append(post)
+        # if res["event"] == 'flood':
+        #     floods.append(post)
+        # elif res["event"] == 'fire':
+        #     fires.append(post)
+        # else:
+        #     nones.append(post)
+        labels.append(res["event"])
         scores.append(res["score"])
 
+    df["model_label"] = labels
+    df["model_score"] = scores
+    
     model_confidence = round(mean(scores), 5)   
-    fire_related = gr.CheckboxGroup(choices=fires)
-    flood_related = gr.CheckboxGroup(choices=floods)
-    not_related = gr.CheckboxGroup(choices=nones)
+    fire_related = gr.CheckboxGroup(choices=df[df["model_label"]=="fire"][text_field].to_list()) #fires
+    flood_related = gr.CheckboxGroup(choices=df[df["model_label"]=="flood"][text_field].to_list())
+    not_related = gr.CheckboxGroup(choices=df[df["model_label"]=="none"][text_field].to_list())
     
-    return flood_related, fire_related, not_related, model_confidence, len(df[text_field].to_list())
+    return flood_related, fire_related, not_related, model_confidence, len(df[text_field].to_list()), df, df.columns.tolist()
 
 def analyze_selected_texts(selections):
     selected_texts = selections
@@ -105,14 +110,16 @@ with gr.Blocks() as demo:
                 gr.Markdown("\n\n\n")
                 model_confidence = gr.Number(label="Model Confidence")
             with gr.Column(scale=5):
-                correct = gr.Number(label="Number of correct classifications") #, value=0
+                correct = gr.Number(label="Number of correct classifications")
                 incorrect = gr.Number(label="Number of incorrect classifications")
                 accuracy = gr.Number(label="Model Accuracy (%)")
 
         accuracy_button = gr.Button("Calculate Accuracy")
         num_posts = gr.Number(visible=False)
+        data = gr.DataFrame(headers=datacols) #visible=False
+        
         predict_button.click(load_and_analyze_csv, inputs=[file_input, text_field, event_model], 
-                         outputs=[flood_checkbox_output, fire_checkbox_output, none_checkbox_output, model_confidence, num_posts])
+                         outputs=[flood_checkbox_output, fire_checkbox_output, none_checkbox_output, model_confidence, num_posts, data, datacols])
         accuracy_button.click(calculate_accuracy, inputs=[flood_checkbox_output, fire_checkbox_output, none_checkbox_output, num_posts], 
                              outputs=[incorrect, correct, accuracy])