Update app.py

app.py

@@ -8,60 +8,60 @@ model_dir = 'temp'
 tokenizer = RobertaTokenizer.from_pretrained(model_dir)
 model = RobertaModel.from_pretrained(model_dir)
 
-@app.route('/gradio_app')
-def gradio_app():
-    def process_text(input_text, input_file):
-        if input_text:
-            text = input_text
-        elif input_file is not None:
-            text = input_file.read().decode('utf-8')
-        inputs = tokenizer(text, return_tensors="pt")
-        with torch.no_grad():
-            logits = model(**inputs).logits
-        probs = softmax(logits, dim=1)
-        max_prob, predicted_class_id = torch.max(probs, dim=1)
-        prob = str(round(max_prob.item() * 100, 2))
-        label = model.config.id2label[predicted_class_id.item()]
-        final_label='Human' if model.config.id2label[predicted_class_id.item()]=='LABEL_0' else 'Chat-GPT'
-        processed_result = text
 
-        def search(text):
-            query = text
-            api_key = 'AIzaSyClvkiiJTZrCJ8BLqUY9I38WYmbve8g-c8'
-            search_engine_id = '53d064810efa44ce7'
-            url = f'https://www.googleapis.com/customsearch/v1?key={api_key}&cx={search_engine_id}&q={query}'
 
-            try:
-                response = requests.get(url)
-                data = response.json()
-                return data
-            except Exception as e:
-                return {'error': str(e)}
+def process_text(input_text, input_file):
+    if input_text:
+        text = input_text
+    elif input_file is not None:
+        text = input_file.read().decode('utf-8')
+    inputs = tokenizer(text, return_tensors="pt")
+    with torch.no_grad():
+        logits = model(**inputs).logits
+    probs = softmax(logits, dim=1)
+    max_prob, predicted_class_id = torch.max(probs, dim=1)
+    prob = str(round(max_prob.item() * 100, 2))
+    label = model.config.id2label[predicted_class_id.item()]
+    final_label='Human' if model.config.id2label[predicted_class_id.item()]=='LABEL_0' else 'Chat-GPT'
+    processed_result = text
+    def search(text):
+        query = text
+        api_key = 'AIzaSyClvkiiJTZrCJ8BLqUY9I38WYmbve8g-c8'
+        search_engine_id = '53d064810efa44ce7'
+        url = f'https://www.googleapis.com/customsearch/v1?key={api_key}&cx={search_engine_id}&q={query}'
 
-        def find_plagiarism(text):
-            search_results = search(text)
-            if 'items' not in search_results:
-                return []
-            similar_articles = []
-            for item in search_results['items']:
-                title = item.get('title', '')
-                link = item.get('link', '')
-                similar_articles.append({'title': title, 'link': link})
-            return similar_articles[:5]
+        try:
+            response = requests.get(url)
+            data = response.json()
+            return data
+        except Exception as e:
+            return {'error': str(e)}
+    def find_plagiarism(text):
+        search_results = search(text)
+        if 'items' not in search_results:
+            return []
+        similar_articles = []
+        for item in search_results['items']:
+            title = item.get('title', '')
+            link = item.get('link', '')
+            similar_articles.append({'title': title, 'link': link})
+        return similar_articles[:5]
 
-        pipe = pipeline('text-classification', model=model, tokenizer=tokenizer)
-        prediction = pipe([text])
-        explainer = shap.Explainer(pipe)
-        shap_values = explainer([text])
-        text_plot = shap.plots.text(shap_values, display=True)
-        similar_articles = find_plagiarism(text)
+    pipe = pipeline('text-classification', model=model, tokenizer=tokenizer)
+    prediction = pipe([text])
+    explainer = shap.Explainer(pipe)
+    shap_values = explainer([text])
+    text_plot = shap.plots.text(shap_values, display=True)
+    similar_articles = find_plagiarism(text)
 
-        return processed_result, prob, final_label, text_plot,similar_articles
+    return processed_result, prob, final_label, text_plot,similar_articles
 
     text_input = gr.inputs.Textbox(label="Enter text")
     file_input = gr.inputs.File(label="Upload a text file")
     outputs = [gr.Textbox(label="Processed text"), gr.Textbox(label="Probability"), gr.Textbox(label="Label"), gr.HTML(label="SHAP Plot"),gr.Table(label="Similar Articles", columns=["Title", "Link"])]
-
+title = "Group 2- ChatGPT text detection module"
+description = '''Please upload text files and text input responsibly and await the explainable results. The approach in place includes finetuning a Roberta model for text classification.Once the classifications are done the decision is exaplined thorugh the SHAP text plot.
+The probability is particularly explained by the attention plots through SHAP'''
     gr.Interface(fn=process_text, inputs=[text_input, file_input], outputs=outputs).launch()
 
     return ''