Update app-backup.py

app-backup.py

@@ -79,20 +79,19 @@ def predict(title, abstract):
         # 1) Load config
         config = AutoConfig.from_pretrained(model_path)
         
-        # 2) Remove quantization_config if it exists to avoid `NoneType` error in PEFT
-        #    This ensures that 'quantization_config.to_dict()' won't be called
+        # 2) Remove quantization_config if it exists (avoid NoneType error in PEFT)
         if hasattr(config, "quantization_config"):
             del config.quantization_config
 
-        # 3) (Optional) We can still set config.num_labels = 1 if needed
+        # 3) Optionally set number of labels
         config.num_labels = 1
 
         # 4) Load the model
         model_loaded = AutoModelForSequenceClassification.from_pretrained(
             model_path,
-            config=config,            # pass config
-            torch_dtype=torch.float32,
-            device_map=None,          # manual device
+            config=config,
+            torch_dtype=torch.float32,  # float32 for stable cublasLt
+            device_map=None,
             low_cpu_mem_usage=False
         )
         model_loaded.to(device)
@@ -104,7 +103,6 @@ def predict(title, abstract):
         # Assign to globals
         model, tokenizer = model_loaded, tokenizer_loaded
 
-    # Construct the input text prompt
     text = (
         f"Given a certain paper,\n"
         f"Title: {title.strip()}\n"
@@ -129,15 +127,23 @@ def predict(title, abstract):
 # Grading
 ##################################################
 def get_grade_and_emoji(score):
-    """Map a 0–1 score to an A/B/C style grade with an emoji."""
-    if score >= 0.900: return "AAA 🌟"
-    if score >= 0.800: return "AA ⭐"
-    if score >= 0.650: return "A ✨"
-    if score >= 0.600: return "BBB 🔵"
-    if score >= 0.550: return "BB 📘"
-    if score >= 0.500: return "B 📖"
-    if score >= 0.400: return "CCC 📝"
-    if score >= 0.300: return "CC ✏️"
+    """Map a 0–1 score to an A/B/C style grade with an emoji indicator."""
+    if score >= 0.900:
+        return "AAA 🌟"
+    if score >= 0.800:
+        return "AA ⭐"
+    if score >= 0.650:
+        return "A ✨"
+    if score >= 0.600:
+        return "BBB 🔵"
+    if score >= 0.550:
+        return "BB 📘"
+    if score >= 0.500:
+        return "B 📖"
+    if score >= 0.400:
+        return "CCC 📝"
+    if score >= 0.300:
+        return "CC ✏️"
     return "C 📑"
 
 ##################################################
@@ -266,7 +272,6 @@ with gr.Blocks(theme=gr.themes.Default(), css=css) as iface:
     gr.Markdown("<div class='main-title'>Papers Impact: AI-Powered Research Impact Predictor</div>")
     gr.Markdown("**Predict the potential research impact (0–1) from title & abstract.**")
 
-    # Row with input column + output column
     with gr.Row():
         with gr.Column(elem_classes="input-section"):
             gr.Markdown("### Import from arXiv")
@@ -277,6 +282,14 @@ with gr.Blocks(theme=gr.themes.Default(), css=css) as iface:
                     label="arXiv URL or ID",
                     value="2504.11651"
                 )
+                gr.Markdown(
+                    """
+                    <p>
+                      Enter an arXiv ID or URL. For example: 
+                      <code>2504.11651</code> or <code>https://arxiv.org/pdf/2504.11651</code>
+                    </p>
+                    """
+                )
                 fetch_btn = gr.Button("🔍 Fetch Paper Details", variant="secondary")
 
             gr.Markdown("### Or Enter Manually")
@@ -297,6 +310,49 @@ with gr.Blocks(theme=gr.themes.Default(), css=css) as iface:
             score_box = gr.Number(label="Impact Score")
             grade_box = gr.Textbox(label="Grade", elem_classes="grade-display")
 
+    ############## METHODOLOGY EXPLANATION ##############
+    gr.Markdown(
+        """
+        ### Scientific Methodology
+        - **Training Data**: Model trained on an extensive dataset of published papers in CS.CV, CS.CL, CS.AI
+        - **Optimization**: NDCG optimization with Sigmoid activation and MSE loss
+        - **Validation**: Cross-validated against historical citation data
+        - **Architecture**: Advanced transformer-based (LLaMA derivative) textual encoder
+        - **Metrics**: Quantitative analysis of citation patterns and research influence
+        """
+    )
+
+    ############## RATING SCALE ##############
+    gr.Markdown(
+        """
+        ### Rating Scale
+        | Grade | Score Range | Description         | Emoji |
+        |-------|-------------|---------------------|-------|
+        | AAA   | 0.900–1.000 | **Exceptional**     | 🌟    |
+        | AA    | 0.800–0.899 | **Very High**       | ⭐    |
+        | A     | 0.650–0.799 | **High**            | ✨    |
+        | BBB   | 0.600–0.649 | **Above Average**   | 🔵    |
+        | BB    | 0.550–0.599 | **Moderate**        | 📘    |
+        | B     | 0.500–0.549 | **Average**         | 📖    |
+        | CCC   | 0.400–0.499 | **Below Average**   | 📝    |
+        | CC    | 0.300–0.399 | **Low**             | ✏️    |
+        | C     | <0.300      | **Limited**         | 📑    |
+        """
+    )
+
+    ############## EXAMPLE PAPERS ##############
+    gr.Markdown("### Example Papers")
+    for paper in example_papers:
+        gr.Markdown(
+            f"**{paper['title']}**  \n"
+            f"Score: {paper['score']} | Grade: {get_grade_and_emoji(paper['score'])}  \n"
+            f"{paper['abstract']}  \n"
+            f"*{paper['note']}*\n---"
+        )
+
+    ##################################################
+    # Events
+    ##################################################
     # Validation triggers
     title_input.change(update_button_status, [title_input, abs_input], [status_box, predict_btn])
     abs_input.change(update_button_status, [title_input, abs_input], [status_box, predict_btn])
@@ -311,16 +367,6 @@ with gr.Blocks(theme=gr.themes.Default(), css=css) as iface:
 
     predict_btn.click(run_predict, [title_input, abs_input], [score_box, grade_box])
 
-    # Example papers
-    gr.Markdown("### Example Papers")
-    for paper in example_papers:
-        gr.Markdown(
-            f"**{paper['title']}**  \n"
-            f"Score: {paper['score']} | Grade: {get_grade_and_emoji(paper['score'])}  \n"
-            f"{paper['abstract']}  \n"
-            f"*{paper['note']}*\n---"
-        )
-
 ##################################################
 # Launch
 ##################################################