Code Changes - Advanced Options

app.py

@@ -158,6 +158,47 @@ def plot_chunking_strategies(subset_name):
 initial_subset = available_subsets[0] if available_subsets else None
 initial_plot = plot_subset_metrics(initial_subset) if initial_subset else "No data available"
 
+
+def generate_advance_report(subset_dropdown, dataset_type_dropdown, chunking_dropdown, embed_dropdown, retriever_dropdown,
+                            chunk_count, retriever_type,
+                            reranking_checkbox, evaluator_dropdown)
+    
+    export_data = {
+            "metadata": {
+                "timestamp": datetime.datetime.now().isoformat(),
+                "format_version": "1.0"
+            },
+            "subset": {
+                "subset": subset_dropdown,
+                "dataset_type": dataset_type_dropdown,
+            },
+            "model_details": {
+                "strategy": chunking_dropdown,
+                "embed_model": embed_dropdown,
+                "generator_model": retriever_dropdown,
+                "chunk_count": chunk_count,
+                "noOfQuestions": noOfQuestions,
+                "retriever_type": retriever_type,
+                "reranking": reranking_checkbox,
+                "evaluator_model": evaluator_dropdown
+            }
+        }
+    
+    export_data["questions"] = []
+    # for sq in advanced_analysis[subset_dropdown][QUESTIONS]:
+    #     export['questions'].append(
+
+    # Step 2: Convert to JSON string for display
+    json_str = json.dumps(export_data, indent=2)
+    
+    fileName = f"{subset_dropdown}_{chunking_dropdown}_{embed_dropdown}_output_{datetime.datetime.now().strftime('%d-%B-%Y-%H-%M-%S')}.json"
+    # Save to file inside Space
+    with open(fileName, "w") as f:
+        f.write(json_str)
+
+    return json_str, fileName
+
+
 def generate_file(subset_dropdown, dataset_type_dropdown, 
                         chunking_dropdown, embed_dropdown, retriever_dropdown,
                         chunk_count, retriever_type,
@@ -474,7 +515,7 @@ with gr.Blocks(
 
     with gr.Row():
         # Add export button
-        with gr.Column(scale=2):
+        with gr.Column(scale=3):
             with gr.Accordion("Advanced Options", open=False):
                 noOfQuestions = gr.Slider(
                     minimum=1,
@@ -486,6 +527,9 @@ with gr.Blocks(
                 
                 # Create button with initial label
                 evaluate_btn = gr.Button("🚀 Run RAG Evaluation for 5 Questions", variant="primary")
+
+                # Create button with initial label
+                evaluate_report = gr.Button("🚀 Generated report", variant="primary")
                 
                 # Function to update button label
                 def update_button_label(value):
@@ -509,13 +553,14 @@ with gr.Blocks(
                     outputs=[evaluator_json_output]
                 )
 
+
+
         generate_btn = gr.Button("Generate JSON & Download")
         with gr.Column(scale=2):
             json_output = gr.Code(label="JSON Output", max_lines=50, language="json")
             download_file = gr.File(label="Download Link")
         
 
-
     submit_btn.click(
         fn=run_rag_pipeline,
         inputs=[
@@ -527,8 +572,6 @@ with gr.Blocks(
         outputs=[gen_ans_display, y_pred_metrics_display, chunks_retrieved_display, evaluator_json_output, download_file]
     )
 
-
-
     generate_btn.click(
         fn=generate_file,
         inputs=[subset_dropdown, dataset_type_dropdown, 
@@ -543,5 +586,10 @@ with gr.Blocks(
             outputs=[json_output, download_file]
         )
 
-
+    evaluate_report.click(
+        fn=generate_file,
+        inputs=[
+                ],
+            outputs=[json_output, download_file]
+        )
 demo.launch(debug=True, prevent_thread_lock=True, share=True)

ragbench.py

@@ -138,9 +138,9 @@ class RAGSystem:
                 repo_type=HF_REPO_TYPE,
             )
 
-            print(f"Repository {HF_DATASET_REPO_NAME} is accessible. Files:")
-            for index, item in enumerate(repo_files):
-                print(f"Index, {index}, File: {item}")
+            print(f"Repository {HF_DATASET_REPO_NAME} is accessible. No of Files: {len(repo_files)}")
+            # for index, item in enumerate(repo_files):
+            #     print(f"Index, {index}, File: {item}")
 
         except Exception as e:
             print(f"Error accessing Hugging Face repo: {e}")