Code Changes - Reranking Implementation

app.py

@@ -7,18 +7,9 @@ from datetime import datetime
 import numpy as np
 import matplotlib
 
-# ==== Metrics Calculation
-from sklearn.metrics import roc_auc_score
-from sklearn.metrics import mean_squared_error
-
-# === HuggingFace & Transformers ===
-from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
-
-
 # === Misc ===
 import json
 import gradio as gr
-from json_repair import repair_json
 
 import datetime
 
@@ -114,54 +105,6 @@ def get_info_from_sample_questions(subset, question):
             return orig_ans, y_metrics
     return "No answer found.", "No metrics found."
 
-def plot_subset_metrics_old(subset_name):
-    summary = ragbench_details[subset_name]["summary"]
-
-    # Create a DataFrame for plotting
-    keys = ['Entries', 'TotalDocs', 'TotalUniqueIds', 'TotalUniqueDocs', 'UniqueDocsPercent']
-    values = [summary.get(k, 0) for k in keys]
-
-    fig, ax = plt.subplots(figsize=(8, 4))
-    bars = ax.bar(keys, values, color='skyblue')
-    plt.tight_layout()  # Apply tight layout after plotting
-
-
-    # ✅ Add count on top of bars
-    ax.bar_label(bars, fmt='%.0f', padding=3)
-    
-    # ax.bar(keys, values, color="mediumseagreen")
-    ax.set_title(f"Metrics for Subset: {subset_name}, Domain: {summary.get('Domain')}", fontsize=14)
-    ax.set_ylabel("Value")
-    ax.grid(axis='y')
-    ax.set_xticks(range(len(keys)))
-    ax.set_xticklabels(keys, rotation=21, ha='right')
-    return fig
-
-def plot_chunking_strategies_old(subset_name):
-    chunking_data = ragbench_details[subset_name]["chunking"]
-
-    plt.figure(figsize=(8, 4))
-    strategies = list(chunking_data.keys())
-    counts = list(chunking_data.values())
-
-    bars = plt.bar(strategies, counts, color="skyblue")
-
-    # Add value labels on top of bars
-    for bar in bars:
-        yval = bar.get_height()
-        plt.text(bar.get_x() + bar.get_width()/2, yval + 20, int(yval), ha='center', va='bottom', fontsize=10)
-
-    # plt.xlabel("Chunking Strategies")
-    plt.tight_layout()  # Apply tight layout after plotting
-
-    plt.ylabel("Number of Chunks")
-    plt.title(f"Chunking Strategy Distribution - {subset_name}")
-    plt.xticks(rotation=30)
-    plt.tight_layout()
-
-    # Return plot as figure (Gradio accepts it)
-    return plt.gcf()
-
 
 def plot_subset_metrics_old1(subset_name):
     """Generate a bar plot of key metrics for a given subset with proper title display."""
@@ -221,7 +164,6 @@ def plot_subset_metrics_old1(subset_name):
     except Exception as e:
         print(f"Plotting error: {str(e)}")
 
-
 def plot_chunking_strategies_old1(subset_name):
     """Visualize chunking strategy distribution with enhanced formatting."""
     try:
@@ -539,7 +481,7 @@ def generate_advance_report(subset_dropdown, dataset_type_dropdown, chunking_dro
     rmEmbedName = embed_dropdown.replace("/", ":")
     rmGenName = generator_dropdown.replace("/", ":")
     
-    fileName = f"{subset_dropdown}_{chunking_dropdown}_{rmEmbedName}_{rmGenName}_output_{datetime.datetime.now().strftime('%d-%B-%Y-%H-%M-%S')}.json"
+    fileName = f"{subset_dropdown}_{noOfQuestions}_{chunking_dropdown}_{rmEmbedName}_{rmGenName}_output_{datetime.datetime.now().strftime('%d-%B-%H-%M')}.json"
     # Save to file inside Space
     with open(fileName, "w") as f:
         f.write(json_str)