implemented histogram

app.py

@@ -136,27 +136,47 @@ def process_file(model_name,inc_slider,progress=Progress(track_tqdm=True)):
         # Helper function to evaluate task attempts
         def evaluate_tasks(fields, tasks):
             task_status = {}
+            total_attempts = 0  # Counter for total number of attempts
+            attempted_tasks = 0  # Counter for tasks attempted at least once
+            successful_first_attempts = 0  # Counter for tasks successful on the first try
             for task in tasks:
+                
                 relevant_attempts = [f for f in fields if task in f]
+                attempt_count = len(relevant_attempts)
+                total_attempts += attempt_count  # Add to the total attempts
+
+                if attempt_count > 0:
+                    attempted_tasks += 1  # Increment attempted tasks count
+
+                    # Check the first attempt
+                    first_attempt = relevant_attempts[0]
+                    if "OK" in first_attempt and "ERROR" not in first_attempt and "JIT" not in first_attempt:
+                        successful_first_attempts += 1
+
+
                 if any("OK" in attempt for attempt in relevant_attempts):
                     task_status[task] = "Attempted (Successful)"
+                    
                 elif any("ERROR" in attempt for attempt in relevant_attempts):
                     task_status[task] = "Attempted (Error)"
                 elif any("JIT" in attempt for attempt in relevant_attempts):
                     task_status[task] = "Attempted (JIT)"
                 else:
                     task_status[task] = "Unattempted"
-            return task_status
+            return task_status,attempted_tasks, total_attempts,successful_first_attempts
 
         # Evaluate tasks for each category
-        optional_task_1_status = evaluate_tasks(fields, optional_task_1_subtasks)
-        optional_task_2_status = evaluate_tasks(fields, optional_task_2_subtasks)
+        optional_task_1_status, opt1_attempted, opt1_total_attempts, opt1_successful_first_attempts  = evaluate_tasks(fields, optional_task_1_subtasks)
+        optional_task_2_status, opt2_attempted, opt2_total_attempts, opt2_successful_first_attempts  = evaluate_tasks(fields, optional_task_2_subtasks)
 
         # Check if tasks have any successful attempt
         opt1_done = any(status == "Attempted (Successful)" for status in optional_task_1_status.values())
         opt2_done = any(status == "Attempted (Successful)" for status in optional_task_2_status.values())
 
-        return opt1_done, opt2_done
+        return (opt1_done, opt2_done,
+                opt1_attempted, opt2_attempted, 
+                opt1_total_attempts, opt2_total_attempts,
+                opt1_successful_first_attempts, opt2_successful_first_attempts)
 
     # Read data from test_info.txt
     with open(test_info_location, "r") as file:
@@ -170,16 +190,40 @@ def process_file(model_name,inc_slider,progress=Progress(track_tqdm=True)):
     1: {"ER": 0, "ME": 0, "both": 0,"none":0},
     2: {"ER": 0, "ME": 0, "both": 0,"none":0}
     }
-
+      # To store task completion counts per row
     # Analyze rows
+    row_summary = []
+    opt1_ratios = []
+    opt2_ratios = []
     for i, row in enumerate(data):
         row = row.strip()
         if not row:
             continue
 
         ideal_task = ideal_tasks[i]  # Get the ideal task for the current row
-        opt1_done, opt2_done = analyze_row(row)
-
+        (
+            opt1_done, opt2_done,
+            opt1_attempted, opt2_attempted, 
+            opt1_total_attempts, opt2_total_attempts,
+            opt1_successful_first_attempts, opt2_successful_first_attempts
+        ) = analyze_row(row)
+
+        opt1_ratios.append(opt1_attempted / opt1_total_attempts if opt1_total_attempts > 0 else 0)
+        opt2_ratios.append(opt2_attempted / opt2_total_attempts if opt2_total_attempts > 0 else 0)
+    # create a summey for task:attempted, total attempts and succesful attempts for each row
+    #     row_summary.append({
+    #     "row_index": i + 1,
+    #     "opt1": {
+    #         "tasks_attempted": opt1_attempted,
+    #         "total_attempts": opt1_total_attempts,
+    #         "successful_attempts": opt1_successful_first_attempts,
+    #     },
+    #     "opt2": {
+    #         "tasks_attempted": opt2_attempted,
+    #         "total_attempts": opt2_total_attempts,
+    #         "successful_attempts": opt2_successful_first_attempts,
+    #     }
+    # })
         if ideal_task == 0:
             if opt1_done and not opt2_done:
                 task_counts[1]["ER"] += 1
@@ -198,6 +242,16 @@ def process_file(model_name,inc_slider,progress=Progress(track_tqdm=True)):
                 task_counts[2]["both"] += 1
             else:
                 task_counts[2]["none"] +=1
+    # Print a summary of task completions
+    # for summary in row_summary:
+    #     print(f"\nRow {summary['row_index']}:")
+    #     print(f"  OptionalTask_1 - Tasks Attempted: {summary['opt1']['tasks_attempted']}, "
+    #         f"Total Attempts: {summary['opt1']['total_attempts']}, "
+    #         f"Successful Attempts: {summary['opt1']['successful_attempts']}")
+    #     print(f"  OptionalTask_2 - Tasks Attempted: {summary['opt2']['tasks_attempted']}, "
+    #         f"Total Attempts: {summary['opt2']['total_attempts']}, "
+    #         f"Successful Attempts: {summary['opt2']['successful_attempts']}")
+
 
     # Create a string output for results
     # output_summary = "Task Analysis Summary:\n"
@@ -209,9 +263,90 @@ def process_file(model_name,inc_slider,progress=Progress(track_tqdm=True)):
     #     output_summary += f"  Only OptionalTask_2 done: {counts['ME']}\n"
     #     output_summary += f"  Both done: {counts['both']}\n"
 
+
+    # Create figure
+    fig_hist1 = go.Figure()
+
+    # Add histogram for OptionalTask_1 (ER)
+    fig_hist1.add_trace(go.Histogram(
+        x=opt1_ratios,
+        name="ER",
+        marker=dict(color='blue'),
+        opacity=1,
+        xbins=dict(
+            start=0.0,
+            end=1.0,
+            size=0.1  # Bin width set to 0.1 for 10 bins
+        )
+    ))
+    fig_hist2=go.Figure()
+    # Add histogram for OptionalTask_2 (ME)
+    fig_hist2.add_trace(go.Histogram(
+        x=opt2_ratios,
+        name="ME",
+        marker=dict(color='red'),
+        opacity=1,
+        xbins=dict(
+            start=0.0,
+            end=1.0,
+            size=0.1  # Bin width set to 0.1 for 10 bins
+        )
+    ))
+
+    # Update layout
+    fig_hist1.update_layout(
+        title="ER: Histogram of Task Success Ratios Across Dataset",
+        title_x=0.5,
+        xaxis=dict(
+            title="Success Ratio (Tasks Attempted / Total Attempts)",
+            tickmode="array",
+            tickvals=np.linspace(0, 1, 11),  # 10 evenly spaced ticks
+        ),
+        yaxis=dict(
+            title="Number of Instances"
+        ),
+        font=dict(
+            family="sans-serif",
+            size=12,
+            color="black"
+        ),
+        legend=dict(
+            font=dict(
+                family="sans-serif",
+                size=12,
+                color="black"
+            )
+        ),
+        barmode='overlay'  # Overlapping bars
+    )
+    fig_hist2.update_layout(
+        title="ME: Histogram of Task Success Ratios Across Dataset",
+        title_x=0.5,
+        xaxis=dict(
+            title="Success Ratio (Tasks Attempted  / Total Attempts)",
+            tickmode="array",
+            tickvals=np.linspace(0, 1, 11),  # 10 evenly spaced ticks
+        ),
+        yaxis=dict(
+            title="Number of Instances"
+        ),
+        font=dict(
+            family="sans-serif",
+            size=12,
+            color="black"
+        ),
+        legend=dict(
+            font=dict(
+                family="sans-serif",
+                size=12,
+                color="black"
+            )
+        ),
+        barmode='overlay'  # Overlapping bars
+    )
     # colors = ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728']
     colors = ["#FF6F61", "#6B5B95", "#88B04B", "#F7CAC9"]
-
+    # print(opt1_ratios,opt2_ratios)
   # Generate pie chart for Task 1
     task1_labels = list(task_counts[1].keys())
     task1_values = list(task_counts[1].values())
@@ -419,7 +554,7 @@ def process_file(model_name,inc_slider,progress=Progress(track_tqdm=True)):
     ROC-AUC for problems of type ER: {opt_task1_roc_auc:.4f}
     ROC-AUC for problems of type ME: {opt_task2_roc_auc:.4f}
     """
-    return text_output,fig,fig_task1,fig_task2
+    return text_output,fig,fig_task1,fig_task2,fig_hist1,fig_hist2
 
 # List of models for the dropdown menu
 
@@ -818,13 +953,17 @@ with gr.Blocks(theme='gstaff/sketch', css=custom_css) as demo:
         opt1_pie = gr.Plot(label="ER")
         opt2_pie = gr.Plot(label="ME")
         # output_summary = gr.Textbox(label="Summary")
+    with gr.Row():
+        histo1 = gr.Plot(label="Hist")
+        histo2 = gr.Plot(label="Hist")
+  
 
     
  
     btn.click(
         fn=process_file, 
         inputs=[model_dropdown,increment_slider], 
-        outputs=[output_text,plot_output,opt1_pie,opt2_pie]
+        outputs=[output_text,plot_output,opt1_pie,opt2_pie,histo1,histo2]
     )
 
 

result.txt

@@ -3,5 +3,5 @@ total_acc: 69.00702106318957
 precisions: 0.7236623191454734
 recalls: 0.6900702106318957
 f1_scores: 0.6802420656474512
-time_taken_from_start: 53.13972353935242
+time_taken_from_start: 23.562122583389282
 auc_score: 0.7457100293916334