AutoSeg/gen_wandb_table.py at master · githubbermoon/AutoSeg · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
import matplotlib.pyplot as plt
import matplotlib
import numpy as np

def main():
    # Data extracted from all W&B run summaries
    runs = [
        {"run_id": "bs5uq9j0", "date": "2025-12-17", "safety_score": 4.37, "safe_pct": 4.37, "hazard_pct": 20.08, "time_ms": 464.68, "confidence": 0.8773},
        {"run_id": "0ks3uxr2", "date": "2025-12-18", "safety_score": 16.25, "safe_pct": 16.25, "hazard_pct": 60.57, "time_ms": 958.00, "confidence": 0.8203},
        {"run_id": "lmc2hl19", "date": "2025-12-18", "safety_score": 16.25, "safe_pct": 16.25, "hazard_pct": 60.57, "time_ms": 950.94, "confidence": 0.8203},
        {"run_id": "hqp1d7rd", "date": "2025-12-18", "safety_score": 16.25, "safe_pct": 16.25, "hazard_pct": 60.57, "time_ms": 928.56, "confidence": 0.8203},
        {"run_id": "hidtc9wu", "date": "2025-12-19", "safety_score": 14.54, "safe_pct": 14.54, "hazard_pct": 61.05, "time_ms": 1107.99, "confidence": 0.9246},
        {"run_id": "yv7o9wuj", "date": "2026-02-05", "safety_score": 52.54, "safe_pct": 52.54, "hazard_pct": 4.03, "time_ms": 693.23, "confidence": 0.9029},
        {"run_id": "4n8ht60w", "date": "2026-02-19", "safety_score": 25.70, "safe_pct": 25.70, "hazard_pct": 50.24, "time_ms": 683.59, "confidence": 0.8085},
        {"run_id": "k7fkeu69", "date": "2026-02-23", "safety_score": 25.70, "safe_pct": 25.70, "hazard_pct": 50.24, "time_ms": 683.59, "confidence": 0.8085},
    ]

    # Table columns
    columns = ["#", "Run ID", "Date", "Safety\nScore (%)", "Safe\nPixels (%)", "Hazard\nPixels (%)", "Inference\nTime (ms)", "Mean\nConfidence"]

    # Table data
    table_data = []
    for i, r in enumerate(runs):
        table_data.append([
            str(i+1),
            r["run_id"],
            r["date"],
            f"{r['safety_score']:.2f}",
            f"{r['safe_pct']:.2f}",
            f"{r['hazard_pct']:.2f}",
            f"{r['time_ms']:.2f}",
            f"{r['confidence']:.4f}"
        ])

    # Create figure
    fig, ax = plt.subplots(figsize=(18, 8))
    ax.axis('off')
    fig.patch.set_facecolor('#FFFFFF')

    # Title
    ax.text(0.5, 0.97, "W&B Experiment Logging Results — Inference Metrics",
            ha='center', va='top', fontsize=18, fontweight='bold', color='#1a1a2e',
            transform=ax.transAxes)

    ax.text(0.5, 0.92, "Model: nvidia/segformer-b2-finetuned-ade-512-512  |  Device: CPU (Apple M4)  |  Dataset: ADE20K (150 classes)",
            ha='center', va='top', fontsize=11, color='#555555',
            transform=ax.transAxes, fontstyle='italic')

    # Create the table
    table = ax.table(
        cellText=table_data,
        colLabels=columns,
        loc='center',
        cellLoc='center',
        bbox=[0.02, 0.05, 0.96, 0.80]
    )

    # Style the table
    table.auto_set_font_size(False)
    table.set_fontsize(11)

    # Header styling
    for j in range(len(columns)):
        cell = table[0, j]
        cell.set_facecolor('#1565C0')
        cell.set_text_props(color='white', fontweight='bold', fontsize=11)
        cell.set_edgecolor('#0D47A1')
        cell.set_linewidth(1.5)
        cell.set_height(0.12)

    # Data row styling
    for i in range(len(table_data)):
        for j in range(len(columns)):
            cell = table[i+1, j]
            cell.set_edgecolor('#BBDEFB')
            cell.set_linewidth(0.8)
            cell.set_height(0.08)

            # Alternate row colors
            if i % 2 == 0:
                cell.set_facecolor('#F5F9FF')
            else:
                cell.set_facecolor('#FFFFFF')

            # Color-code Safety Score
            if j == 3:  # Safety Score column
                val = float(table_data[i][j])
                if val >= 40:
                    cell.set_text_props(color='#2E7D32', fontweight='bold')  # Green
                elif val >= 15:
                    cell.set_text_props(color='#E65100', fontweight='bold')  # Orange
                else:
                    cell.set_text_props(color='#C62828', fontweight='bold')  # Red

            # Color-code Hazard %
            if j == 5:  # Hazard column
                val = float(table_data[i][j])
                if val >= 50:
                    cell.set_text_props(color='#C62828', fontweight='bold')
                elif val >= 20:
                    cell.set_text_props(color='#E65100', fontweight='bold')
                else:
                    cell.set_text_props(color='#2E7D32', fontweight='bold')

            # Color-code Confidence
            if j == 7:  # Confidence column
                val = float(table_data[i][j])
                if val >= 0.90:
                    cell.set_text_props(color='#2E7D32', fontweight='bold')
                elif val >= 0.80:
                    cell.set_text_props(color='#1565C0', fontweight='bold')

            # Highlight best run row (yv7o9wuj with 52.54%)
            if table_data[i][1] == "yv7o9wuj":
                cell.set_facecolor('#E8F5E9')
                cell.set_edgecolor('#4CAF50')

    # Set column widths
    col_widths = [0.04, 0.10, 0.10, 0.10, 0.10, 0.10, 0.12, 0.10]
    for j, w in enumerate(col_widths):
        for i in range(len(table_data) + 1):
            table[i, j].set_width(w)

    # Summary stats at bottom
    scores = [r["safety_score"] for r in runs]
    times = [r["time_ms"] for r in runs]
    confs = [r["confidence"] for r in runs]

    summary = (
        f"Summary:  Avg Safety Score: {np.mean(scores):.2f}%  |  "
        f"Avg Inference Time: {np.mean(times):.2f}ms  |  "
        f"Avg Confidence: {np.mean(confs):.4f}  |  "
        f"Best Score: {max(scores):.2f}% (Run yv7o9wuj)  |  "
        f"Total Runs: {len(runs)}"
    )
    ax.text(0.5, 0.01, summary,
            ha='center', va='bottom', fontsize=10, color='#333333',
            transform=ax.transAxes, fontstyle='italic',
            bbox=dict(boxstyle='round,pad=0.4', fc='#FFF3E0', ec='#FF9800', lw=1))

    plt.savefig('wandb_results_table.png', dpi=250, bbox_inches='tight', facecolor='#FFFFFF')
    plt.close()
    print("Successfully generated wandb_results_table.png!")

if __name__ == "__main__":
    main()