File size: 5,889 Bytes
8c10cf2 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 | #!/usr/bin/env python3
import pandas as pd
import numpy as np
from pathlib import Path
def aggregate_llm_share():
base_dir = Path(__file__).parent
series_order = [
(
"Email Responder",
["EmailResponder", "EmailResponder-MCP"],
),
(
"Recruitment",
[
"RecruitmentAssistant-A2A",
"RecruitmentAssistant-H_A2A",
"RecruitmentAssistant-MCP",
],
),
("Markdown Val.", ["MarkdownValidator", "MarkdownValidator-MCP"]),
("Game Builder", ["GameBuilder", "GameBuilder-MCP"]),
(
"SQL Asst.",
["SQLAssistant-A2A", "SQLAssistant-H_A2A", "SQLAssistant-MCP"],
),
(
"Landing Pg.",
[
"LandingPageGenerator-A2A",
"LandingPageGenerator-H_A2A",
"LandingPageGenerator-MCP",
],
),
(
"Book Writer",
[
"BookWriter-A2A",
"BookWriter-H_A2A",
"BookWriter-MCP",
],
),
(
"Social M. M.",
[
"SocialMediaManager-A2A",
"SocialMediaManager-H_A2A",
"SocialMediaManager-MCP",
],
),
]
projects = []
csv_files = []
project_to_display_name = {}
for series_name, project_list in series_order:
for project_name in project_list:
project_dir = base_dir / project_name
if project_dir.is_dir():
csv_file = project_dir / "performance_breakdown_summary_by_model.csv"
if csv_file.exists():
if "-A2A_mix" in project_name:
display_name = f"{series_name} (A2A_mix)"
elif "-A2A" in project_name:
display_name = f"{series_name} (A2A)"
elif "-MCP" in project_name:
display_name = f"{series_name} (MCP)"
else:
# Projects without a suffix are Pure CrewAI baselines
display_name = f"{series_name} (CrewAI)"
projects.append(project_name)
project_to_display_name[project_name] = display_name
csv_files.append(csv_file)
else:
print(f"Warning: {project_name} missing CSV file")
print(f"Found {len(projects)} projects with CSV files")
models = [
"GPT-5",
"GPT-4o-mini",
"DeepSeek-V3-1",
"DeepSeek-R1",
"Gemini-2.5-flash",
"Gemini-2.5-flash-nothinking",
"Qwen3-235b",
]
data_dict = {model: [] for model in models}
weight_dict = {model: [] for model in models}
total_weighted_sum = 0.0
total_weight = 0.0
for project, csv_file in zip(projects, csv_files):
try:
df = pd.read_csv(csv_file)
for model in models:
model_data = df[df["model"] == model]
if not model_data.empty:
llm_share = model_data["LLM_share"].values[0]
llm_share = min(llm_share, 1.0)
llm_share = round(llm_share, 4)
data_dict[model].append(llm_share)
comp_time = model_data["total_components_time"].values[0]
weight = float(comp_time) if pd.notna(comp_time) else 0.0
weight_dict[model].append(weight)
if pd.notna(llm_share) and weight > 0:
total_weighted_sum += llm_share * weight
total_weight += weight
else:
data_dict[model].append(None)
weight_dict[model].append(0.0)
except Exception as e:
print(f"Error reading {csv_file}: {e}")
for model in models:
data_dict[model].append(None)
weight_dict[model].append(0.0)
result_df = pd.DataFrame(data_dict, index=projects)
result_df.index = result_df.index.map(project_to_display_name)
result_df.index.name = "Model"
result_df = result_df.T
# Time-weighted overall average across all models and projects
overall_avg = (
round(total_weighted_sum / total_weight, 4) if total_weight > 0 else np.nan
)
overall_row_name = "Overall Average (time-weighted, all models & projects)"
overall_row = pd.Series(
{col: overall_avg for col in result_df.columns}, name=overall_row_name
)
result_df = pd.concat([result_df, overall_row.to_frame().T])
output_file = base_dir / "llm_share_summary.csv"
result_df.to_csv(output_file, float_format="%.4f")
print(f"\nCSV saved to: {output_file}")
print(f"\nShape: {result_df.shape[0]} models × {result_df.shape[1]} projects")
return result_df
def generate_latex_table(df):
num_cols = len(df.columns)
latex = []
latex.append("\\begin{table*}[htbp]")
latex.append("\\centering")
latex.append("\\small")
latex.append(f"\\begin{{tabular}}{{l{'c' * num_cols}}}")
latex.append("\\toprule")
header = "Model & " + " & ".join(df.columns) + " \\\\"
latex.append(header)
latex.append("\\midrule")
for idx, row in df.iterrows():
row_str = (
str(idx)
+ " & "
+ " & ".join([f"{v:.4f}" if pd.notna(v) else "-" for v in row])
+ " \\\\"
)
latex.append(row_str)
latex.append("\\bottomrule")
latex.append("\\end{tabular}")
latex.append("\\caption{LLM Share by Model and Project}")
latex.append("\\label{tab:llm_share}")
latex.append("\\end{table*}")
return "\n".join(latex)
if __name__ == "__main__":
aggregate_llm_share()
|