(benchmark_results, cl_packets, config)
| 572 | |
| 573 | |
| 574 | def generate_all_charts(benchmark_results, cl_packets, config): |
| 575 | fig = plt.figure(figsize=(18, 12), facecolor="white") |
| 576 | fig.suptitle("Control Layer — Production Benchmark Results", |
| 577 | fontsize=15, fontweight="bold", color=CHART_COLOR_DARK, y=0.98) |
| 578 | gs = gridspec.GridSpec(2, 3, figure=fig, hspace=0.42, wspace=0.38, |
| 579 | left=0.06, right=0.97, top=0.92, bottom=0.08) |
| 580 | axes = [fig.add_subplot(gs[r, c]) for r in range(2) for c in range(3)] |
| 581 | |
| 582 | chart_1_pass_rate(benchmark_results["naive_pass_rate"], |
| 583 | benchmark_results["cl_pass_rate"], axes[0]) |
| 584 | chart_2_failure_dist(benchmark_results["cl_audit"], axes[1]) |
| 585 | chart_3_retry_dist(benchmark_results["cl_attempts_dist"], axes[2]) |
| 586 | chart_4_latency(benchmark_results["naive_latencies"], |
| 587 | benchmark_results["cl_latencies"], axes[3]) |
| 588 | chart_5_token_budget(config, axes[4]) |
| 589 | chart_6_quality_scores(cl_packets, axes[5]) |
| 590 | |
| 591 | plt.savefig("control_layer_benchmark.png", dpi=150, |
| 592 | bbox_inches="tight", facecolor="white") |
| 593 | print("\n Benchmark chart saved: control_layer_benchmark.png") |
| 594 | plt.show() |
| 595 | |
| 596 | |
| 597 | # ============================================================================= |
no test coverage detected