openai / evals - Conditional Complexity

Intro

Conditional complexity (also called cyclomatic complexity) is a term used to measure the complexity of software. The term refers to the number of possible paths through a program function. A higher value ofter means higher maintenance and testing costs (infosecinstitute.com).
Conditional complexity is calculated by counting all conditions in the program that can affect the execution path (e.g. if statement, loops, switches, and/or operators, try and catch blocks...).
Conditional complexity is measured at the unit level (methods, functions...).
Units are classified in four categories based on the measured McCabe index: 1-5 (simple units), 6-10 (medium complex units), 11-25 (complex units), 26+ (very complex units).

Learn more...

Conditional Complexity Overall

There are 1,775 units with 17,123 lines of code in units (40.8% of code).

0 very complex units (0 lines of code)
3 complex units (205 lines of code)
54 medium complex units (2,292 lines of code)
134 simple units (3,101 lines of code)
1,584 very simple units (11,525 lines of code)

Legend:

51+

26-50

11-25

6-10

1-5

Alternative Visuals

3D view of all units

Conditional Complexity per Extension

51+

26-50

11-25

6-10

1-5

Conditional Complexity per Logical Component

primary logical decomposition

51+

26-50

11-25

6-10

1-5

Most Complex Units

Top 50 most complex units

Unit	# lines	McCabe index	# params
def solve_diagram() in evals/registry/data/simple_physics_engine/solver.py	77	31	1
def plot_rewards() in evals/elsuite/incontext_rl/scripts/plot_experiments.py	56	28	5
def create_id_based_action() in evals/elsuite/multistep_web_tasks/webarena/browser_env/actions.py	72	26	1
def browser_is_equivalent() in evals/elsuite/multistep_web_tasks/webarena/browser_env/actions.py	48	25	2
def run() in evals/cli/oaieval.py	98	25	2
def action2create_function() in evals/elsuite/multistep_web_tasks/webarena/browser_env/actions.py	63	24	1
def plot_results() in evals/elsuite/bugged_tools/scripts/plot_experiments.py	73	23	3
def solve() in evals/registry/data/solve-for-variable/tools/solve.py	51	23	3
def _process_msgs() in evals/solvers/providers/together/together_solver.py	18	20	3
def run() in evals/elsuite/bluff/eval.py	61	20	2
def retrieve_viewport_info() in evals/elsuite/multistep_web_tasks/webarena/browser_env/processors.py	56	19	2
def _get_sample_std() in evals/elsuite/function_deduction/eval.py	32	19	2
def get_all_winning_hands() in evals/elsuite/bluff/bluff/cards.py	20	18	1
def make_bar_plot() in evals/elsuite/already_said_that/scripts/make_plots.py	67	18	3
def __post_init__() in evals/elsuite/make_me_say/core.py	38	18	1
def test() in evals/registry/data/solve-for-variable/tools/tester.py	41	18	1
def run() in evals/elsuite/error_recovery/eval.py	33	17	2
def dfs() in evals/elsuite/multistep_web_tasks/webarena/browser_env/processors.py	60	17	3
def eval_sample() in evals/elsuite/ballots/eval.py	87	17	3
def get_stockfish_move() in evals/elsuite/cant_do_that_anymore/scripts/diagonal_dataset_creation.py	25	17	2
def _initialize_task_env() in evals/elsuite/hr_ml_agent_bench/environment.py	39	17	1
def extract_results() in evals/elsuite/incontext_rl/scripts/plot_experiments.py	48	17	1
def run() in evals/elsuite/make_me_pay/eval.py	49	16	2
def run() in evals/elsuite/self_prompting/eval.py	65	16	2
def execute() in evals/elsuite/hr_ml_agent_bench/environment.py	62	16	3
def make_bar_plot() in evals/elsuite/track_the_stat/scripts/make_plots.py	71	16	4
def current_viewport_html() in evals/elsuite/multistep_web_tasks/webarena/browser_env/processors.py	52	14	2
def forward() in evals/elsuite/hr_ml_agent_bench/benchmarks/vectorization/env/train.py	65	14	2
def _log_additional_metrics() in evals/elsuite/bugged_tools/eval.py	23	13	3
def precision_recall_fscore() in evals/elsuite/bugged_tools/utils.py	12	13	1
def count_tokens() in evals/elsuite/identifying_variables/scripts/make_plots.py	58	13	2
def _compute_agg_metrics() in evals/elsuite/identifying_variables/eval.py	19	13	2
def eval_sample() in evals/elsuite/incontext_rl/eval.py	70	13	4
def run() in evals/elsuite/skill_acquisition/eval.py	46	13	2
def _get_complexity_tests() in evals/elsuite/function_deduction/eval.py	13	13	2
def calculate_probabilities() in evals/registry/data/poker_analysis/poker_analysis_sample_generator.py	27	13	2
def get_bishop_diagonal_moves() in evals/elsuite/cant_do_that_anymore/scripts/diagonal_dataset_creation.py	25	12	2
def determine_sample_type() in evals/elsuite/identifying_variables/renderers/corrset.py	24	12	2
def _compute_grouped_metrics() in evals/elsuite/identifying_variables/eval.py	57	12	2
def execute_script() in evals/elsuite/hr_ml_agent_bench/low_level_actions.py	45	12	3
def process_miskito_page() in evals/elsuite/skill_acquisition/scraping/scrape_miskito.py	31	12	0
def run() in evals/elsuite/twenty_questions/eval.py	28	12	2
def record_event() in evals/record.py	24	12	4
def _cut_messages_to_fit() in evals/elsuite/multistep_web_tasks/solvers/strong_solver/strong_solver.py	37	11	2
def parse_playwright_code() in evals/elsuite/multistep_web_tasks/webarena/browser_env/actions.py	40	11	1
def make_plot() in evals/elsuite/theory_of_mind/scripts/make_plots.py	50	11	2
def get_model() in evals/elsuite/already_said_that/scripts/make_plots.py	19	11	1
def _solver_used_bugged_input() in evals/elsuite/bugged_tools/eval.py	17	11	3
def _evaluate_sample() in evals/elsuite/identifying_variables/eval.py	35	11	4
def num_tokens_from_messages() in evals/elsuite/steganography/scripts/dataset/utils.py	28	11	2