azure / azure-llm-fine-tuning - 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 45 units with 1,152 lines of code in units (20.2% of code).

0 very complex units (0 lines of code)
1 complex units (52 lines of code)
5 medium complex units (368 lines of code)
8 simple units (279 lines of code)
31 very simple units (453 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 45 most complex units

Unit	# lines	McCabe index	# params
def validate_json() in aoai/token_count_utils.py	52	35	1
def main() in phi3/olive/phi3.py	35	16	1
def main() in phi3/src_train/train.py	135	14	1
def num_tokens_from_functions() in aoai/token_count_utils.py	40	13	2
def _num_token_from_messages() in aoai/token_count_utils.py	32	11	3
def main() in phi3/src_train/train_mlflow.py	126	11	1
def generate_config() in phi3/olive/phi3.py	43	10	1
def validate_jsonl() in aoai/token_count_utils.py	26	9	1
def train_model() in florence2-VQA/src_train/train_mlflow.py	90	9	3
def use_passes() in phi3/olive/phi3.py	17	8	2
def combine_tokenizers() in phi3/dataset-preparation/combine_tokenizer.py	21	8	1
def print_stats_tokens() in aoai/token_count_utils.py	20	6	2
def log_params_from_dict() in phi3/src_train/train_mlflow.py	10	6	3
def genai_run() in phi3/olive/phi3.py	52	6	3
def num_assistant_tokens_from_messages() in aoai/token_count_utils.py	11	5	2
def main() in phi3/dataset-preparation/train_tokenizer.py	34	5	0
def create_image_with_text() in florence2-VQA/src_train/train_mlflow.py	20	5	4
def count_token() in aoai/token_count_utils.py	7	4	4
def save_dataset_to_txt() in phi3/dataset-preparation/train_tokenizer.py	8	3	3
def load_model() in florence2-VQA/src_train/train_mlflow.py	17	3	2
def _num_token_from_text() in aoai/token_count_utils.py	7	2	2
def print_distribution() in aoai/token_count_utils.py	6	2	2
def init() in phi3/src_serve/score.py	11	2	0
def __getitem__() in florence2-VQA/src_train/train_mlflow.py	8	2	2
def run_example() in florence2-VQA/src_train/train_mlflow.py	14	2	3
def run_example_base64() in florence2-VQA/src_serve/score.py	17	2	4
def init() in florence2-VQA/src_serve/score.py	18	2	0
def get_max_token_limit() in aoai/token_count_utils.py	31	1	1
def percentile_used() in aoai/token_count_utils.py	2	1	2
def token_left() in aoai/token_count_utils.py	2	1	4
def load_model() in phi3/src_train/train.py	15	1	1
def apply_chat_template() in phi3/src_train/train.py	3	1	0
def parse_args() in phi3/src_train/train.py	28	1	0
def load_model() in phi3/src_train/train_mlflow.py	15	1	1
def apply_chat_template() in phi3/src_train/train_mlflow.py	3	1	0
def parse_args() in phi3/src_train/train_mlflow.py	27	1	0
def get_args() in phi3/olive/phi3.py	89	1	1
def main() in phi3/dataset-preparation/combine_tokenizer.py	7	1	0
def run() in phi3/src_serve/score.py	11	1	1
def __init__() in florence2-VQA/src_train/train_mlflow.py	2	1	2
def __len__() in florence2-VQA/src_train/train_mlflow.py	2	1	1
def collate_fn() in florence2-VQA/src_train/train_mlflow.py	4	1	2
def parse_args() in florence2-VQA/src_train/train_mlflow.py	18	1	0
def main() in florence2-VQA/src_train/train_mlflow.py	6	1	1
def run() in florence2-VQA/src_serve/score.py	10	1	1