deepseek-ai / DeepSeek-V3 - 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 38 units with 478 lines of code in units (66.9% of code).

0 very complex units (0 lines of code)
0 complex units (0 lines of code)
2 medium complex units (86 lines of code)
2 simple units (64 lines of code)
34 very simple units (328 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 38 most complex units

Unit	# lines	McCabe index	# params
def main() in inference/convert.py	37	14	4
def main() in inference/fp8_cast_bf16.py	49	11	2
def forward() in inference/model.py	40	6	5
def forward() in inference/model.py	24	6	2
def __init__() in inference/model.py	14	5	2
def forward() in inference/model.py	16	5	3
def linear() in inference/model.py	12	4	3
def __init__() in inference/model.py	15	4	5
def __init__() in inference/model.py	31	4	2
def forward() in inference/model.py	16	4	2
def weight_dequant() in inference/kernel.py	8	3	3
def fp8_gemm() in inference/kernel.py	10	3	4
def forward() in inference/model.py	10	3	2
def forward() in inference/model.py	7	3	2
def precompute_freqs_cis() in inference/model.py	28	3	1
def __init__() in inference/model.py	13	3	2
def fp8_gemm_kernel() in inference/kernel.py	33	2	11
def __init__() in inference/model.py	10	2	2
def __init__() in inference/model.py	6	2	3
def act_quant_kernel() in inference/kernel.py	9	1	4
def act_quant() in inference/kernel.py	8	1	2
def weight_dequant_kernel() in inference/kernel.py	12	1	6
def sample() in inference/generate.py	4	1	2
def generate() in inference/generate.py	6	1	0
def main() in inference/generate.py	7	1	0
def __init__() in inference/model.py	9	1	3
def forward() in inference/model.py	2	1	2
def __init__() in inference/model.py	4	1	5
def forward() in inference/model.py	3	1	2
def __init__() in inference/model.py	4	1	5
def __init__() in inference/model.py	5	1	3
def forward() in inference/model.py	2	1	2
def apply_rotary_emb() in inference/model.py	6	1	2
def __init__() in inference/model.py	5	1	3
def forward() in inference/model.py	2	1	2
def __init__() in inference/model.py	5	1	3
def forward() in inference/model.py	2	1	2
def forward() in inference/model.py	4	1	5