apache / singa - Unit Size

Intro

Unit size measurements show the distribution of size of units of code (methods, functions...).
Units are classified in four categories based on their size (lines of code): 1-20 (small units), 20-50 (medium size units), 51-100 (long units), 101+ (very long units).
You should aim at keeping units small (< 20 lines). Long units may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.

Learn more...

Unit Size Overall

There are 2,008 units with 20,586 lines of code in units (52.9% of code).

6 very long units (689 lines of code)
44 long units (2,951 lines of code)
189 medium size units (5,872 lines of code)
343 small units (5,049 lines of code)
1,426 very small units (6,025 lines of code)

Legend:

101+

51-100

21-50

11-20

1-10

Unit Size per Extension

101+

51-100

21-50

11-20

1-10

Unit Size per Logical Component

primary logical decomposition

101+

51-100

21-50

11-20

1-10

Alternative Visuals

3D view of all units

Longest Units

Top 50 longest units

Unit	# lines	McCabe index	# params
inline float half2float_impl() in include/half.hpp	159	3	3
void Graph::Debug() in src/core/scheduler/scheduler.cc	114	27	0
Tensor resize() in src/io/image_transformer.cc	108	40	4
def tensordot() in python/singa/tensor.py	104	34	3
void CudnnConvolution::InitCudnn() in src/model/layer/cudnn_convolution.cc	102	11	1
def einsum() in python/singa/tensor.py	102	19	2
template unsigned int gamma() in include/half.hpp	91	31	1
const vector CudnnRNN::Forward() in src/model/layer/cudnn_rnn.cc	91	13	2
void NetworkThread::onRecv() in src/io/network/endpoint.cc	85	18	1
void Graph::AddOperation() in src/core/scheduler/scheduler.cc	85	19	4
def run() in python/singa/sonnx.py	83	35	3
const Tensor BatchNorm::Forward() in src/model/layer/batchnorm.cc	82	6	2
template unsigned int float2half_impl() in include/half.hpp	80	11	2
int tinydir_readfile() in include/singa/utils/tinydir.h	78	15	2
Tensor mirror() in src/io/image_transformer.cc	78	24	4
def initialize() in python/singa/layer.py	77	10	2
void Communicator::fusedSynch() in src/io/communicator.cc	76	7	2
const Tensor CudnnBatchNorm::Forward() in src/model/layer/cudnn_batchnorm.cc	75	8	2
void Communicator::fusedSynchHalf() in src/io/communicator.cc	72	5	2
void Graph::AnalyzeNodes() in src/core/scheduler/scheduler.cc	72	19	0
Tensor CpuConvForward() in src/model/operation/convolution.cc	71	8	4
Tensor crop() in src/io/image_transformer.cc	69	13	6
vector GpuRNNBackwardx() in src/model/operation/rnn.cc	69	3	8
void RepeatDataToFrom() in src/core/tensor/tensor.cc	69	16	6
def __init__() in python/singa/layer.py	69	19	12
def backward_and_sparse_update() in python/singa/opt.py	68	20	6
void Communicator::valSparsAllReduce() in src/io/communicator.cc	67	7	3
void Convolution::Setup() in src/model/layer/convolution.cc	67	12	2
void Mult() in src/core/tensor/tensor.cc	67	11	5
template unsigned int mod() in include/half.hpp	65	23	3
vector GpuRNNBackwardxEx() in src/model/operation/rnn.cc	64	5	9
def prepare_extension_options() in setup.py	63	6	0
inline half fma() in include/half.hpp	62	33	3
void Communicator::topKSparsAllReduce() in src/io/communicator.cc	61	5	3
const std::vector CpuBatchNormBackwardx() in src/model/operation/batchnorm.cc	61	1	8
Tensor CpuConvBackwardW() in src/model/operation/convolution.cc	60	5	4
vector GpuRNNForwardTraining() in src/model/operation/rnn.cc	60	4	5
int tinydir_file_open() in include/singa/utils/tinydir.h	59	11	2
def main() in tool/release/release.py	59	15	1
void NetworkThread::afterConnEst() in src/io/network/endpoint.cc	58	16	3
inline void sincos() in include/half.hpp	56	13	3
vector GpuRNNForwardInference() in src/model/operation/rnn.cc	56	4	5
def backward() in python/singa/autograd.py	56	18	2
inline half hypot() in include/half.hpp	55	18	3
def repeat() in python/singa/tensor.py	54	2	3
Tensor CpuConvBackwardx() in src/model/operation/convolution.cc	53	6	4
void Pooling::Setup() in src/model/layer/pooling.cc	53	9	2
int NetworkThread::asyncSend() in src/io/network/endpoint.cc	52	13	1
vector GpuRNNForwardTrainingEx() in src/model/operation/rnn.cc	52	5	6
vector GpuRNNForwardInferenceEx() in src/model/operation/rnn.cc	51	5	6