facebookresearch / FAMBench
File Size

The distribution of size of files (measured in lines of code).

Intro
  • File size measurements show the distribution of size of files.
  • Files are classified in four categories based on their size (lines of code): 1-100 (very small files), 101-200 (small files), 201-500 (medium size files), 501-1000 (long files), 1001+(very long files).
  • It is a good practice to keep files small. Long files may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.
Learn more...
File Size Overall
  • There are 108 files with 18,179 lines of code.
    • 3 very long files (4,324 lines of code)
    • 8 long files (5,801 lines of code)
    • 6 medium size files (2,064 lines of codeclsfd_ftr_w_mp_ins)
    • 20 small files (2,811 lines of code)
    • 71 very small files (3,179 lines of code)
23% | 31% | 11% | 15% | 17%
Legend:
1001+
501-1000
201-500
101-200
1-100


explore: zoomable circles | sunburst | 3D view
File Size per Extension
1001+
501-1000
201-500
101-200
1-100
py23% | 28% | 7% | 18% | 21%
cc39% | 26% | 24% | 3% | 4%
h0% | 69% | 0% | 13% | 17%
cu0% | 100% | 0% | 0% | 0%
cpp0% | 0% | 100% | 0% | 0%
yml0% | 0% | 0% | 100% | 0%
toml0% | 0% | 0% | 0% | 100%
yaml0% | 0% | 0% | 0% | 100%
File Size per Logical Decomposition
primary
1001+
501-1000
201-500
101-200
1-100
benchmarks/dlrm/ootb47% | 47% | 4% | 0% | 1%
benchmarks/rnnt/ootb/inference20% | 23% | 17% | 14% | 23%
benchmarks/dlrm/ootb/tools0% | 100% | 0% | 0% | 0%
benchmarks/cudnn_multihead_attn0% | 76% | 0% | 23% | 0%
benchmarks/rnnt/ootb/train0% | 0% | 24% | 39% | 35%
benchmarks/dlrm/ubench0% | 0% | 0% | 65% | 34%
fb5logging0% | 0% | 0% | 68% | 31%
benchmarks/xlmr/ootb0% | 0% | 0% | 54% | 45%
benchmarks/dlrm/ootb/tricks0% | 0% | 0% | 0% | 100%
bmlogging0% | 0% | 0% | 0% | 100%
benchmarks/dlrm/ootb/optim0% | 0% | 0% | 0% | 100%
benchmarks/dlrm/ootb/cython0% | 0% | 0% | 0% | 100%
Longest Files (Top 50)
File# lines# units
dlrm_s_pytorch.py
in benchmarks/dlrm/ootb
 1885 29
cc
loadgen.cc
in benchmarks/rnnt/ootb/inference/loadgen
 1281 30
dlrm_s_caffe2.py
in benchmarks/dlrm/ootb
 1158 26
dlrm_data_pytorch.py
in benchmarks/dlrm/ootb
 973 29
data_utils.py
in benchmarks/dlrm/ootb
 950 8
cc
logging.cc
in benchmarks/rnnt/ootb/inference/loadgen
 850 56
visualize.py
in benchmarks/dlrm/ootb/tools
 712 12
logging.h
in benchmarks/rnnt/ootb/inference/loadgen
 618 40
cu
cudnn_multihead_attn_benchmark.cu
in benchmarks/cudnn_multihead_attn
 599 -
dlrm_data_caffe2.py
in benchmarks/dlrm/ootb
 578 21
extend_distributed.py
in benchmarks/dlrm/ootb
 521 25
train.py
in benchmarks/rnnt/ootb/train
 492 4
cc
issue_query_controller.cc
in benchmarks/rnnt/ootb/inference/loadgen
 472 16
cc
python_api.cc
in benchmarks/rnnt/ootb/inference/loadgen/bindings
 325 17
data_loader_terabyte.py
in benchmarks/dlrm/ootb
 286 13
repro.cpp
in benchmarks/rnnt/ootb/inference/loadgen/benchmark
 258 19
inference.py
in benchmarks/rnnt/ootb/train
 231 4
multihead_attn_make_ref.py
in benchmarks/cudnn_multihead_attn
 189 3
dlrm_ubench_train_embeddingbag_driver.py
in benchmarks/dlrm/ubench
 173 2
features.py
in benchmarks/rnnt/ootb/train/common/data
 169 14
features.py
in benchmarks/rnnt/ootb/inference/pytorch/parts
 168 12
result_summarizer.py
in fb5logging
 161 14
pipeline.py
in benchmarks/rnnt/ootb/train/common/data/dali
 157 7
model.py
in benchmarks/rnnt/ootb/train/rnnt
 156 9
dataset.py
in benchmarks/rnnt/ootb/train/common/data
 153 12
helpers.py
in benchmarks/rnnt/ootb/train/common
 142 12
model_separable_rnnt.py
in benchmarks/rnnt/ootb/inference/pytorch
 138 9
optimizers.py
in benchmarks/rnnt/ootb/train/common
 136 7
audio.py
in benchmarks/rnnt/ootb/train/common
 128 18
dlrm_ubench_train_driver.py
in benchmarks/dlrm/ubench
 128 -
environment.yml
in benchmarks/rnnt/ootb/inference
 127 -
cc
c_api.cc
in benchmarks/rnnt/ootb/inference/loadgen/bindings
 127 13
tb_dllogger.py
in benchmarks/rnnt/ootb/train/common
 119 12
manifest.py
in benchmarks/rnnt/ootb/inference/pytorch/parts
 117 10
issue_query_controller.h
in benchmarks/rnnt/ootb/inference/loadgen
 115 5
xlmr.py
in benchmarks/xlmr/ootb
 107 6
segment.py
in benchmarks/rnnt/ootb/inference/pytorch/parts
 101 14
dataset.py
in benchmarks/rnnt/ootb/inference/pytorch
 100 7
dlrm_ubench_comms_driver.py
in benchmarks/dlrm/ubench
 96 2
pytorch_SUT.py
in benchmarks/rnnt/ootb/inference
 94 6
cc
utils.cc
in benchmarks/rnnt/ootb/inference/loadgen
 92 5
data_loader.py
in benchmarks/rnnt/ootb/train/common/data/dali
 79 8
rwsadagrad.py
in benchmarks/dlrm/ootb/optim
 78 3
version_generator.py
in benchmarks/rnnt/ootb/inference/loadgen
 76 7
run.py
in benchmarks/rnnt/ootb/inference
 74 2
qr_embedding_bag.py
in benchmarks/dlrm/ootb/tricks
 74 4
bmlogger.py
in bmlogging
 73 17
sampler.py
in benchmarks/rnnt/ootb/train/common/data/dali
 71 11
mlperf_logger.py
in benchmarks/dlrm/ootb
 70 8
decoder.py
in benchmarks/rnnt/ootb/train/rnnt
 68 5
Files With Most Units (Top 20)
File# lines# units
cc
logging.cc
in benchmarks/rnnt/ootb/inference/loadgen
 850 56
logging.h
in benchmarks/rnnt/ootb/inference/loadgen
 618 40
cc
loadgen.cc
in benchmarks/rnnt/ootb/inference/loadgen
 1281 30
dlrm_data_pytorch.py
in benchmarks/dlrm/ootb
 973 29
dlrm_s_pytorch.py
in benchmarks/dlrm/ootb
 1885 29
dlrm_s_caffe2.py
in benchmarks/dlrm/ootb
 1158 26
extend_distributed.py
in benchmarks/dlrm/ootb
 521 25
dlrm_data_caffe2.py
in benchmarks/dlrm/ootb
 578 21
repro.cpp
in benchmarks/rnnt/ootb/inference/loadgen/benchmark
 258 19
audio.py
in benchmarks/rnnt/ootb/train/common
 128 18
bmlogger.py
in bmlogging
 73 17
cc
python_api.cc
in benchmarks/rnnt/ootb/inference/loadgen/bindings
 325 17
cc
issue_query_controller.cc
in benchmarks/rnnt/ootb/inference/loadgen
 472 16
result_summarizer.py
in fb5logging
 161 14
features.py
in benchmarks/rnnt/ootb/train/common/data
 169 14
segment.py
in benchmarks/rnnt/ootb/inference/pytorch/parts
 101 14
cc
c_api.cc
in benchmarks/rnnt/ootb/inference/loadgen/bindings
 127 13
data_loader_terabyte.py
in benchmarks/dlrm/ootb
 286 13
tb_dllogger.py
in benchmarks/rnnt/ootb/train/common
 119 12
dataset.py
in benchmarks/rnnt/ootb/train/common/data
 153 12
Files With Long Lines (Top 18)

There are 18 files with lines longer than 120 characters. In total, there are 37 long lines.

File# lines# units# long lines
visualize.py
in benchmarks/dlrm/ootb/tools
 712 12 12
dlrm_ubench_train_driver.py
in benchmarks/dlrm/ubench
 128 - 5
xlmr.py
in benchmarks/xlmr/ootb
 107 6 2
run.py
in benchmarks/rnnt/ootb/inference
 74 2 2
data_utils.py
in benchmarks/dlrm/ootb
 950 8 2
dlrm_s_pytorch.py
in benchmarks/dlrm/ootb
 1885 29 2
xlmr_data.py
in benchmarks/xlmr/ootb
 39 4 1
symbols.py
in benchmarks/rnnt/ootb/train/common/text
 10 - 1
iterator.py
in benchmarks/rnnt/ootb/train/common/data/dali
 62 7 1
pipeline.py
in benchmarks/rnnt/ootb/train/common/data/dali
 157 7 1
cc
python_api.cc
in benchmarks/rnnt/ootb/inference/loadgen/bindings
 325 17 1
decoders.py
in benchmarks/rnnt/ootb/inference/pytorch
 61 6 1
model_separable_rnnt.py
in benchmarks/rnnt/ootb/inference/pytorch
 138 9 1
toml
rnnt.toml
in benchmarks/rnnt/ootb/inference/pytorch/configs
 65 - 1
accuracy_eval.py
in benchmarks/rnnt/ootb/inference
 43 2 1
rwsadagrad.py
in benchmarks/dlrm/ootb/optim
 78 3 1
extend_distributed.py
in benchmarks/dlrm/ootb
 521 25 1
dlrm_ubench_infer_driver.py
in benchmarks/dlrm/ubench
 63 - 1