facebookresearch / param
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 41 files with 7,015 lines of code.
    • 1 very long files (1,039 lines of code)
    • 4 long files (2,750 lines of code)
    • 5 medium size files (1,257 lines of codeclsfd_ftr_w_mp_ins)
    • 6 small files (870 lines of code)
    • 25 very small files (1,099 lines of code)
14% | 39% | 17% | 12% | 15%
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
py14% | 39% | 17% | 12% | 15%
File Size per Logical Decomposition
primary
1001+
501-1000
201-500
101-200
1-100
train/comms/pt25% | 66% | 0% | 7% | 0%
train/compute/python/lib/pytorch0% | 0% | 35% | 17% | 46%
train/compute/python/workloads/pytorch0% | 0% | 78% | 0% | 21%
train/compute/pt0% | 0% | 33% | 23% | 42%
inference/compute/pt0% | 0% | 100% | 0% | 0%
train/compute/python/lib0% | 0% | 42% | 19% | 38%
train/compute/python/pytorch0% | 0% | 0% | 68% | 31%
train/compute/python/tools0% | 0% | 0% | 0% | 100%
train/compute/python0% | 0% | 0% | 0% | 100%
train/compute/python/workloads0% | 0% | 0% | 0% | 100%
Longest Files (Top 41)
File# lines# units
comms.py
in train/comms/pt
 1039 22
dlrm.py
in train/comms/pt
 880 40
comms_utils.py
in train/comms/pt
 798 64
commsTraceReplay.py
in train/comms/pt
 559 16
pytorch_dist_backend.py
in train/comms/pt
 513 45
build_executor.py
in train/compute/python/lib/pytorch
 273 14
split_table_batched_embeddings_ops.py
in train/compute/python/workloads/pytorch
 270 11
pytorch_linear.py
in train/compute/pt
 243 8
pytorch_linear.py
in inference/compute/pt
 242 8
iterator.py
in train/compute/python/lib
 229 16
dlrm_data.py
in train/comms/pt
 190 13
pytorch_emb.py
in train/compute/pt
 170 8
pytorch_tpu_backend.py
in train/comms/pt
 135 27
data_impl.py
in train/compute/python/lib/pytorch
 135 5
run_benchmark.py
in train/compute/python/pytorch
 135 1
config.py
in train/compute/python/lib
 105 20
op_executor.py
in train/compute/python/lib/pytorch
 100 6
pytorch_gemm.py
in train/compute/pt
 97 5
config_util.py
in train/compute/python/lib/pytorch
 96 8
generator.py
in train/compute/python/lib
 94 18
dataset.py
in train/compute/pt
 85 -
pytorch_cutlass.py
in train/compute/pt
 80 4
operator_impl.py
in train/compute/python/lib/pytorch
 77 15
run_batch.py
in train/compute/python/pytorch
 61 1
benchmark.py
in train/compute/python/lib/pytorch
 60 4
run_ncu.py
in train/compute/python/tools
 58 2
init_helper.py
in train/compute/python/lib
 49 4
driver.py
in train/compute/pt
 48 -
operator.py
in train/compute/python/lib
 41 9
alex_net.py
in train/compute/python/workloads/pytorch
 39 2
native_basic_ops.py
in train/compute/python/workloads/pytorch
 34 -
setup.py
in train/compute/python
 30 1
data.py
in train/compute/python/lib
 25 4
timer.py
in train/compute/python/lib/pytorch
 18 4
__init__.py
in train/compute/python/tools
 1 -
__init__.py
in train/compute/python
 1 -
__init__.py
in train/compute/python/workloads
 1 -
__init__.py
in train/compute/python/workloads/pytorch
 1 -
__init__.py
in train/compute/python/lib
 1 -
__init__.py
in train/compute/python/lib/pytorch
 1 -
__init__.py
in train/compute/python/pytorch
 1 -
Files With Most Units (Top 20)
File# lines# units
comms_utils.py
in train/comms/pt
 798 64
pytorch_dist_backend.py
in train/comms/pt
 513 45
dlrm.py
in train/comms/pt
 880 40
pytorch_tpu_backend.py
in train/comms/pt
 135 27
comms.py
in train/comms/pt
 1039 22
config.py
in train/compute/python/lib
 105 20
generator.py
in train/compute/python/lib
 94 18
commsTraceReplay.py
in train/comms/pt
 559 16
iterator.py
in train/compute/python/lib
 229 16
operator_impl.py
in train/compute/python/lib/pytorch
 77 15
build_executor.py
in train/compute/python/lib/pytorch
 273 14
dlrm_data.py
in train/comms/pt
 190 13
split_table_batched_embeddings_ops.py
in train/compute/python/workloads/pytorch
 270 11
operator.py
in train/compute/python/lib
 41 9
config_util.py
in train/compute/python/lib/pytorch
 96 8
pytorch_emb.py
in train/compute/pt
 170 8
pytorch_linear.py
in train/compute/pt
 243 8
pytorch_linear.py
in inference/compute/pt
 242 8
op_executor.py
in train/compute/python/lib/pytorch
 100 6
data_impl.py
in train/compute/python/lib/pytorch
 135 5
Files With Long Lines (Top 7)

There are 7 files with lines longer than 120 characters. In total, there are 53 long lines.

File# lines# units# long lines
dlrm.py
in train/comms/pt
 880 40 32
commsTraceReplay.py
in train/comms/pt
 559 16 9
comms_utils.py
in train/comms/pt
 798 64 5
comms.py
in train/comms/pt
 1039 22 4
pytorch_dist_backend.py
in train/comms/pt
 513 45 1
driver.py
in train/compute/pt
 48 - 1
pytorch_linear.py
in inference/compute/pt
 242 8 1