openai/mle-bench
File Size
The distribution of size of files (measured in lines of code).
File Size Overall
  • There are 382 files with 14,775 lines of code.
    • 0 very long files (0 lines of code)
    • 0 long files (0 lines of code)
    • 8 medium size files (1,947 lines of code)
    • 22 small files (3,050 lines of code)
    • 352 very small files (9,778 lines of code)
0% | 0% | 13% | 20% | 66%
Legend:
1001+
501-1000
201-500
101-200
1-100


explore: grouped by folders | grouped by size | sunburst | 3D view
File Size per Extension
1001+
501-1000
201-500
101-200
1-100
py0% | 0% | 15% | 23% | 60%
yaml0% | 0% | 0% | 5% | 94%
mjs0% | 0% | 0% | 0% | 100%
File Size per Logical Decomposition
primary
1001+
501-1000
201-500
101-200
1-100
mlebench0% | 0% | 8% | 20% | 71%
experiments0% | 0% | 80% | 0% | 19%
extras0% | 0% | 35% | 36% | 28%
ROOT0% | 0% | 94% | 0% | 5%
agents0% | 0% | 0% | 47% | 52%
environment0% | 0% | 0% | 0% | 100%
Longest Files (Top 50)
File# lines# units
prepare.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 324 1
282 12
data.py
in mlebench
 275 14
analyze.py
in extras/rule_violation_detector
 232 3
mAP_evaluation.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 215 18
familiarity.py
in experiments/familiarity
 213 5
203 -
prepare.py
in mlebench/competitions/inaturalist-2019-fgvc6
 203 2
cli.py
in mlebench
 200 1
utils.py
in mlebench
 190 21
prepare.py
in mlebench/competitions/herbarium-2021-fgvc8
 176 1
prepare.py
in mlebench/competitions/herbarium-2020-fgvc7
 176 1
grade_helpers.py
in mlebench
 171 7
utils.py
in mlebench/competitions
 167 9
prepare.py
in mlebench/competitions/herbarium-2022-fgvc9
 156 1
prepare.py
in mlebench/competitions/rsna-2022-cervical-spine-fracture-detection
 153 1
prepare.py
in mlebench/competitions/mlsp-2013-birds
 143 2
grade.py
in mlebench/competitions/vinbigdata-chest-xray-abnormalities-detection
 136 9
start.py
in agents/opendevin
 134 3
config.yaml
in agents/aide
 130 -
prepare.py
in mlebench/competitions/google-research-identify-contrails-reduce-global-warming
 122 1
download_kernels.py
in extras/kernels
 121 2
analyze.py
in extras/plagiarism_detector
 118 4
registry.py
in mlebench
 113 10
prepare.py
in mlebench/competitions/tgs-salt-identification-challenge
 113 1
prepare.py
in mlebench/competitions/iwildcam-2020-fgvc7
 112 1
grade.py
in mlebench
 110 4
prepare.py
in mlebench/competitions/rsna-breast-cancer-detection
 104 1
prepare.py
in mlebench/competitions/smartphone-decimeter-2022
 104 2
classes.py
in mlebench/competitions/leaf-classification
 101 -
run.py
in agents
 100 4
prepare.py
in mlebench/competitions/hms-harmful-brain-activity-classification
 98 1
grade.py
in mlebench/competitions/siim-covid19-detection
 98 4
prepare.py
in mlebench/competitions/siim-isic-melanoma-classification
 97 1
prepare.py
in mlebench/competitions/icecube-neutrinos-in-deep-ice
 96 1
grade.py
in mlebench/competitions/jigsaw-unintended-bias-in-toxicity-classification
 95 6
prepare.py
in mlebench/competitions/cdiscount-image-classification-challenge
 95 1
prepare.py
in mlebench/competitions/stanford-covid-vaccine
 95 1
grade.py
in mlebench/competitions/uw-madison-gi-tract-image-segmentation
 94 4
grade.py
in mlebench/competitions/tgs-salt-identification-challenge
 93 3
prepare.py
in mlebench/competitions/denoising-dirty-documents
 93 2
utils.py
in environment
 90 5
prepare.py
in mlebench/competitions/uw-madison-gi-tract-image-segmentation
 89 2
prepare.py
in mlebench/competitions/text-normalization-challenge-english-language
 89 1
prepare.py
in mlebench/competitions/text-normalization-challenge-russian-language
 89 1
prepare.py
in mlebench/competitions/vesuvius-challenge-ink-detection
 89 1
prepare.py
in mlebench/competitions/vinbigdata-chest-xray-abnormalities-detection
 86 1
grade.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 85 3
make_submission.py
in experiments
 84 1
prepare.py
in mlebench/competitions/cassava-leaf-disease-classification
 84 1
Large Files That Change Often (Top 1)
To change a file, an AI coding agent (or a person) has to read it: these files are read the most for the changes in the past year (lines × changes).
File# lines≈ tokens
to read		# changes
(1y)		≈ lines edited
per change		lines read per
line changed
282 ~1.5k–5k 1 ~300 ~1

Tokens are lines × 7 to 14 (analysis.fileReadsForChanges). Lines edited per change are the file's lines added and deleted divided by its commits over the whole history. Agents read about 2,000 lines at a time: when a 17,000-line file was split into modules, the tokens an agent read for the same change fell by 83% (The Economic Benefit of Refactoring, 2026).

Files With Most Units (Top 50)
File# lines# units
utils.py
in mlebench
 190 21
mAP_evaluation.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 215 18
data.py
in mlebench
 275 14
282 12
registry.py
in mlebench
 113 10
grade.py
in mlebench/competitions/vinbigdata-chest-xray-abnormalities-detection
 136 9
utils.py
in mlebench/competitions
 167 9
grade_helpers.py
in mlebench
 171 7
grade.py
in mlebench/competitions/jigsaw-unintended-bias-in-toxicity-classification
 95 6
notebook.py
in mlebench/competitions/smartphone-decimeter-2022
 60 6
familiarity.py
in experiments/familiarity
 213 5
grade.py
in mlebench/competitions/rsna-2022-cervical-spine-fracture-detection
 62 5
utils.py
in environment
 90 5
run.py
in agents
 100 4
registry.py
in agents
 78 4
grade.py
in mlebench/competitions/tweet-sentiment-extraction
 37 4
grade.py
in mlebench/competitions/uw-madison-gi-tract-image-segmentation
 94 4
grade.py
in mlebench/competitions/siim-covid19-detection
 98 4
prepare.py
in mlebench/competitions/billion-word-imputation
 72 4
grade.py
in mlebench/competitions/alaska2-image-steganalysis
 64 4
grade.py
in mlebench
 110 4
analyze.py
in extras/plagiarism_detector
 118 4
utils.py
in agents
 22 3
start.py
in agents/opendevin
 134 3
metrics.py
in mlebench
 29 3
grade.py
in mlebench/competitions/osic-pulmonary-fibrosis-progression
 52 3
grade.py
in mlebench/competitions/bms-molecular-translation
 25 3
grade.py
in mlebench/competitions/AI4Code
 48 3
kaggle_metric_utilities.py
in mlebench/competitions/hms-harmful-brain-activity-classification
 56 3
grade.py
in mlebench/competitions/rsna-breast-cancer-detection
 45 3
grade.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 85 3
grade.py
in mlebench/competitions/multi-modal-gesture-recognition
 35 3
grade.py
in mlebench/competitions/freesound-audio-tagging-2019
 37 3
grade.py
in mlebench/competitions/chaii-hindi-and-tamil-question-answering
 37 3
grade.py
in mlebench/competitions/tgs-salt-identification-challenge
 93 3
grade.py
in mlebench/competitions/champs-scalar-coupling
 35 3
grade.py
in mlebench/competitions/icecube-neutrinos-in-deep-ice
 59 3
grading_server.py
in environment
 28 3
analyze.py
in extras/rule_violation_detector
 232 3
grade.py
in mlebench/competitions/predict-volcanic-eruptions-ingv-oe
 29 2
prepare.py
in mlebench/competitions/osic-pulmonary-fibrosis-progression
 72 2
grade.py
in mlebench/competitions/nomad2018-predict-transparent-conductors
 39 2
prepare.py
in mlebench/competitions/bms-molecular-translation
 43 2
prepare.py
in mlebench/competitions/AI4Code
 55 2
grade.py
in mlebench/competitions/h-and-m-personalized-fashion-recommendations
 27 2
prepare.py
in mlebench/competitions/uw-madison-gi-tract-image-segmentation
 89 2
grade.py
in mlebench/competitions/lmsys-chatbot-arena
 39 2
grade.py
in mlebench/competitions/text-normalization-challenge-english-language
 34 2
kuzushiji_metric.py
in mlebench/competitions/kuzushiji-recognition
 73 2
grade.py
in mlebench/competitions/kuzushiji-recognition
 30 2
Files With Long Lines (Top 29)

There are 29 files with lines longer than 120 characters. In total, there are 52 long lines.

File# lines# units# long lines
prepare.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 324 1 8
prepare.py
in mlebench/competitions/paddy-disease-classification
 46 2 5
prepare.py
in mlebench/competitions/vinbigdata-chest-xray-abnormalities-detection
 86 1 4
prepare.py
in mlebench/competitions/iwildcam-2019-fgvc6
 82 1 4
203 - 3
make_submission.py
in experiments
 84 1 2
prepare.py
in mlebench/competitions/plant-seedlings-classification
 57 2 2
prepare.py
in mlebench/competitions/herbarium-2020-fgvc7
 176 1 2
run.py
in extras/rule_violation_detector
 42 - 2
run.py
in agents
 100 4 1
start.py
in agents/opendevin
 134 3 1
grade.py
in mlebench/competitions/predict-volcanic-eruptions-ingv-oe
 29 2 1
prepare.py
in mlebench/competitions/bms-molecular-translation
 43 2 1
prepare.py
in mlebench/competitions/cdiscount-image-classification-challenge
 95 1 1
prepare.py
in mlebench/competitions/stanford-covid-vaccine
 95 1 1
prepare.py
in mlebench/competitions/facebook-recruiting-iii-keyword-extraction
 27 1 1
prepare.py
in mlebench/competitions/billion-word-imputation
 72 4 1
grade.py
in mlebench/competitions/3d-object-detection-for-autonomous-vehicles
 85 3 1
grade.py
in mlebench/competitions/multi-modal-gesture-recognition
 35 3 1
utils.py
in mlebench/competitions
 167 9 1
prepare.py
in mlebench/competitions/plant-pathology-2020-fgvc7
 54 1 1
prepare.py
in mlebench/competitions/random-acts-of-pizza
 57 1 1
prepare.py
in mlebench/competitions/statoil-iceberg-classifier-challenge
 80 1 1
grade.py
in mlebench/competitions/nfl-player-contact-detection
 25 2 1
prepare.py
in mlebench/competitions/icecube-neutrinos-in-deep-ice
 96 1 1
cli.py
in mlebench
 200 1 1
data.py
in mlebench
 275 14 1
grade.py
in mlebench
 110 4 1
analyze.py
in extras/plagiarism_detector
 118 4 1
Correlations

File Size vs. Commits (all time): 5 points

experiments/aggregate_grading_reports.py x: 1 commits (all time) y: 282 lines of code mlebench/cli.py x: 1 commits (all time) y: 200 lines of code mlebench/registry.py x: 2 commits (all time) y: 113 lines of code mlebench/competitions/the-icml-2013-whale-challenge-right-whale-redux/prepare.py x: 1 commits (all time) y: 60 lines of code mlebench/data.py x: 1 commits (all time) y: 275 lines of code
282.0
lines of code
  min: 60.0
  average: 186.0
  25th percentile: 86.5
  median: 200.0
  75th percentile: 278.5
  max: 282.0
0 2.0
commits (all time)
min: 1.0 | average: 1.2 | 25th percentile: 1.0 | median: 1.0 | 75th percentile: 1.5 | max: 2.0

File Size vs. Contributors (all time): 5 points

experiments/aggregate_grading_reports.py x: 1 contributors (all time) y: 282 lines of code mlebench/cli.py x: 1 contributors (all time) y: 200 lines of code mlebench/registry.py x: 2 contributors (all time) y: 113 lines of code mlebench/competitions/the-icml-2013-whale-challenge-right-whale-redux/prepare.py x: 1 contributors (all time) y: 60 lines of code mlebench/data.py x: 1 contributors (all time) y: 275 lines of code
282.0
lines of code
  min: 60.0
  average: 186.0
  25th percentile: 86.5
  median: 200.0
  75th percentile: 278.5
  max: 282.0
0 2.0
contributors (all time)
min: 1.0 | average: 1.2 | 25th percentile: 1.0 | median: 1.0 | 75th percentile: 1.5 | max: 2.0

File Size vs. Commits (30 days): 0 points

No data for "commits (30d)" vs. "lines of code".

File Size vs. Contributors (30 days): 0 points

No data for "contributors (30d)" vs. "lines of code".

File Size vs. Commits (90 days): 0 points

No data for "commits (90d)" vs. "lines of code".

File Size vs. Contributors (90 days): 0 points

No data for "contributors (90d)" vs. "lines of code".

About This Analysis
  • 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...