Dataset Viewer
Auto-converted to Parquet Duplicate
file_name
stringlengths
5
44
extension
stringclasses
121 values
size_bytes
int64
99
33.6B
folder_hint
stringclasses
10 values
label
stringclasses
7 values
sunset_2022.jpg
jpg
2,223,361
Downloads
images
paul_draft.md
md
570
Downloads
documents
source_13.zst
zst
47,371,601
Downloads
archives
expenses_summary_2023.pdf
pdf
378,023
Downloads
documents
documents_2022-10-16.zip
zip
2,315,558
Backups
archives
core.c
c
23,666
Projects
code
music_2024-05-09.zip
zip
5,595,570
Downloads
archives
export_2024-12.toml
toml
815
Downloads
code
S08E01.mov
mov
1,790,082,181
Videos
video
AppAuth.sh
sh
2,166
Projects
code
nina_invoice.txt
txt
4,466
Downloads
documents
icon_final.webp
webp
131,645
Pictures
images
old_2025-09-04.zip
zip
9,552,367
Projects
archives
Screen Recording 2025-09-15 at 09.36.38.webm
webm
114,064,554
Videos
video
Review Guide.txt
txt
23,532
Downloads
documents
ApiDatabase.js
js
3,746
Projects
code
system_7.0.8.iso
iso
1,410,209,803
Downloads
other
setup.dat
dat
351,285
Downloads
other
schedule_q2.numbers
numbers
418,317
Downloads
documents
VID_20220707_232121.mpg
mpg
96,386,077
Videos
video
app11.js
js
28,283
Projects
code
GOPR8807.mp4
mp4
263,043,204
Desktop
video
trailer_1.mp4
mp4
483,469,079
Downloads
video
router.rs
rs
1,902
Work
code
index_train.ts
ts
9,963
Downloads
code
old (4).gz
gz
860,567
Work
archives
track14.m4a
m4a
339,434
Work
audio
vmware.exe
exe
5,240,242
Downloads
other
DSC_1373.jpeg
jpeg
1,266,608
Pictures
images
checklist-2022-07.epub
epub
3,279,155
Downloads
documents
IMG_6793.png
png
9,189,421
Downloads
images
family_2022.png
png
23,786
Desktop
images
clip126.mpeg
mpeg
1,554,466,930
School
video
Screenshot 2025-11-06 at 16.39.14.heic
heic
1,180,800
Pictures
images
PXL_20230625_212880.jpg
jpg
2,358,856
Downloads
images
GOPR8496.MP4
mp4
32,788,442
Videos
video
main_config.ipynb
ipynb
97,388
Downloads
code
05 - Dreams Sky.m4a
m4a
17,908,040
Desktop
audio
cache_7.9.1.log
log
20,500
Downloads
other
review_quote_2026.xlsx
xlsx
599,155
Documents
documents
router6.swift
swift
5,585
School
code
fonts_2023-06-11.zip
zip
7,830,343
Downloads
archives
node-linux.dmg
dmg
78,881,945
Desktop
other
agenda_copy.pdf
pdf
1,985,736
Downloads
documents
podcast_party_52.mp3
mp3
2,266,795
Music
audio
Screen Recording 2024-01-20 at 02.30.51.mp4
mp4
502,951,787
Videos
video
IMG_0068.bmp
bmp
385,547
Pictures
images
orders_4.sql
sql
644,894
Projects
code
PXL_20241006_159658.svg
svg
19,247
Downloads
images
backup_2022-06-10.zip
zip
63,631,750
Backups
archives
license_20230519.apk
apk
23,766,319
Downloads
other
voice_memo_20240806.m4a
m4a
2,790,126
Music
audio
Screenshot 2022-07-21 at 11.06.19.gif
gif
2,259,550
Desktop
images
Agenda Lecture.docx
docx
44,039
Downloads
documents
archive (4).zip
zip
81,212,047
Downloads
archives
thesis (2).rtf
rtf
133,306
Downloads
documents
license_20231014.pkg
pkg
139,866,377
Work
other
DSC_6354.tiff
tiff
21,075,800
Work
images
S06E15.m4v
m4v
4,294,967,296
Videos
video
letter_v2.docx
docx
130,295
Documents
documents
PXL_20230311_130343.jpg
jpg
2,867,974
School
images
patch.deb
deb
27,873,166
Downloads
other
SETUP_1.4.7.DEB
deb
162,154,575
Downloads
other
music_2026-02-23.zip
zip
18,456,902
Backups
archives
deploy.jsx
jsx
2,040
Projects
code
TIMESHEET LECTURE.PDF
pdf
886,588
School
documents
john_offer.docx
docx
468,999
Downloads
documents
Screenshot 2026-07-22 at 21.07.34.ico
ico
27,927
Work
images
clip151.avi
avi
559,196,433
Videos
video
setup.lnk
lnk
975
Downloads
other
IMG_7855.gif
gif
274,618
School
images
maria_research.txt
txt
6,168
Downloads
documents
backup_4.tar.gz
tar.gz
195,122,008
Downloads
archives
track11.mp3
mp3
895,129
Music
audio
budget_nov.pptx
pptx
588,326
Documents
documents
GOPR8191.mp4
mp4
73,813,012
Videos
video
core.cpp
cpp
2,286
Work
code
main.py
py
3,307
Projects
code
david_contract.docx
docx
49,816
Documents
documents
GOPR9078.wmv
wmv
253,746,152
Videos
video
concert_34.webp
webp
183,571
Work
images
documents_2023-04-02.zip
zip
43,662,241
Downloads
archives
holiday_2025.webp
webp
142,655
Desktop
images
intro_23.mp4
mp4
303,647,165
Videos
video
PXL_20230330_164559.HEIC
heic
2,246,569
Downloads
images
debug-x64.pkg
pkg
202,181,685
Downloads
other
Screenshot 2022-11-12 at 05.55.09.png
png
808,089
Desktop
images
export_8.zip
zip
13,082,679
Downloads
archives
helper.xml
xml
5,222
Projects
code
config_run.ts
ts
3,931
Projects
code
lecture_mar.pdf
pdf
1,728,604
Downloads
documents
release.tar.gz
tar.gz
12,626,053
Downloads
archives
DSC_2238.jpg
jpg
648,191
Downloads
images
ringtone_8.m4a
m4a
9,390,694
Music
audio
recording_45.mp3
mp3
1,073,132
Projects
audio
notes_copy.txt
txt
24,494
Documents
documents
cover_v2.png
png
7,652,327
Pictures
images
S04E02.mov
mov
71,080,316
Downloads
video
banner_light.gif
gif
289,953
Downloads
images
PXL_20221124_226976.bmp
bmp
440,346
Pictures
images
End of preview. Expand in Data Studio

Smart Downloads Organiser Metadata

A small synthetic dataset of realistic file-metadata records from a typical "Downloads" folder. It is designed for training and prototyping simple file-classification models, such as a smart downloads organizer that sorts files into folders by type.

Dataset summary

  • Rows: 670
  • Format: CSV with a header row (downloads_metadata.csv)
  • Classes: 7, roughly balanced
  • Size range: 147 bytes to about 7.9 GB (log-uniform sampling per file type, so small text files and very large media/ISO files are both well represented)
  • Data origin: programmatically generated, not scraped from real users. It contains no personal data.

Columns

Column Type Description
file_name string Full file name including extension, with realistic variants such as _final_v2, (1), camera-style names (IMG_4821.jpg, PXL_...) and WhatsApp-style names
extension string File extension without the dot (e.g. pdf, mp4, tar.gz)
size_bytes int File size in bytes
folder_hint string Folder the file was found in or downloaded to (e.g. Downloads, Pictures, Voice Memos, Downloads/work)
label string Target class (see below)

Labels

Label Rows Typical extensions
documents 100 pdf, docx, doc, txt, xlsx, xls, pptx, rtf, md, odt
audio 98 mp3, wav, m4a, flac, ogg, aac
code 98 py, js, ts, tsx, csv, ipynb, json, html, css, java, cpp, sql, sh, yaml, go, rs
images 97 jpg, jpeg, png, gif, webp, bmp, svg, heic
archives 94 zip, tar.gz, tgz, tar, rar, 7z, gz, tar.bz2
video 94 mp4, mov, mkv, avi, webm, wmv, flv
other 89 exe, msi, dmg, iso, apk, torrent, ttf, otf, deb, rpm, db, sqlite, pt, onnx, psd, epub, crdownload, and similar

Labelling convention

Labels follow a simple extension-based rule:

  • PDF, DOC(X), TXT, XLS(X), PPTX, RTF, MD, ODT map to documents
  • JPG, PNG, GIF, WEBP and similar map to images
  • MP3, WAV, M4A and similar map to audio
  • MP4, MOV, MKV and similar map to video
  • ZIP, TAR, RAR and similar map to archives
  • PY, JS, TS, CSV, IPYNB and similar map to code (note that CSV is deliberately treated as a code/data file)
  • Anything unusual maps to other

Quick start

from datasets import load_dataset

ds = load_dataset("AnodemTheGr8t/file_organiser_metadata", split="train")
print(ds[0])

Or with pandas and scikit-learn:

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression

df = pd.read_csv("downloads_metadata.csv")
df["text"] = df["extension"] + " " + df["folder_hint"] + " " + df["file_name"]

X_train, X_test, y_train, y_test = train_test_split(
    df["text"], df["label"], test_size=0.2, stratify=df["label"], random_state=42
)

vec = TfidfVectorizer(analyzer="char_wb", ngram_range=(2, 4))
clf = LogisticRegression(max_iter=1000).fit(vec.fit_transform(X_train), y_train)
print(clf.score(vec.transform(X_test), y_test))

Intended uses

  • Prototyping and teaching file or text classification
  • Benchmarking lightweight rule-based vs. ML approaches for file organization
  • Feature-engineering exercises on mixed tabular and text-like features (extension, size, path, file name)

Limitations

  • Synthetic data. Real download folders are messier than this dataset.
  • Labels are largely determined by extension, so a model can reach very high accuracy from that column alone. For a harder task, drop extension and classify from file_name stems, size_bytes and folder_hint.
  • Small size (670 rows), so it is not suitable for training production models.
  • English-language file names only.
  • The other class is a catch-all and is intentionally heterogeneous.

License

CC0 1.0 Universal (public domain dedication).

Citation

@dataset{file_organiser_metadata,
  author = {AnodemTheGr8},
  title  = {Smart Downloads Organiser Metadata},
  year   = {2026},
  url    = {https://huggingface.co/datasets/AnodemTheGr8t/file_organiser_metadata}
}
Downloads last month
38