For Data Scientists & ML Engineers · Footwear
Footwear Data for Data Scientists
Footwear data for data scientists: 7 datasets on one shelf. Every one delivered as API, files, or warehouse rows.
financial time series api for backtesting · alternative data for quantitative research · where to get training data for footwear models
API, files, or your warehouse. Daily, weekly, or hourly.
What can you train on shoe image corpora?
The relative-attribute pair supervision lets you train ranking-style heads, not just flat classifiers, and because the release is static you can pin one snapshot in your experiment tracker and get bit-reproducible runs.
Hugging Face lexington/Sneakers - Sneaker Image Dataset plays a narrower role: quality 4/10, a single-image imagefolder repository shipping JPG plus Parquet. Its value is operational rather than statistical - it works as a zero-setup fixture when you are testing datasets loading code or CI smoke tests before pointing them at the 50k-image corpus.
Where do resale and assortment signals come from?
Zappos - Structured Footwear Catalog (50k+ SKUs).org Product JSON-LD reachable through 22 navigation facets plus XML sitemaps, which turns assortment, price and rating extraction into a parsing problem instead of a reverse-engineering one.
How should you assemble the stack?
Register a USITC DataWeb account and schedule monthly pulls of the 10-digit-HTS panels; join the OEC's annual bilateral HS64 series for non-US corridors, keeping Parquet as the interchange format. Digitize the APICCAPS Yearbook once per edition for country aggregates.
This page is the footwear slice of all data-scientists resources; the footwear data hub applies the same rubric to all 7 datasets we catalog in the industry.
Straight answers
Is there a financial time series API for backtesting footwear strategies?
No equity tape exists in this slice, but two series serve as backtest inputs. The USITC DataWeb returns monthly footwear imports by 10-digit HTS, country and district from 1989 as JSON through its registered-account REST API, while the OEC serves 1995-2024 bilateral HS64 flows queryable as JSON, CSV, Parquet or TSV.
Where can I find alternative data for quantitative research on footwear?
Pair it with Zappos' daily-refreshed structured catalog of 50k+ SKUs for assortment, price and rating features keyed on product identifiers.
Where to get training data for footwear models?
Start with UT-Zappos50K's 50,025 labeled images for fine-grained visual classification, add the lexington/Sneakers Hugging Face repo as a zero-setup loader fixture, then wire USITC DataWeb and OEC trade series as demand-side targets.
Rows before rollout
Sample rows from any shelf entry — the field dictionary and coverage notes ride along. If the shelf misses what you need, say so; sourcing requests are half our job.
Talk to us