Generated on August 26, 2026.
This catalog is based on:
Cleaning status is intentionally conservative:
no: looks like a raw or near-raw download/exportmixed: mostly upstream data, but converted, lightly
renamed, trimmed, or reformattedyes: clearly classroom-prepped, subsetted, aggregated,
or feature-engineeredunclear: provenance/source is not pinned down
confidently enough to say moreConfidence is about provenance, not data quality.
| File | Shape | Likely source family | Probable upstream source | Cleaned? | Confidence | Notes |
|---|---|---|---|---|---|---|
2022 election cycle fundraising.csv |
537 x 9 | Kaggle / campaign-finance classroom dataset | Likely the #MakeoverMonday / Kaggle version of 2022 congressional fundraising data built from FEC/OpenSecrets context: VizWiz prompt | yes |
medium | Small summary table by member, not raw FEC filings. |
CardiacPatientData.csv |
5,906 x 20 | Academic clinical dataset | Likely Cardiac Patient Bed Head Ticket Dataset | mixed |
medium | Columns match the published cardiac variables very closely; local file looks like a flat CSV export of the research data. |
ForeignGifts_edu.csv |
28,221 x 10 | U.S. government open data | U.S. Department of Education Section 117 / Foreign Gifts and Contracts Report | no |
high | Columns match the DOE spreadsheet export almost exactly. |
SIPRI Military Expenditure Database.csv |
5,882 x 7 | SIPRI official data | SIPRI Military Expenditure Database | mixed |
high | Looks like a tidy/exported slice with an added index column. |
USA_cars_datasets.csv |
2,499 x 13 | Kaggle | US Cars Dataset | no |
high | Exact filename match. |
airbnb_NYC.csv |
30,478 x 13 | Inside Airbnb classroom/community extract | Core upstream source is Inside Airbnb | mixed |
medium | Already reduced to a teaching-friendly 13-column NYC listing table; not a full raw Inside Airbnb dump. |
airbnb_hw.csv |
30,478 x 13 | Duplicate classroom copy | Same likely upstream as airbnb_NYC.csv: Inside Airbnb |
yes |
high | Same shape/columns as airbnb_NYC.csv; likely the
homework working copy. |
ames_prices.csv |
2,930 x 82 | Ames Housing teaching dataset | Dean De Cock’s Ames Housing data: Ames paper reference and common Kaggle mirror | yes |
high | Local codebook matches Ames; area/price
naming suggests class-side renaming/cleanup. |
brazil_default.csv |
50,000 x 54 | Brazilian credit/default dataset | Upstream page not pinned down confidently; local
brazil_codebook.XLS strongly suggests a Brazilian credit
application/default dataset |
mixed |
low | Kept in catalog because the codebook is detailed, but I could not verify a clean public landing page from the web. |
car_prices.csv |
92 x 20 | Local/classroom derivative | Likely derived from local Charlottesville Craigslist car data rather than a public canonical dataset | yes |
medium | Enriched version of the local Craigslist car scrape with extra parsed fields and free text. |
cars_env.csv |
524 x 13 | EPA model input / policy modeling | EPA CVCM / OMEGA vehicle-sheet format documented in Consumer Vehicle Choice Model documentation | mixed |
medium | predicted mpg and baseline mpg/sales
suggest model-ready input rather than raw market data. |
cars_hw.csv |
976 x 12 | Used-car classroom dataset | Likely a Kaggle/community used-car dataset, but no exact upstream page confirmed | yes |
low | Filename and structure strongly suggest a homework-prepped file. |
college_completion.csv |
3,798 x 63 | Kaggle | College Completion Dataset | no |
high | Kaggle page says the data draw from NCES IPEDS and the Voluntary System of Accountability. |
corporate_ratings.csv |
2,029 x 31 | Kaggle | Corporate Credit Rating | no |
high | Exact schema match. |
craigslist_cville_cars.csv |
46 x 7 | Local scrape | Charlottesville Craigslist listings, locally scraped | yes |
high | Not a third-party packaged dataset; this looks like your own small scrape/export. |
crimebystatecombinedwithunemployment.csv |
1,989 x 13 | Kaggle | Crime rate and unemployment rate by state | no |
high | Exact filename match. |
cville_weather.csv |
411 x 13 | NOAA climate data | NOAA Climate Data Online / Daily Summaries, Charlottesville area: CDO Daily Summaries location details | no |
high | Local file uses NOAA station-style columns and station
US1VACRC002. |
diabetes-dataset.csv |
2,000 x 9 | Kaggle | Likely diadata or a very similar Kaggle Pima-style repost | no |
medium | Exact filename appears on Kaggle; not the original 768-row UCI Pima file. |
divorce/divorce_data.csv |
170 x 55 | UCI | UCI Divorce Predictors data set | no |
high | Exact row count and 54 predictors + target. |
drilling_rigs.csv |
623 x 10 | U.S. government energy data | EIA monthly drilling/rig statistics family, likely from the Monthly Energy Review or a related EIA table | no |
medium | Structure looks like a straight EIA table export. |
electricity_data_validation.csv |
108 x 14 | EIA API extract | EIA electricity retail sales metadata/API: EIA open data docs | yes |
high | Small Virginia-only recent-period validation slice, clearly not the full national dataset. |
energy.csv |
632 x 10 | EIA spreadsheet export | Monthly Energy Review | no |
high | The first rows are title/release-date rows from an EIA spreadsheet export. |
heart_failure_clinical_records_dataset.csv |
299 x 13 | UCI | UCI Heart Failure Clinical Records | no |
high | Exact filename match. |
iowa.csv |
159,904 x 14 | Iowa government open data | Iowa Liquor Sales | mixed |
high | Column names match the official dataset, but this file is a classroom-sized subset/export rather than the full live table. |
justice_data.parquet |
22,986 x 709 | Virginia official pretrial data | Virginia Pre-Trial Data Project | mixed |
high | Very likely the wide official public dataset converted from CSV/XLS to Parquet for easier work. |
land_mines.csv |
338 x 4 | UCI | UCI Land Mines | yes |
high | Row count matches UCI, but headers have been made friendlier
(voltage, height, soil,
mine_type). |
mammogram.csv |
89,835 x 2 | OpenIntro teaching dataset | OpenIntro-style teaching distribution; see local companion doc mammogram.txt |
yes |
high | Highly distilled two-column teaching version, not a raw clinical file. |
metabric.csv |
1,343 x 14 | Kaggle / derived METABRIC subset | Likely based on Breast Cancer (METABRIC) | yes |
medium | Much smaller and tidier than the common full Kaggle METABRIC tables, so this looks class-prepared. |
mn_police_use_of_force.csv |
12,925 x 13 | OpenIntro from Minneapolis open data | OpenIntro dataset page; original city source: Minneapolis use-of-force data | mixed |
high | Looks like the OpenIntro teaching distribution of Minneapolis public data. |
mnist/Z_train.parquet,
mnist/Z_test.parquet, mnist/y_train.parquet,
mnist/y_test.parquet |
60,000/10,000 image rows + labels | Official benchmark dataset, locally converted | Official MNIST page | yes |
high | Raw MNIST image/label split converted into Parquet matrices and label vectors. |
nhanes_data_17_18.csv |
8,366 x 198 | NHANES | CDC NHANES 2017-2018: data portal | yes |
high | This is a merged, cleaned, human-readable NHANES extract, not a raw single XPT file. |
nhanes_meta_17_18.csv |
197 x 18 | NHANES metadata | CDC NHANES 2017-2018: questionnaires/datasets page | yes |
high | Custom variable dictionary/metadata table layered on top of NHANES docs and recodes. |
patents.csv |
161 x 7 | WIPO-derived custom extract | WIPO IP Statistics Data Center | yes |
medium | Small country-by-country 2020-2021 extract; clearly a hand-selected summary rather than a raw WIPO bulk table. |
pierce_county_house_sales.csv |
16,814 x 19 | OpenIntro/usdata from county open data | OpenIntro/usdata page | mixed |
high | Distributed as a teaching dataset, sourced from Pierce County, Washington. |
pretrial_data.csv |
22,986 x 17 | Derived from Virginia pretrial data | Virginia Pre-Trial Data Project | yes |
high | Clearly a feature-selected/engineered modeling subset of
justice_data.parquet. |
sharks.csv |
6,462 x 257 | Shark attack registry / likely community mirror | Likely derived from the Global Shark Attack File | no |
medium | Messy wide structure with many empty columns suggests a raw-ish export, not a cleaned teaching table. |
tech_stocks.csv |
32,224 x 10 | Market API export / classroom aggregation | Schema matches Alpha Vantage TIME_SERIES_DAILY_ADJUSTED | mixed |
medium | Columns like close_adjusted and
split_coefficient strongly suggest an Alpha Vantage-style
export combined across symbols. |
time_use.csv |
67 x 7 | ATUS/BLS-derived summary | Likely derived from the American Time Use Survey / BLS ATUS releases | yes |
medium | Already aggregated to age-level summary measures; not raw diary records. |
tuna.csv |
10,032 x 1 | Biosequence classroom data | Upstream not confidently identified | mixed |
low | One-column nucleotide sequence table; almost certainly preprocessed for class use, but I could not pin a canonical public source. |
us_power_consumption.csv |
75 x 13 | EIA-derived annual energy summary | EIA energy data family / likely Monthly Energy Review | yes |
medium | Tidy annual totals table; much more processed than the raw
energy.csv export. |
wages_hw.csv |
731 x 6 | Kaggle-derived salary homework file | Likely derived from a Glassdoor salary dataset such as Data Science Job Posting on Glassdoor | yes |
medium | Minimal modeling subset with avg_salary,
job_state, and a few employer descriptors. |
weapons-trade-register.csv |
CSV with SIPRI preamble + transfer rows | SIPRI official data | SIPRI Arms Transfers Database / register export interface | no |
high | Straight export with explanatory lines above the header; definitely near-raw. |
zoo.csv |
101 x 18 | UCI | UCI Zoo | no |
high | Classic UCI dataset with standard 101-row shape. |
These are not primary datasets themselves, but they explain or support the tabular files above.
| File | Relates to | Notes |
|---|---|---|
ames_codebook.txt |
ames_prices.csv |
Local codebook for Ames Housing; confirms provenance. |
brazil_codebook.XLS |
brazil_default.csv |
Local variable dictionary for the Brazilian credit/default file. |
heart_failure_readme.txt |
heart_failure_clinical_records_dataset.csv |
Local UCI-style README with provenance and citation info. |
VirginiaPretrialData2017Codebook.pdf |
justice_data.parquet,
pretrial_data.csv |
Official Virginia codebook for the October 2017 cohort. |
mammogram.txt |
mammogram.csv |
Local dataset description and original study citation. |
use_of_force.txt |
mn_police_use_of_force.csv |
Local dataset description matching OpenIntro wording. |
divorce/about.txt |
divorce/divorce_data.csv |
Local description of the divorce predictors data. |
divorce/questions.txt |
divorce/divorce_data.csv |
Human-readable questionnaire text for Q1-Q54. |
divorce/reference.tsv |
divorce/divorce_data.csv |
Variable-to-question lookup table. |
mnist/example_solution.ipynb |
MNIST Parquet files | Teaching notebook/example solution using the local MNIST files. |
cpr.txt |
no matching CSV in this directory | Dataset description for a CPR class example; the tabular file itself is not present here. |
Effectiveness of Bystander Cardiopulmonary Resuscitation and Survival Following Out-of-Hospital Cardiac Arrest.pdf |
CPR/cardiac arrest teaching material | Source article or background reading; not a tabular dataset. |
UCI shows up clearly in
heart_failure_clinical_records_dataset.csv,
zoo.csv, divorce/divorce_data.csv, and
land_mines.csv.NHANES is clearly represented by
nhanes_data_17_18.csv and
nhanes_meta_17_18.csv, and both are already
classroom-cleaned/merged.Kaggle/community mirrors show up heavily in
USA_cars_datasets.csv, corporate_ratings.csv,
college_completion.csv,
crimebystatecombinedwithunemployment.csv, and likely
several of the homework-prepped derivatives.Official/public-sector data is a large share of the
directory too: DOE Section 117, NOAA weather, Iowa liquor sales,
Virginia pretrial, EIA energy/electricity/rig data, SIPRI, and
WIPO.airbnb_hw.csv, ames_prices.csv,
car_prices.csv, craigslist_cville_cars.csv,
electricity_data_validation.csv,
pretrial_data.csv, the MNIST Parquet set, both NHANES CSVs,
patents.csv, us_power_consumption.csv, and
wages_hw.csv.brazil_default.csv,
cars_hw.csv, and tuna.csv; those are the ones
I would label as inferred rather than fully verified if someone
asks.