India Open Government Data — Bharat Economy Datasets
Collection
Research and non-commercial use (CC BY-NC-SA 4.0). Commercial: surya@judicialmind.ai / hello@judicialmind.ai • 3 items • Updated
You need to agree to share your contact information to access this dataset
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
This dataset is released under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 licence (CC BY-NC-SA 4.0): research, academic, personal and other non-commercial use is permitted with attribution; adaptations must be shared under the same licence; commercial use is not permitted. The original public source data remains under the Government Open Data License - India (GODL-India). Commercial use: contact surya@judicialmind.ai or hello@judicialmind.ai.
Log in or Sign Up to review the conditions and access this dataset content.
Bharat-Workforce-Atlas
Licence: CC BY-NC-SA 4.0 — research and non-commercial use only; derivatives must be shared under the same licence. Commercial use: contact surya@judicialmind.ai or hello@judicialmind.ai
Access is approved automatically when the email you enter matches your Hugging Face account email and belongs to the organisation you name; other requests are reviewed manually.
District-level counts of unorganised workers registered on the eShram portal, by occupation, gender, age band and education, built from 233,068,163 eShram worker records published on data.gov.in (snapshot May 2026). Two companion aggregates sit alongside: PM-KISAN beneficiaries per district x installment x gender, and UDYAM MSME registrations per district x month (and x NIC-2 division). Aggregates only: no row-level data is included, every count below 10 is suppressed, and districts carry Local Government Directory (LGD) codes so they join to other district datasets.
Quickstart
All files are Parquet. Download the repository once (it is gated: accept the terms on the dataset page and log in
with huggingface-cli login first), then read the tables with pandas or query them with DuckDB.
# pip install pandas pyarrow duckdb huggingface_hub
import duckdb
import pandas as pd
from huggingface_hub import snapshot_download
root = snapshot_download("judicialmind/Bharat-Workforce-Atlas", repo_type="dataset")
tot = pd.read_parquet(f"{root}/atlas/district_total.parquet")
print(tot.nlargest(3, "workers")[["state_name", "district_name", "district_lgd", "workers"]].to_string(index=False))
# female share of registrants in each sector, nationally (suppressed cells are null and skipped by SUM)
print(duckdb.sql(f"""
SELECT occupation, SUM(workers) AS workers,
ROUND(SUM(workers) FILTER (WHERE gender = 'FEMALE') / SUM(workers), 3) AS female_share
FROM '{root}/atlas/district_occupation_gender_age_education.parquet'
GROUP BY occupation ORDER BY workers DESC LIMIT 3
""").df().to_string(index=False))
comp = pd.read_parquet(f"{root}/atlas/state_completeness.parquet")
print(comp.sort_values("share").head(3)[["state_name", "rows_fetched", "api_total", "share"]].to_string(index=False))
Output:
Top 3 districts by registered workers
| State | District | LGD code | Workers |
|---|---|---|---|
| WEST BENGAL | MURSHIDABAD | 319 | 2,424,210 |
| UTTAR PRADESH | PRAYAGRAJ | 120 | 2,332,328 |
| WEST BENGAL | 24 PARAGANAS SOUTH | 304 | 2,290,466 |
Top 3 sectors nationally, with female share of registrants
| Occupation | Workers | Female share |
|---|---|---|
| AGRICULTURE | 118,512,734 | 0.491 |
| DOMESTIC AND HOUSEHOLD WORKERS | 21,944,761 | 0.959 |
| CONSTRUCTION | 20,877,440 | 0.247 |
States with the lowest fetch completeness
| State | Rows fetched | Portal total | Share fetched |
|---|---|---|---|
| RAJASTHAN | 280,000 | 17,047,448 | 0.0164 |
| JHARKHAND | 360,000 | 11,087,381 | 0.0325 |
| TAMIL NADU | 360,000 | 10,173,361 | 0.0354 |
A single table also loads with datasets:
from datasets import load_dataset
ds = load_dataset("judicialmind/Bharat-Workforce-Atlas", "district_total", split="train")
Files
| file | rows | suppressed cells |
|---|---|---|
atlas/district_age_band.parquet |
4,636 | 211 |
atlas/district_disabled.parquet |
1,518 | 116 |
atlas/district_education.parquet |
11,393 | 1,327 |
atlas/district_gender.parquet |
1,983 | 651 |
atlas/district_occupation.parquet |
22,918 | 3,382 |
atlas/district_occupation_gender_age_education.parquet |
1,393,212 | 784,320 |
atlas/district_total.parquet |
770 | 6 |
atlas/national_age_band.parquet |
7 | 0 |
atlas/national_disabled.parquet |
2 | 0 |
atlas/national_education.parquet |
16 | 0 |
atlas/national_gender.parquet |
3 | 0 |
atlas/national_occupation.parquet |
35 | 1 |
atlas/state_age_band.parquet |
218 | 8 |
atlas/state_disabled.parquet |
70 | 0 |
atlas/state_education.parquet |
552 | 31 |
atlas/state_gender.parquet |
99 | 16 |
atlas/state_occupation.parquet |
1,128 | 70 |
atlas/state_completeness.parquet |
35 | - |
pmkisan/district_installment_gender.parquet |
49,164 | - |
udyam/district_month.parquet |
53,263 | - |
udyam/district_month_nic2.parquet |
2,775,245 | - |
udyam/district_year_nic2.parquet |
384,827 | - |
The main table is atlas/district_occupation_gender_age_education.parquet (one row per non-empty district x
occupation x gender x age band x education cell). The other atlas/ files are marginals: district / state / national
x each single dimension, and district totals.
Schema
eShram tables (atlas/; marginal files carry the subset of columns they need):
| column | type | meaning |
|---|---|---|
state_lgd, district_lgd |
int | Local Government Directory state / district code (matched by name) |
src_state_code, src_district_code |
int | state / district code as published by eShram |
state_name, district_name |
str | names as published by eShram (upper case) |
occupation |
str | eShram sector (35 values, e.g. AGRICULTURE, CONSTRUCTION, DOMESTIC AND HOUSEHOLD WORKERS) |
gender |
str | FEMALE, MALE, TRANSGENDER |
age_band |
str | under_18, 18-25, 26-35, 36-45, 46-55, 56-60, 60+ (age as registered) |
education |
str | eShram education level (16 values, NOT LITERATE to POST-GRADUATE AND ABOVE and technical degrees) |
disabled |
bool | disability status (only in the *_disabled marginals) |
workers |
int | registered workers in the cell; null when suppressed |
suppressed |
bool | true when workers is withheld (see Privacy) |
atlas/state_completeness.parquet: state_name, rows_fetched (records returned by the portal),
api_total (records the portal reported for the state), share (= rows_fetched / api_total), fetch_errors,
source.
pmkisan/district_installment_gender.parquet: state_lgd, district_lgd, state_name, district_name, installment
(release number 1-22), fin_year_id, quadrimester, period_from, period_to (installment period, 2018-12-01 to
2026-03-31), n_villages (villages reporting), gender (female / male / transgender), beneficiaries,
suppressed.
udyam/district_month.parquet: state_lgd, district_lgd, state_name, district_name, year, month, registrations
(MSME registrations dated in that month, 2020-2026), suppressed.
udyam/district_month_nic2.parquet and udyam/district_year_nic2.parquet: state_lgd, district_lgd, year[, month],
nic2 (NIC-2008 two-digit division), enterprises_with_activity, suppressed. An enterprise with activities in
several NIC-2 divisions counts once in each, so NIC-2 counts do not sum to registrations.
Privacy
- Every count below 10 is replaced by null with
suppressed = true(primary suppression). - Complementary suppression: wherever a published total would let a reader recover a single suppressed cell by subtraction, the next-smallest cell in that group is suppressed too, so suppressed cells can only be bounded, never recovered. Cube: 784,320 primary and 0 complementary suppressions.
- Disability status is published only as a district/state/national marginal, never crossed with other fields.
- Gender: where a district has fewer than 10 transgender registrants, that cell is suppressed and so is the MALE cell (keeping FEMALE visible), because the district total would otherwise reveal it. State and national gender counts are complete.
- UDYAM enterprise names, addresses and PIN codes are never read.
Coverage
- India, 770 eShram districts in 35 states and union territories; LGD district codes assigned for 100.0% of workers: 99.1% by district name, the rest from eShram's own district code (eShram codes equal LGD codes wherever the name matched). Rows without a code keep the eShram district name.
- eShram: 233,068,163 registrations, snapshot May 2026. eShram is a voluntary registry: it counts who registered.
- Per-state completeness is in
atlas/state_completeness.parquet. The portal API stopped early (HTTP 429) for some states, so their counts are partial: Rajasthan 2% of 17,047,448; Jharkhand 3% of 11,087,381; Tamil Nadu 4% of 10,173,361; Karnataka 4% of 15,457,484; Andhra Pradesh 6% of 8,981,973; Uttar Pradesh 60% of 85,880,774; Gujarat 77% of 13,082,609; Odisha 88% of 14,880,985; Maharashtra 94% of 24,469,440; completeness unknown (stopped before the API reported a total): Punjab. For Uttar Pradesh, Uttarakhand and West Bengal the per-state shard download held more rows and was used; no worker is counted twice. - PM-KISAN: 49,164 district x installment x gender cells from 8,237,357 village rows (installments 1-22).
- UDYAM: 39,648,419 MSME registrations, 2020-2026.
Source data
Derived from 3 resources published on the Open Government Data Platform India (data.gov.in), snapshot May 2026, under the Government Open Data License - India. GODL-India does not cover personal information; this release holds only aggregates.
Providers: Ministry of Agriculture and Farmers Welfare, Ministry of Labour and Employment, Ministry of Micro, Small and Medium Enterprises.
- Ministry of Labour and Employment. "District-wise Demographic Data of Unorganised Workers registered on eShram as on previous day". data.gov.in resource 1d4d1c5a-feea-423b-be4f-c3636fdd1d82, https://api.data.gov.in/resource/1d4d1c5a-feea-423b-be4f-c3636fdd1d82.
- Ministry of Micro, Small and Medium Enterprises. "List of MSME Registered Units under UDYAM". data.gov.in resource 8b68ae56-84cf-4728-a0a6-1be11028dea7, https://api.data.gov.in/resource/8b68ae56-84cf-4728-a0a6-1be11028dea7.
- Ministry of Agriculture and Farmers Welfare. "Village and Gender - wise Beneficiaries Count under PM - KISAN Scheme". data.gov.in resource 388208c6-d82a-4190-90df-91aa2c326fec, https://api.data.gov.in/resource/388208c6-d82a-4190-90df-91aa2c326fec.
Original data: published by Ministry of Labour and Employment (eShram), Ministry of Micro, Small and Medium Enterprises (UDYAM) and Ministry of Agriculture and Farmers Welfare (PM-KISAN) on the Open Government Data Platform India (data.gov.in) and, where marked, NITI Aayog's National Data and Analytics Platform (NDAP), under the Government Open Data License - India (GODL-India). The original public data remains under GODL-India; the CC BY-NC-SA 4.0 licence of this repository covers JudicialMind's compilation, derived fields, annotations, benchmarks and models.
Citation
@misc{judicialmind2026workforce,
title={Bharat-Workforce-Atlas: District-level aggregates of India's eShram, PM-KISAN and UDYAM registries},
author={JudicialMind},
year={2026},
publisher={Hugging Face},
howpublished={\url{https://huggingface.co/datasets/judicialmind/Bharat-Workforce-Atlas}}
}
License
CC BY-NC-SA 4.0 (see LICENSE): research, academic, personal and other non-commercial use, with attribution to JudicialMind and to the original data publishers; adaptations must be shared under the same licence. Commercial use requires a separate licence.
Commercial use: contact surya@judicialmind.ai or hello@judicialmind.ai.
- Downloads last month
- 22