AgentPublic/data-gouv-datasets-catalog

---

📢 Sondage 2026 : Utilisation des datasets publiques de MediaTech

Vous utilisez ce dataset ou d’autres datasets de notre collection MediaTech ? Votre avis compte ! Aidez-nous à améliorer nos datasets publiques en répondant à ce sondage rapide (5 min) : 👉 https://grist.numerique.gouv.fr/o/albert/forms/gF4hLaq9VvUog6c5aVDuMw/11 Merci pour votre contribution ! 🙌

---

🇫🇷 Data.gouv.fr Datasets Catalog

This dataset contains a processed and embedded version of the catalog of datasets published on data.gouv.fr, the French open data platform. The dataset was published by data.gouv.fr on the dedicated dataset page.

It includes rich metadata about each public dataset: title, URL, publisher organization, description, tags, licensing, update frequency, usage metrics, and more. The dataset provides semantic-ready and structured for semantic indexing and retrieval.

Each chunk has been embedded using the BAAI/bge-m3 model, making this catalog ready for search, classification, or retrieval-augmented generation (RAG) pipelines.

🗂️ Dataset Contents

The dataset is provided in Parquet format and contains the following columns:

Column Name	Type	Description
chunk_id	str	Unique source based identifier of the chunk
doc_id	str	Document identifier from the source site (slug).
chunk_xxh64	str	XXH64 hash of the chunk_text value.
title	str	Title of the dataset.
acronym	str	Acronym of the dataset (if available).
url	str	URL of the dataset on data.gouv.fr.
organization	str	Name of the associated organization.
organization_id	str	Unique ID of the organization.
owner	str	Name of the dataset owner.
owner_id	str	Unique ID of the dataset owner.
description	str	Full description of the dataset.
frequency	str	Update frequency of the dataset.
license	str	License type (e.g. Etalab-2.0).
temporal_coverage_start	str	Start of the temporal coverage (if applicable).
temporal_coverage_end	str	End of the temporal coverage (if applicable).
spatial_granularity	str	Spatial granularity level (e.g. country, region, etc.).
spatial_zones	str	Names of the spatial zone covered.
featured	bool	Whether the dataset is featured.
created_at	str	Dataset creation date (standard ISO 8601).
last_modified	str	Last modification date (standard ISO 8601).
tags	str	Comma-separated list of tags associated with the dataset.
archived	str	Whether the dataset is archived.
resources_count	int	Total number of attached resources.
main_resources_count	int	Number of primary resources.
resources_formats	str	Formats used by the dataset's ressources (e.g., CSV, JSON, PDF)
harvest_backend	str	Name of the harvest backend.
harvest_domain	str	Domain source of the harvested dataset.
harvest_created_at	str	Harvest creation date (standard ISO 8601).
harvest_modified_at	str	Harvest last update date (standard ISO 8601).
harvest_remote_url	str	Remote source URL of the harvested dataset.
quality_score	float	Quality score assigned by data.gouv.fr.
metric_discussions	int	Number of discussions related to the dataset.
metric_reuses	int	Number of declared reuses.
metric_reuses_by_months	str	Monthly reuse statistics (as JSON string).
metric_followers	int	Number of users following the dataset.
metric_followers_by_months	str	Monthly follower statistics (as JSON string or number).
metric_views	int	Number of views.
metric_resources_downloads	float	Number of resource downloads.
chunk_text	str	Text used for semantic embedding (title + organization + cropped description).
embeddings_bge-m3	str (stringified list)	Embedding of chunk_text using BAAI/bge-m3. Stored as JSON array string.

---

🛠️ Data Processing Methodology

📥 1. Field Extraction

The original dataset was retrieved directly from the official dedicated dataset page. This dataset only includes data.gouv.fr datasets that have at least a 100 characters description to remove as much noise as possible from incomplete datasets.

✂️ 2. Text Chunking

The chunk_text field was created by combining the title, organization name, description.

The description was here cropped to a maximum length of +- 1000 characters. The Langchain's RecursiveCharacterTextSplitter function was used to crop the description.

The parameters used are :

• chunk_size = 1000 (in order to limit as much noise as possible)
• length_function = len

Then, only the first splitted text was keeped. Which leads to have a cropped description of a maximum of +- 1000 characters.

🧠 3. Embedding Generation

Each chunk_text was embedded using the BAAI/bge-m3 model. The resulting embedding vector is stored in the embeddings_bge-m3 column as a string, but can easily be parsed back into a list[float] or NumPy array.

🎓 Tutorials

🔄 1. The chunking doesn't fit your use case?

If you need to reconstitute the original, un-chunked dataset, you can follow this tutorial notebook available on our GitHub repository.

⚠️ The tutorial is only relevant for datasets that were chunked without overlap.

🤖 2. How to load MediaTech's datasets from Hugging Face and use them in a RAG pipeline ?

To learn how to load MediaTech's datasets from Hugging Face and integrate them into a Retrieval-Augmented Generation (RAG) pipeline, check out our step-by-step RAG tutorial available on our GitHub repository !

📌 3. Embedding Use Notice

⚠️ The embeddings_bge-m3 column is stored as a stringified list of floats (e.g., "[-0.03062629,-0.017049594,...]"). To use it as a vector, you need to parse it into a list of floats or NumPy array.

Using the datasets library:

import pandas as pd
import json
from datasets import load_dataset
# The Pyarrow library must be installed in your Python environment for this example. By doing => pip install pyarrow

dataset = load_dataset("AgentPublic/data-gouv-datasets-catalog")
df = pd.DataFrame(dataset['train'])
df["embeddings_bge-m3"] = df["embeddings_bge-m3"].apply(json.loads)

Using downloaded local Parquet files:

import pandas as pd
import json
# The Pyarrow library must be installed in your Python environment for this example. By doing => pip install pyarrow

df = pd.read_parquet(path="data-gouv-datasets-catalog-latest/") # Assuming that all parquet files are located into this folder
df["embeddings_bge-m3"] = df["embeddings_bge-m3"].apply(json.loads)

You can then use the dataframe as you wish, such as by inserting the data from the dataframe into the vector database of your choice.

🐱 GitHub repository :

The project MediaTech is open source ! You are free to contribute or see the complete code used to build the dataset by checking the GitHub repository

📚 Source & License

🔗 Source :

• Data.gouv.fr : Catalogue des données de data.gouv.fr

📄 Licence :

Open License (Etalab) — This dataset is publicly available and can be reused under the conditions of the Etalab open license.