Resources¶
Every dataset used in class, in one place. If your file from last session is missing, broken or on another computer, fetch it again from here.
Right-click a file name and choose Save link as... — or just click it. Then, in Superset: + → Data → Upload CSV to database, database Dades classe, schema public.
Class datasets¶
Files hosted on this site. They are the exact copies we use in class, so your results will match the ones on the board.
| File | What is inside | Rows | Size | Used in | Original source |
|---|---|---|---|---|---|
pokemon_df.csv |
One row per Pokémon: types, height, weight, base stats, generation. Upload it as the table pokemon. |
949 | 204 KB | UT1 · RA1 |
TidyTuesday 2025-04-01 |
generations.csv |
One row per Pokémon generation: region, first game, release year. The second table for the JOIN session. |
8 | 1 KB | UT1 · RA1 |
Written for this course |
video_games.csv |
Steam games: release date, price, owners, developer, publisher, playtime, Metascore. Deliberately left dirty. Upload it as the table games. |
26,688 | 2.6 MB | UT1 practical · UT2 · RA2 |
TidyTuesday 2019-07-30 |
anscombe.xlsx |
Anscombe's quartet (1973): four tiny datasets with almost the same statistics and completely different charts. | 4 × 11 | 11 KB | First session | Anscombe, Graphs in Statistical Analysis |
api_to_csv.py |
A short Python script: asks the PokéAPI for 20 Pokémon and writes pokeapi_sample.csv. Run it with python3 api_to_csv.py. |
— | 1 KB | UT1 · RA1 c) |
Written for this course |
Open a CSV in a text editor first
Before you upload a file, open it in a text editor, not in a spreadsheet. You will see what the tool will see: the separator (, or ;), the decimal mark and the header row. Most failed uploads are a wrong separator.
Where to find more data¶
For your project you choose your own dataset. These are the sources we use in class. All of them are free and none needs an account.
| Source | What you get | Formats | Watch out for |
|---|---|---|---|
| TidyTuesday | A new public dataset every week, on everyday topics: music, games, films, sport. Each one lives in data/<year>/<date>/. |
CSV | Open the .csv on GitHub and press Download raw file. |
| GOIB open-data portal | Data published by the Government of the Balearic Islands: tourism, transport, environment, education. | CSV, JSON, XLSX | Files often use ; as separator and a decimal comma. |
| IBESTAT | The official statistics institute of the Balearic Islands: population, economy, labour market. | CSV, XLSX | Statistical tables are made for reading. Check the header rows before you upload one. |
| datos.gob.es | The Spanish national open-data catalogue. It links to datasets from ministries, regions and town councils. | Varies | Quality varies a lot from one publisher to another. |
| PokéAPI | An API: you ask a web address and it answers in JSON. Try /api/v2/pokemon?limit=20 in your browser. |
JSON | Superset cannot read an API directly. Extract it to a CSV first — that is what api_to_csv.py does. |
TidyTuesday picks¶
Four datasets that work well for a first project:
| Dataset | Folder | File |
|---|---|---|
| Pokémon | data/2025/2025-04-01 |
pokemon_df.csv — types, stats, height, weight |
| Spotify songs | data/2020/2020-01-21 |
spotify_songs.csv — popularity, danceability, energy, tempo |
| Taylor Swift | data/2023/2023-10-17 |
taylor_album_songs.csv — every album track with audio features |
| Steam games | data/2021/2021-03-16 |
games.csv — monthly average and peak players per game |
Open data is free, not lawless
Open data is data published so that anyone can download, reuse and share it. It still comes with a licence that says how. Many only ask you to credit the source. Before you use a dataset in your project, find its licence and write the source in your report.
Tools¶
| Tool | What we use it for | Reference |
|---|---|---|
| Apache Superset | The main tool: connections, SQL Lab, datasets, charts and dashboards. It runs on your class VM. | superset.apache.org/docs |
| PostgreSQL | The database where your tables live. | postgresql.org/docs |
| Power BI | The commercial tool we compare Superset with. Power BI Desktop needs Windows. | learn.microsoft.com/power-bi |
Not sure which outcome a dataset belongs to? The Learning outcomes page maps every RA to its units.