---
license: cc-by-4.0
language:
- en
- fr
pretty_name: skycmd commands (MASSIVE intents, en + fr)
task_categories:
- text-classification
tags:
- skycmd
- typed-decisions
- system-one
configs:
- config_name: massive
  data_files: data/massive.parquet
---

# skycmd commands (MASSIVE intents, en + fr)

MASSIVE intents in English and French as command records (`{src, lang, user, intent, split}`), official splits.

Part of [skycmd](https://github.com/maxgfr/skycmd), tiny decision models trained from scratch in Rust that run in the
browser. Written by `sky-data fetch-commands` at commit [`6cbcce8c`](https://github.com/maxgfr/skycmd/commit/6cbcce8cbbbf27e8e952e61b6cb1ed03ee1fca58).
Licences of every set: [DATA_LICENSES.md](https://github.com/maxgfr/skycmd/blob/main/DATA_LICENSES.md).


## Source and licence

MASSIVE (FitzGerald et al., 2022), CC-BY-4.0, via mteb/amazon_massive_intent. The conversion is ours.

## Files

| file | rows |
|---|---:|
| `data/massive.parquet` | 32,341 |

```python
from datasets import load_dataset
ds = load_dataset("maxgfr/skycmd-commands", "massive")
```

skycmd wrote no row with a language model: the targets come from the upstream labels (some upstream
sets carry their own teacher soft labels, see their cards), the games' oracles or our generators.