> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://fr.nvidia-localization.ferndocs.com/fr/nemo/rl/latest/nemo-rl-data-hf-datasets/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://fr.nvidia-localization.ferndocs.com/_mcp/server.

# nemo\_rl.data.hf\_datasets

## Sous-modules

* **[`prompt_response_dataset`](nemo-rl-data-hf-datasets-prompt-response-dataset)**
* **[`helpsteer3`](nemo-rl-data-hf-datasets-helpsteer3)**
* **[`clevr`](nemo-rl-data-hf-datasets-clevr)**
* **[`geometry3k`](nemo-rl-data-hf-datasets-geometry3k)**
* **[`preference_dataset`](nemo-rl-data-hf-datasets-preference-dataset)**
* **[`tulu3`](nemo-rl-data-hf-datasets-tulu3)**
* **[`squad`](nemo-rl-data-hf-datasets-squad)**
* **[`oasst`](nemo-rl-data-hf-datasets-oasst)**
* **[`chat_templates`](nemo-rl-data-hf-datasets-chat-templates)**
* **[`deepscaler`](nemo-rl-data-hf-datasets-deepscaler)**
* **[`openmathinstruct2`](nemo-rl-data-hf-datasets-openmathinstruct2)**
* **[`dpo`](nemo-rl-data-hf-datasets-dpo)**
* **[`refcoco`](nemo-rl-data-hf-datasets-refcoco)**
* **[`oai_format_dataset`](nemo-rl-data-hf-datasets-oai-format-dataset)**

## Contenu du module

### Données

`__all__`

### API

```python
nemo_rl.data.hf_datasets.__all__
```

**Valeur**: `['DPODataset', 'HelpSteer3Dataset', 'OasstDataset', 'OpenAIFormatDataset', 'OpenMathInstruct2Dataset', 'PreferenceDataset', 'PromptResponseDataset', 'SquadDataset', 'Tulu3PreferenceDataset', 'COMMON_CHAT_TEMPLATES', 'CLEVRCoGenTDataset']`