# Data Loading with FiftyOneTorchDataset This recipe introduces [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset), which lets you turn any FiftyOne dataset or view directly into a PyTorch-compatible dataset — no data copying, no format conversion. Specifically, it covers: - Loading a dataset from the [Dataset Zoo](https://docs.voxel51.com/user_guide/dataset_zoo/index.html) and creating a view - Converting a FiftyOne view to a [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset) with a custom `get_item` function - Visualizing samples from the dataset - Wrapping the dataset in a `torch.utils.data.DataLoader` for use in a training loop **API references:** [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset) · [GetItem](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.GetItem) ## Setup If you haven’t already, install FiftyOne: In this tutorial, we’ll use [PyTorch](https://pytorch.org/) for working with tensors and inspecting sample data. To follow along, you’ll need to install `torch` and `torchvision`, if necessary: This recipe requires a helper file, `utils.py`, which contains reusable functions for building `get_item` methods and creating dataloaders. The following cell downloads it into your working directory. ## Import Libraries ## Load Dataset ## Curate Your View One of the key advantages of [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset) is that you can pass any FiftyOne *view* — not just a full dataset. Here we take a 100-sample subset, but any filter, sort, or tag expression works equally well. ## Converting to a FiftyOneTorchDataset To convert a view to a [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset), call `.to_torch(get_item)` on any dataset or view. The `get_item` argument is a callable that receives a `fiftyone.core.sample.Sample` and returns whatever your model expects. To best understand what’s happening, start with the identity function — this shows you exactly what [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset) hands to your `get_item` callable: The `get_item` callable can return anything — here’s a minimal example that returns just the sample ID: ## Writing a Real get_item Here’s a detection-ready `get_item` that converts FiftyOne bounding boxes to `tv_tensors.BoundingBoxes` and applies augmentations: ## Visualizing Samples This is also a good place to debug your `get_item` function before hooking it into a training loop: ## Creating a DataLoader [FiftyOneTorchDataset](https://docs.voxel51.com/api/fiftyone.utils.torch.html#fiftyone.utils.torch.FiftyOneTorchDataset) is fully compatible with `torch.utils.data.DataLoader`, including multi-worker loading. The only required addition is `worker_init_fn=FiftyOneTorchDataset.worker_init`, which lets each worker process open its own FiftyOne database connection: ```python def simple_collate_fn(batch): return tuple(zip(*batch)) dataloader = torch.utils.data.DataLoader( torch_dataset, batch_size=5, shuffle=True, num_workers=2, worker_init_fn=FiftyOneTorchDataset.worker_init, collate_fn=simple_collate_fn, ) ``` The cells below use `utils.create_dataloader_simple()`, which wraps this same pattern: