Super Gradients Integration ¶

FiftyOne integrates natively with Deci AI’s SuperGradients library, so you can run inference with YOLO-NAS architectures on your FiftyOne datasets with just a few lines of code!

Setup ¶

To get started with SuperGradients, just install the super-gradients package:

pip install super-gradients

Inference ¶

You can directly pass SuperGradients YOLO-NAS models to your FiftyOne dataset’s apply_model() method:

import fiftyone as fo
import fiftyone.zoo as foz

from super_gradients.training import models

dataset = foz.load_zoo_dataset("quickstart", max_samples=25)
dataset.select_fields().keep_fields()

model = models.get("yolo_nas_m", pretrained_weights="coco")
# model = models.get("yolo_nas_l", pretrained_weights="coco")
# model = models.get("yolo_nas_s", pretrained_weights="coco")

dataset.apply_model(model, label_field="yolo_nas", confidence_thresh=0.7)

session = fo.launch_app(dataset)

Model zoo ¶

SuperGradients YOLO-NAS is also available directly from the FiftyOne Model Zoo!

import fiftyone as fo
import fiftyone.zoo as foz

model = foz.load_zoo_model("yolo-nas-torch")

dataset = foz.load_zoo_dataset("quickstart")
dataset.apply_model(model, label_field="yolo_nas")

session = fo.launch_app(dataset)