YOLOv5

A very fast and easy to use PyTorch model that achieves state of the art (or near state of the art) results.

Overview

YOLOv5 is Here

YOLOv5 was released by Glenn Jocher on June 9, 2020. It follows the recent releases of YOLOv4 (April 23, 2020) and EfficientDet (March 18, 2020).

YOLOv5 Performance

YOLOv5 is smaller and generally easier to use in production. Given it is natively implemented in PyTorch (rather than Darknet), modifying the architecture and exporting to many deploy environments is straightforward.

SIZE: YOLOv5s is about 88% smaller than big-YOLOv4 (27 MB vs 244 MB)

SPEED: YOLOv5 performs batch inference at about 140 FPS by default.
ACCURACY: YOLOv5 is roughly as accurate as YOLOv4 on small tasks (0.895 mAP vs 0.892 mAP on BCCD). On larger tasks like COCO, YOLOv4 is more performant.

Using YOLOv5

We've written both a YOLOv5 tutorial and YOLOv5 Colab notebook for training YOLOv5 on your own custom data.

2022 YOLOv5 releases Classification and Instance Segmentation

YOLOv5 launched supporting bounding boxes for object detection. Now you can use YOLOv5 for classification and instance segmentation as well.

‍

Performance

Model	size ^(pixels)	mAP^box 50-95	mAP^mask 50-95	Train time ^{300 epochs A100 (hours)}	Speed ^{ONNX CPU (ms)}	Speed ^{TRT A100 (ms)}	params ^(M)	FLOPs ^{@640 (B)}
YOLOv5n-seg	640	27.6	23.4	80:17	62.7	1.2	2.0	7.1
YOLOv5s-seg	640	37.6	31.7	88:16	173.3	1.4	7.6	26.4
YOLOv5m-seg	640	45.0	37.1	108:36	427.0	2.2	22.0	70.8
YOLOv5l-seg	640	49.0	39.9	66:43 (2x)	857.4	2.9	47.9	147.7
YOLOv5x-seg	640	50.7	41.4	62:56 (3x)	1579.2	4.5	88.8	265.7

Use This Model

You can run fine-tuned YOLOv5 object detection models with Inference.

First, install Inference:‍


pip install inference

Retrieve your Roboflow API key and save it in an environment variable called ROBOFLOW_API_KEY:


export ROBOFLOW_API_KEY="your-api-key"

To use your model, run the following code:


import inference
model = inference.load_roboflow_model("model-name/version")
results = model.infer(image="YOUR_IMAGE.jpg")

Above, replace:

YOUR_IMAGE.jpg with the path to your image.
model_id/version with the YOLOv5 model ID and version you want to use. Learn how to retrieve your model and version ID

Label Data Automatically with YOLOv5

You can automatically label a dataset using YOLOv5 with help from Autodistill, an open source package for training computer vision models. You can label a folder of images automatically with only a few lines of code. Below, see our tutorials that demonstrate how to use YOLOv5 to train a computer vision model.

OWL-ViT and YOLOv5

Grounding DINO and YOLOv5

SAM-CLIP and YOLOv5

Grounded SAM and YOLOv5