[中文](https://docs.ultralytics.com/zh) | [한국어](https://docs.ultralytics.com/ko) | [日本語](https://docs.ultralytics.com/ja) | [Русский](https://docs.ultralytics.com/ru) | [Deutsch](https://docs.ultralytics.com/de) | [Français](https://docs.ultralytics.com/fr) | [Español](https://docs.ultralytics.com/es) | [Português](https://docs.ultralytics.com/pt) | [Türkçe](https://docs.ultralytics.com/tr) | [Tiếng Việt](https://docs.ultralytics.com/vi) | [العربية](https://docs.ultralytics.com/ar)

Open In Kaggle

Ultralytics YOLOv5 🚀 is a cutting-edge, state-of-the-art (SOTA) computer vision model developed by [Ultralytics](https://www.ultralytics.com/). Based on the [PyTorch](https://pytorch.org/) framework, YOLOv5 is renowned for its ease of use, speed, and accuracy. It incorporates insights and best practices from extensive research and development, making it a popular choice for a wide range of vision AI tasks, including [object detection](https://docs.ultralytics.com/tasks/detect/), [image segmentation](https://docs.ultralytics.com/tasks/segment/), and [image classification](https://docs.ultralytics.com/tasks/classify/). We hope the resources here help you get the most out of YOLOv5. Please browse the [YOLOv5 Docs](https://docs.ultralytics.com/yolov5/) for detailed information, raise an issue on [GitHub](https://github.com/ultralytics/yolov5/issues/new/choose) for support, and join our [Discord community](https://discord.com/invite/ultralytics) for questions and discussions! To request an Enterprise License, please complete the form at [Ultralytics Licensing](https://www.ultralytics.com/license).

## 🚀 YOLO11: The Next Evolution We are excited to announce the launch of **Ultralytics YOLO11** 🚀, the latest advancement in our state-of-the-art (SOTA) vision models! Available now at the [Ultralytics YOLO GitHub repository](https://github.com/ultralytics/ultralytics), YOLO11 builds on our legacy of speed, precision, and ease of use. Whether you're tackling [object detection](https://docs.ultralytics.com/tasks/detect/), [instance segmentation](https://docs.ultralytics.com/tasks/segment/), [pose estimation](https://docs.ultralytics.com/tasks/pose/), [image classification](https://docs.ultralytics.com/tasks/classify/), or [oriented object detection (OBB)](https://docs.ultralytics.com/tasks/obb/), YOLO11 delivers the performance and versatility needed to excel in diverse applications. Get started today and unlock the full potential of YOLO11! Visit the [Ultralytics Docs](https://docs.ultralytics.com/) for comprehensive guides and resources: [![PyPI version](https://badge.fury.io/py/ultralytics.svg)](https://badge.fury.io/py/ultralytics) [![Downloads](https://static.pepy.tech/badge/ultralytics)](https://www.pepy.tech/projects/ultralytics) ```bash # Install the ultralytics package pip install ultralytics ```

Ultralytics YOLO Performance Comparison

## 📚 Documentation See the [YOLOv5 Docs](https://docs.ultralytics.com/yolov5/) for full documentation on training, testing, and deployment. See below for quickstart examples.

Install

Clone the repository and install dependencies in a [**Python>=3.8.0**](https://www.python.org/) environment. Ensure you have [**PyTorch>=1.8**](https://pytorch.org/get-started/locally/) installed. ```bash # Clone the YOLOv5 repository git clone https://github.com/ultralytics/yolov5 # Navigate to the cloned directory cd yolov5 # Install required packages pip install -r requirements.txt ```

Inference with PyTorch Hub

Use YOLOv5 via [PyTorch Hub](https://docs.ultralytics.com/yolov5/tutorials/pytorch_hub_model_loading/) for inference. [Models](https://github.com/ultralytics/yolov5/tree/master/models) are automatically downloaded from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). ```python import torch # Load a YOLOv5 model (options: yolov5n, yolov5s, yolov5m, yolov5l, yolov5x) model = torch.hub.load("ultralytics/yolov5", "yolov5s") # Default: yolov5s # Define the input image source (URL, local file, PIL image, OpenCV frame, numpy array, or list) img = "https://ultralytics.com/images/zidane.jpg" # Example image # Perform inference (handles batching, resizing, normalization automatically) results = model(img) # Process the results (options: .print(), .show(), .save(), .crop(), .pandas()) results.print() # Print results to console results.show() # Display results in a window results.save() # Save results to runs/detect/exp ```

Inference with detect.py

The `detect.py` script runs inference on various sources. It automatically downloads [models](https://github.com/ultralytics/yolov5/tree/master/models) from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases) and saves the results to the `runs/detect` directory. ```bash # Run inference using a webcam python detect.py --weights yolov5s.pt --source 0 # Run inference on a local image file python detect.py --weights yolov5s.pt --source img.jpg # Run inference on a local video file python detect.py --weights yolov5s.pt --source vid.mp4 # Run inference on a screen capture python detect.py --weights yolov5s.pt --source screen # Run inference on a directory of images python detect.py --weights yolov5s.pt --source path/to/images/ # Run inference on a text file listing image paths python detect.py --weights yolov5s.pt --source list.txt # Run inference on a text file listing stream URLs python detect.py --weights yolov5s.pt --source list.streams # Run inference using a glob pattern for images python detect.py --weights yolov5s.pt --source 'path/to/*.jpg' # Run inference on a YouTube video URL python detect.py --weights yolov5s.pt --source 'https://youtu.be/LNwODJXcvt4' # Run inference on an RTSP, RTMP, or HTTP stream python detect.py --weights yolov5s.pt --source 'rtsp://example.com/media.mp4' ```

Training

The commands below demonstrate how to reproduce YOLOv5 [COCO dataset](https://docs.ultralytics.com/datasets/detect/coco/) results. Both [models](https://github.com/ultralytics/yolov5/tree/master/models) and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) are downloaded automatically from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). Training times for YOLOv5n/s/m/l/x are approximately 1/2/4/6/8 days on a single [NVIDIA V100 GPU](https://www.nvidia.com/en-us/data-center/v100/). Using [Multi-GPU training](https://docs.ultralytics.com/yolov5/tutorials/multi_gpu_training/) can significantly reduce training time. Use the largest `--batch-size` your hardware allows, or use `--batch-size -1` for YOLOv5 [AutoBatch](https://github.com/ultralytics/yolov5/pull/5092). The batch sizes shown below are for V100-16GB GPUs. ```bash # Train YOLOv5n on COCO for 300 epochs python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5n.yaml --batch-size 128 # Train YOLOv5s on COCO for 300 epochs python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5s.yaml --batch-size 64 # Train YOLOv5m on COCO for 300 epochs python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5m.yaml --batch-size 40 # Train YOLOv5l on COCO for 300 epochs python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5l.yaml --batch-size 24 # Train YOLOv5x on COCO for 300 epochs python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5x.yaml --batch-size 16 ``` YOLOv5 Training Results

YOLOv5 Training Results

Tutorials

- **[Train Custom Data](https://docs.ultralytics.com/yolov5/tutorials/train_custom_data/)** 🚀 **RECOMMENDED**: Learn how to train YOLOv5 on your own datasets. - **[Tips for Best Training Results](https://docs.ultralytics.com/guides/model-training-tips/)** ☘️: Improve your model's performance with expert tips. - **[Multi-GPU Training](https://docs.ultralytics.com/yolov5/tutorials/multi_gpu_training/)**: Speed up training using multiple GPUs. - **[PyTorch Hub Integration](https://docs.ultralytics.com/yolov5/tutorials/pytorch_hub_model_loading/)** 🌟 **NEW**: Easily load models using PyTorch Hub. - **[Model Export (TFLite, ONNX, CoreML, TensorRT)](https://docs.ultralytics.com/yolov5/tutorials/model_export/)** 🚀: Convert your models to various deployment formats like [ONNX](https://onnx.ai/) or [TensorRT](https://developer.nvidia.com/tensorrt). - **[NVIDIA Jetson Deployment](https://docs.ultralytics.com/guides/nvidia-jetson/)** 🌟 **NEW**: Deploy YOLOv5 on [NVIDIA Jetson](https://developer.nvidia.com/embedded-computing) devices. - **[Test-Time Augmentation (TTA)](https://docs.ultralytics.com/yolov5/tutorials/test_time_augmentation/)**: Enhance prediction accuracy with TTA. - **[Model Ensembling](https://docs.ultralytics.com/yolov5/tutorials/model_ensembling/)**: Combine multiple models for better performance. - **[Model Pruning/Sparsity](https://docs.ultralytics.com/yolov5/tutorials/model_pruning_and_sparsity/)**: Optimize models for size and speed. - **[Hyperparameter Evolution](https://docs.ultralytics.com/yolov5/tutorials/hyperparameter_evolution/)**: Automatically find the best training hyperparameters. - **[Transfer Learning with Frozen Layers](https://docs.ultralytics.com/yolov5/tutorials/transfer_learning_with_frozen_layers/)**: Adapt pretrained models to new tasks efficiently using [transfer learning](https://www.ultralytics.com/glossary/transfer-learning). - **[Architecture Summary](https://docs.ultralytics.com/yolov5/tutorials/architecture_description/)** 🌟 **NEW**: Understand the YOLOv5 model architecture. - **[Ultralytics HUB Training](https://www.ultralytics.com/hub)** 🚀 **RECOMMENDED**: Train and deploy YOLO models using Ultralytics HUB. - **[ClearML Logging](https://docs.ultralytics.com/yolov5/tutorials/clearml_logging_integration/)**: Integrate with [ClearML](https://clear.ml/) for experiment tracking. - **[Neural Magic DeepSparse Integration](https://docs.ultralytics.com/yolov5/tutorials/neural_magic_pruning_quantization/)**: Accelerate inference with DeepSparse. - **[Comet Logging](https://docs.ultralytics.com/yolov5/tutorials/comet_logging_integration/)** 🌟 **NEW**: Log experiments using [Comet ML](https://www.comet.com/site/).

## 🧩 Integrations Our key integrations with leading AI platforms extend the functionality of Ultralytics' offerings, enhancing tasks like dataset labeling, training, visualization, and model management. Discover how Ultralytics, in collaboration with partners like [Weights & Biases](https://docs.ultralytics.com/integrations/weights-biases/), [Comet ML](https://docs.ultralytics.com/integrations/comet/), [Roboflow](https://docs.ultralytics.com/integrations/roboflow/), and [Intel OpenVINO](https://docs.ultralytics.com/integrations/openvino/), can optimize your AI workflow. Explore more at [Ultralytics Integrations](https://docs.ultralytics.com/integrations/).

Ultralytics active learning integrations

| Ultralytics HUB 🌟 | Weights & Biases | Comet | Neural Magic | | :-----------------------------------------------------------------------------------------------------------------------------: | :---------------------------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------------: | | Streamline YOLO workflows: Label, train, and deploy effortlessly with [Ultralytics HUB](https://hub.ultralytics.com/). Try now! | Track experiments, hyperparameters, and results with [Weights & Biases](https://docs.ultralytics.com/integrations/weights-biases/). | Free forever, [Comet ML](https://docs.ultralytics.com/integrations/comet/) lets you save YOLO models, resume training, and interactively visualize predictions. | Run YOLO inference up to 6x faster with [Neural Magic DeepSparse](https://docs.ultralytics.com/integrations/neural-magic/). | ## ⭐ Ultralytics HUB Experience seamless AI development with [Ultralytics HUB](https://www.ultralytics.com/hub) ⭐, the ultimate platform for building, training, and deploying [computer vision](https://www.ultralytics.com/glossary/computer-vision-cv) models. Visualize datasets, train [YOLOv5](https://docs.ultralytics.com/models/yolov5/) and [YOLOv8](https://docs.ultralytics.com/models/yolov8/) 🚀 models, and deploy them to real-world applications without writing any code. Transform images into actionable insights using our cutting-edge tools and user-friendly [Ultralytics App](https://www.ultralytics.com/app-install). Start your journey for **Free** today!

Ultralytics HUB Platform Screenshot

## 🤔 Why YOLOv5? YOLOv5 is designed for simplicity and ease of use. We prioritize real-world performance and accessibility.

YOLOv5 Performance Chart

YOLOv5-P5 640 Figure

YOLOv5 P5 640 Performance Chart

Figure Notes

- **COCO AP val** denotes the [mean Average Precision (mAP)](https://www.ultralytics.com/glossary/mean-average-precision-map) at [Intersection over Union (IoU)](https://www.ultralytics.com/glossary/intersection-over-union-iou) thresholds from 0.5 to 0.95, measured on the 5,000-image [COCO val2017 dataset](https://docs.ultralytics.com/datasets/detect/coco/) across various inference sizes (256 to 1536 pixels). - **GPU Speed** measures the average inference time per image on the [COCO val2017 dataset](https://docs.ultralytics.com/datasets/detect/coco/) using an [AWS p3.2xlarge V100 instance](https://aws.amazon.com/ec2/instance-types/p4/) with a batch size of 32. - **EfficientDet** data is sourced from the [google/automl repository](https://github.com/google/automl) at batch size 8. - **Reproduce** these results using the command: `python val.py --task study --data coco.yaml --iou 0.7 --weights yolov5n6.pt yolov5s6.pt yolov5m6.pt yolov5l6.pt yolov5x6.pt`

### Pretrained Checkpoints This table shows the performance metrics for various YOLOv5 models trained on the COCO dataset. | Model | Size
^{(pixels) | mAP^{val
50-95 | mAP^{val
50 | Speed
^{CPU b1
(ms) | Speed
^{V100 b1
(ms) | Speed
^{V100 b32
(ms) | Params
^{(M) | FLOPs
^{@640 (B) |
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | --------------------- | -------------------- | ----------------- | ---------------------------- | ----------------------------- | ------------------------------ | ------------------ | ---------------------- |
| [YOLOv5n](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5n.pt) | 640 | 28.0 | 45.7 | **45** | **6.3** | **0.6** | **1.9** | **4.5** |
| [YOLOv5s](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s.pt) | 640 | 37.4 | 56.8 | 98 | 6.4 | 0.9 | 7.2 | 16.5 |
| [YOLOv5m](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5m.pt) | 640 | 45.4 | 64.1 | 224 | 8.2 | 1.7 | 21.2 | 49.0 |
| [YOLOv5l](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5l.pt) | 640 | 49.0 | 67.3 | 430 | 10.1 | 2.7 | 46.5 | 109.1 |
| [YOLOv5x](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5x.pt) | 640 | 50.7 | 68.9 | 766 | 12.1 | 4.8 | 86.7 | 205.7 |
| | | | | | | | | |
| [YOLOv5n6](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5n6.pt) | 1280 | 36.0 | 54.4 | 153 | 8.1 | 2.1 | 3.2 | 4.6 |
| [YOLOv5s6](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s6.pt) | 1280 | 44.8 | 63.7 | 385 | 8.2 | 3.6 | 12.6 | 16.8 |
| [YOLOv5m6](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5m6.pt) | 1280 | 51.3 | 69.3 | 887 | 11.1 | 6.8 | 35.7 | 50.0 |
| [YOLOv5l6](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5l6.pt) | 1280 | 53.7 | 71.3 | 1784 | 15.8 | 10.5 | 76.8 | 111.4 |
| [YOLOv5x6](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5x6.pt)
+ [[TTA]](https://docs.ultralytics.com/yolov5/tutorials/test_time_augmentation/) | 1280
1536 | 55.0
**55.8** | 72.7
**72.7** | 3136
- | 26.2
- | 19.4
- | 140.7
- | 209.8
- |

Table Notes

- All checkpoints were trained for 300 epochs using default settings. Nano (n) and Small (s) models use [hyp.scratch-low.yaml](https://github.com/ultralytics/yolov5/blob/master/data/hyps/hyp.scratch-low.yaml) hyperparameters, while Medium (m), Large (l), and Extra-Large (x) models use [hyp.scratch-high.yaml](https://github.com/ultralytics/yolov5/blob/master/data/hyps/hyp.scratch-high.yaml).
- **mAP^val** values represent single-model, single-scale performance on the [COCO val2017 dataset](https://docs.ultralytics.com/datasets/detect/coco/).
Reproduce using: `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65`
- **Speed** metrics are averaged over COCO val images using an [AWS p3.2xlarge V100 instance](https://aws.amazon.com/ec2/instance-types/p4/). Non-Maximum Suppression (NMS) time (~1 ms/image) is not included.
Reproduce using: `python val.py --data coco.yaml --img 640 --task speed --batch 1`
- **TTA** ([Test Time Augmentation](https://docs.ultralytics.com/yolov5/tutorials/test_time_augmentation/)) includes reflection and scale augmentations for improved accuracy.
Reproduce using: `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment`

## 🖼️ Segmentation

The YOLOv5 [release v7.0](https://github.com/ultralytics/yolov5/releases/v7.0) introduced [instance segmentation](https://docs.ultralytics.com/tasks/segment/) models that achieve state-of-the-art performance. These models are designed for easy training, validation, and deployment. For full details, see the [Release Notes](https://github.com/ultralytics/yolov5/releases/v7.0) and explore the [YOLOv5 Segmentation Colab Notebook](https://github.com/ultralytics/yolov5/blob/master/segment/tutorial.ipynb) for quickstart examples.

Segmentation Checkpoints

YOLOv5 segmentation models were trained on the [COCO dataset](https://docs.ultralytics.com/datasets/segment/coco/) for 300 epochs at an image size of 640 pixels using A100 GPUs. Models were exported to [ONNX](https://onnx.ai/) FP32 for CPU speed tests and [TensorRT](https://developer.nvidia.com/tensorrt) FP16 for GPU speed tests. All speed tests were conducted on Google [Colab Pro](https://colab.research.google.com/signup) notebooks for reproducibility.

| 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](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5n-seg.pt) | 640 | 27.6 | 23.4 | 80:17 | **62.7** | **1.2** | **2.0** | **7.1** |
| [YOLOv5s-seg](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s-seg.pt) | 640 | 37.6 | 31.7 | 88:16 | 173.3 | 1.4 | 7.6 | 26.4 |
| [YOLOv5m-seg](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5m-seg.pt) | 640 | 45.0 | 37.1 | 108:36 | 427.0 | 2.2 | 22.0 | 70.8 |
| [YOLOv5l-seg](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5l-seg.pt) | 640 | 49.0 | 39.9 | 66:43 (2x) | 857.4 | 2.9 | 47.9 | 147.7 |
| [YOLOv5x-seg](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5x-seg.pt) | 640 | **50.7** | **41.4** | 62:56 (3x) | 1579.2 | 4.5 | 88.8 | 265.7 |

- All checkpoints were trained for 300 epochs using the SGD optimizer with `lr0=0.01` and `weight_decay=5e-5` at an image size of 640 pixels, using default settings.
Training runs are logged at [https://wandb.ai/glenn-jocher/YOLOv5_v70_official](https://wandb.ai/glenn-jocher/YOLOv5_v70_official).
- **Accuracy** values represent single-model, single-scale performance on the COCO dataset.
Reproduce using: `python segment/val.py --data coco.yaml --weights yolov5s-seg.pt`
- **Speed** metrics are averaged over 100 inference images using a [Colab Pro A100 High-RAM instance](https://colab.research.google.com/signup). Values indicate inference speed only (NMS adds approximately 1ms per image).
Reproduce using: `python segment/val.py --data coco.yaml --weights yolov5s-seg.pt --batch 1`
- **Export** to ONNX (FP32) and TensorRT (FP16) was performed using `export.py`.
Reproduce using: `python export.py --weights yolov5s-seg.pt --include engine --device 0 --half`}}}}}}}}

Segmentation Usage Examples

### Train

YOLOv5 segmentation training supports automatic download of the [COCO128-seg dataset](https://docs.ultralytics.com/datasets/segment/coco8-seg/) via the `--data coco128-seg.yaml` argument. For the full [COCO-segments dataset](https://docs.ultralytics.com/datasets/segment/coco/), download it manually using `bash data/scripts/get_coco.sh --train --val --segments` and then train with `python train.py --data coco.yaml`.

```bash
# Train on a single GPU
python segment/train.py --data coco128-seg.yaml --weights yolov5s-seg.pt --img 640

# Train using Multi-GPU Distributed Data Parallel (DDP)
python -m torch.distributed.run --nproc_per_node 4 --master_port 1 segment/train.py --data coco128-seg.yaml --weights yolov5s-seg.pt --img 640 --device 0,1,2,3
```

### Val

Validate the mask [mean Average Precision (mAP)](https://www.ultralytics.com/glossary/mean-average-precision-map) of YOLOv5s-seg on the COCO dataset:

```bash
# Download COCO validation segments split (780MB, 5000 images)
bash data/scripts/get_coco.sh --val --segments

# Validate the model
python segment/val.py --weights yolov5s-seg.pt --data coco.yaml --img 640
```

### Predict

Use the pretrained YOLOv5m-seg.pt model to perform segmentation on `bus.jpg`:

```bash
# Run prediction
python segment/predict.py --weights yolov5m-seg.pt --source data/images/bus.jpg
```

```python
# Load model from PyTorch Hub (Note: Inference support might vary)
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5m-seg.pt")
```

| ![Zidane Segmentation Example](https://user-images.githubusercontent.com/26833433/203113421-decef4c4-183d-4a0a-a6c2-6435b33bc5d3.jpg) | ![Bus Segmentation Example](https://user-images.githubusercontent.com/26833433/203113416-11fe0025-69f7-4874-a0a6-65d0bfe2999a.jpg) |
| :-----------------------------------------------------------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------: |

### Export

Export the YOLOv5s-seg model to ONNX and TensorRT formats:

```bash
# Export model
python export.py --weights yolov5s-seg.pt --include onnx engine --img 640 --device 0
```

## 🏷️ Classification

YOLOv5 [release v6.2](https://github.com/ultralytics/yolov5/releases/v6.2) introduced support for [image classification](https://docs.ultralytics.com/tasks/classify/) model training, validation, and deployment. Check the [Release Notes](https://github.com/ultralytics/yolov5/releases/v6.2) for details and the [YOLOv5 Classification Colab Notebook](https://github.com/ultralytics/yolov5/blob/master/classify/tutorial.ipynb) for quickstart guides.

Classification Checkpoints

YOLOv5-cls classification models were trained on [ImageNet](https://docs.ultralytics.com/datasets/classify/imagenet/) for 90 epochs using a 4xA100 instance. [ResNet](https://arxiv.org/abs/1512.03385) and [EfficientNet](https://arxiv.org/abs/1905.11946) models were trained alongside under identical settings for comparison. Models were exported to [ONNX](https://onnx.ai/) FP32 (CPU speed tests) and [TensorRT](https://developer.nvidia.com/tensorrt) FP16 (GPU speed tests). All speed tests were run on Google [Colab Pro](https://colab.research.google.com/signup) for reproducibility.

| Model | Size
^{(pixels) | Acc
^{top1 | Acc
^{top5 | Training
^{90 epochs
4xA100 (hours) | Speed
^{ONNX CPU
(ms) | Speed
^{TensorRT V100
(ms) | Params
^{(M) | FLOPs
^{@224 (B) |
| -------------------------------------------------------------------------------------------------- | --------------------- | ---------------- | ---------------- | -------------------------------------------- | ------------------------------ | ----------------------------------- | ------------------ | ---------------------- |
| [YOLOv5n-cls](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5n-cls.pt) | 224 | 64.6 | 85.4 | 7:59 | **3.3** | **0.5** | **2.5** | **0.5** |
| [YOLOv5s-cls](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5s-cls.pt) | 224 | 71.5 | 90.2 | 8:09 | 6.6 | 0.6 | 5.4 | 1.4 |
| [YOLOv5m-cls](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5m-cls.pt) | 224 | 75.9 | 92.9 | 10:06 | 15.5 | 0.9 | 12.9 | 3.9 |
| [YOLOv5l-cls](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5l-cls.pt) | 224 | 78.0 | 94.0 | 11:56 | 26.9 | 1.4 | 26.5 | 8.5 |
| [YOLOv5x-cls](https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5x-cls.pt) | 224 | **79.0** | **94.4** | 15:04 | 54.3 | 1.8 | 48.1 | 15.9 |
| | | | | | | | | |
| [ResNet18](https://github.com/ultralytics/yolov5/releases/download/v7.0/resnet18.pt) | 224 | 70.3 | 89.5 | **6:47** | 11.2 | 0.5 | 11.7 | 3.7 |
| [ResNet34](https://github.com/ultralytics/yolov5/releases/download/v7.0/resnet34.pt) | 224 | 73.9 | 91.8 | 8:33 | 20.6 | 0.9 | 21.8 | 7.4 |
| [ResNet50](https://github.com/ultralytics/yolov5/releases/download/v7.0/resnet50.pt) | 224 | 76.8 | 93.4 | 11:10 | 23.4 | 1.0 | 25.6 | 8.5 |
| [ResNet101](https://github.com/ultralytics/yolov5/releases/download/v7.0/resnet101.pt) | 224 | 78.5 | 94.3 | 17:10 | 42.1 | 1.9 | 44.5 | 15.9 |
| | | | | | | | | |
| [EfficientNet_b0](https://github.com/ultralytics/yolov5/releases/download/v7.0/efficientnet_b0.pt) | 224 | 75.1 | 92.4 | 13:03 | 12.5 | 1.3 | 5.3 | 1.0 |
| [EfficientNet_b1](https://github.com/ultralytics/yolov5/releases/download/v7.0/efficientnet_b1.pt) | 224 | 76.4 | 93.2 | 17:04 | 14.9 | 1.6 | 7.8 | 1.5 |
| [EfficientNet_b2](https://github.com/ultralytics/yolov5/releases/download/v7.0/efficientnet_b2.pt) | 224 | 76.6 | 93.4 | 17:10 | 15.9 | 1.6 | 9.1 | 1.7 |
| [EfficientNet_b3](https://github.com/ultralytics/yolov5/releases/download/v7.0/efficientnet_b3.pt) | 224 | 77.7 | 94.0 | 19:19 | 18.9 | 1.9 | 12.2 | 2.4 |

Table Notes (click to expand)

- All checkpoints were trained for 90 epochs using the SGD optimizer with `lr0=0.001` and `weight_decay=5e-5` at an image size of 224 pixels, using default settings.
Training runs are logged at [https://wandb.ai/glenn-jocher/YOLOv5-Classifier-v6-2](https://wandb.ai/glenn-jocher/YOLOv5-Classifier-v6-2).
- **Accuracy** values (top-1 and top-5) represent single-model, single-scale performance on the [ImageNet-1k dataset](https://docs.ultralytics.com/datasets/classify/imagenet/).
Reproduce using: `python classify/val.py --data ../datasets/imagenet --img 224`
- **Speed** metrics are averaged over 100 inference images using a Google [Colab Pro V100 High-RAM instance](https://colab.research.google.com/signup).
Reproduce using: `python classify/val.py --data ../datasets/imagenet --img 224 --batch 1`
- **Export** to ONNX (FP32) and TensorRT (FP16) was performed using `export.py`.
Reproduce using: `python export.py --weights yolov5s-cls.pt --include engine onnx --imgsz 224`}}}}}}}}

Classification Usage Examples

### Train

YOLOv5 classification training supports automatic download for datasets like [MNIST](https://docs.ultralytics.com/datasets/classify/mnist/), [Fashion-MNIST](https://docs.ultralytics.com/datasets/classify/fashion-mnist/), [CIFAR10](https://docs.ultralytics.com/datasets/classify/cifar10/), [CIFAR100](https://docs.ultralytics.com/datasets/classify/cifar100/), [Imagenette](https://docs.ultralytics.com/datasets/classify/imagenette/), [Imagewoof](https://docs.ultralytics.com/datasets/classify/imagewoof/), and [ImageNet](https://docs.ultralytics.com/datasets/classify/imagenet/) using the `--data` argument. For example, start training on MNIST with `--data mnist`.

```bash
# Train on a single GPU using CIFAR-100 dataset
python classify/train.py --model yolov5s-cls.pt --data cifar100 --epochs 5 --img 224 --batch 128

# Train using Multi-GPU DDP on ImageNet dataset
python -m torch.distributed.run --nproc_per_node 4 --master_port 1 classify/train.py --model yolov5s-cls.pt --data imagenet --epochs 5 --img 224 --device 0,1,2,3
```

### Val

Validate the accuracy of the YOLOv5m-cls model on the ImageNet-1k validation dataset:

```bash
# Download ImageNet validation split (6.3GB, 50,000 images)
bash data/scripts/get_imagenet.sh --val

# Validate the model
python classify/val.py --weights yolov5m-cls.pt --data ../datasets/imagenet --img 224
```

### Predict

Use the pretrained YOLOv5s-cls.pt model to classify the image `bus.jpg`:

```bash
# Run prediction
python classify/predict.py --weights yolov5s-cls.pt --source data/images/bus.jpg
```

```python
# Load model from PyTorch Hub
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s-cls.pt")
```

### Export

Export trained YOLOv5s-cls, ResNet50, and EfficientNet_b0 models to ONNX and TensorRT formats:

```bash
# Export models
python export.py --weights yolov5s-cls.pt resnet50.pt efficientnet_b0.pt --include onnx engine --img 224
```

## ☁️ Environments

Get started quickly with our pre-configured environments. Click the icons below for setup details.

## 🤝 Contribute

We welcome your contributions! Making YOLOv5 accessible and effective is a community effort. Please see our [Contributing Guide](https://docs.ultralytics.com/help/contributing/) to get started. Share your feedback through the [YOLOv5 Survey](https://www.ultralytics.com/survey?utm_source=github&utm_medium=social&utm_campaign=Survey). Thank you to all our contributors for making YOLOv5 better!

[![Ultralytics open-source contributors](https://raw.githubusercontent.com/ultralytics/assets/main/im/image-contributors.png)](https://github.com/ultralytics/yolov5/graphs/contributors)

## 📜 License

Ultralytics provides two licensing options to meet different needs:

- **AGPL-3.0 License**: An [OSI-approved](https://opensource.org/license/agpl-v3) open-source license ideal for academic research, personal projects, and testing. It promotes open collaboration and knowledge sharing. See the [LICENSE](https://github.com/ultralytics/yolov5/blob/master/LICENSE) file for details.
- **Enterprise License**: Tailored for commercial applications, this license allows seamless integration of Ultralytics software and AI models into commercial products and services, bypassing the open-source requirements of AGPL-3.0. For commercial use cases, please contact us via [Ultralytics Licensing](https://www.ultralytics.com/license).

## 📧 Contact

For bug reports and feature requests related to YOLOv5, please visit [GitHub Issues](https://github.com/ultralytics/yolov5/issues). For general questions, discussions, and community support, join our [Discord server](https://discord.com/invite/ultralytics)!}}}}}}}}