HongChuang/ultralytics-ascend
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 3

Add Docs glossary links (#16448)

Browse source 
Signed-off-by: UltralyticsAssistant <web@ultralytics.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
This commit is contained in:
Glenn Jocher 2024-09-23 23:48:46 +02:00 • committed by GitHub
parent 8b8c25f216
commit 443fbce194
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
193 changed files with 1124 additions and 1124 deletions
Download patch file Download diff file Expand all files Collapse all files

6
docs/en/yolov5/environments/aws_quickstart_tutorial.md
View file

@ -6,7 +6,7 @@ keywords: YOLOv5, AWS, Deep Learning, Machine Learning, AWS EC2, YOLOv5 setup, D
# YOLOv5 🚀 on AWS Deep Learning Instance: Your Complete Guide
Setting up a high-performance deep learning environment can be daunting for newcomers, but fear not! 🛠️ With this guide, we'll walk you through the process of getting YOLOv5 up and running on an AWS Deep Learning instance. By leveraging the power of Amazon Web Services (AWS), even those new to machine learning can get started quickly and cost-effectively. The AWS platform's scalability is perfect for both experimentation and production deployment.
Setting up a high-performance deep learning environment can be daunting for newcomers, but fear not! 🛠️ With this guide, we'll walk you through the process of getting YOLOv5 up and running on an AWS Deep Learning instance. By leveraging the power of Amazon Web Services (AWS), even those new to [machine learning](https://www.ultralytics.com/glossary/machine-learning-ml) can get started quickly and cost-effectively. The AWS platform's scalability is perfect for both experimentation and production deployment.
Other quickstart options for YOLOv5 include our [Colab Notebook](https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb) <a href="https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a> <a href="https://www.kaggle.com/ultralytics/yolov5"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open In Kaggle"></a>, [GCP Deep Learning VM](./google_cloud_quickstart_tutorial.md), and our Docker image at [Docker Hub](https://hub.docker.com/r/ultralytics/yolov5) <a href="https://hub.docker.com/r/ultralytics/yolov5"><img src="https://img.shields.io/docker/pulls/ultralytics/yolov5?logo=docker" alt="Docker Pulls"></a>.
@ -24,7 +24,7 @@ In the EC2 dashboard, you'll find the **Launch Instance** button which is your g
### Selecting the Right Amazon Machine Image (AMI)
Here's where you choose the operating system and software stack for your instance. Type 'Deep Learning' into the search field and select the latest Ubuntu-based Deep Learning AMI, unless your needs dictate otherwise. Amazon's Deep Learning AMIs come pre-installed with popular frameworks and GPU drivers to streamline your setup process.
Here's where you choose the operating system and software stack for your instance. Type '[Deep Learning](https://www.ultralytics.com/glossary/deep-learning-dl)' into the search field and select the latest Ubuntu-based Deep Learning AMI, unless your needs dictate otherwise. Amazon's Deep Learning AMIs come pre-installed with popular frameworks and GPU drivers to streamline your setup process.
![Choose AMI](https://github.com/ultralytics/docs/releases/download/0/choose-ami.avif)
@ -92,4 +92,4 @@ sudo swapon /swapfile # activate swap file
free -h # verify swap memory
```
And that's it! 🎉 You've successfully created an AWS Deep Learning instance and run YOLOv5. Whether you're just starting with object detection or scaling up for production, this setup can help you achieve your machine learning goals. Happy training, validating, and deploying! If you encounter any hiccups along the way, the robust AWS documentation and the active Ultralytics community are here to support you.
And that's it! 🎉 You've successfully created an AWS Deep Learning instance and run YOLOv5. Whether you're just starting with [object detection](https://www.ultralytics.com/glossary/object-detection) or scaling up for production, this setup can help you achieve your machine learning goals. Happy training, validating, and deploying! If you encounter any hiccups along the way, the robust AWS documentation and the active Ultralytics community are here to support you.

2
docs/en/yolov5/environments/azureml_quickstart_tutorial.md
View file

@ -61,7 +61,7 @@ Train the YOLOv5 model:
python train.py
```
Validate the model for Precision, Recall, and mAP
Validate the model for [Precision](https://www.ultralytics.com/glossary/precision), [Recall](https://www.ultralytics.com/glossary/recall), and mAP
```bash
python val.py --weights yolov5s.pt

2
docs/en/yolov5/environments/docker_image_quickstart_tutorial.md
View file

@ -36,7 +36,7 @@ sudo docker run --ipc=host -it ultralytics/yolov5:latest
### Container with local file access:
To run a container with access to local files (e.g., COCO training data in `/datasets`), use the `-v` flag:
To run a container with access to local files (e.g., COCO [training data](https://www.ultralytics.com/glossary/training-data) in `/datasets`), use the `-v` flag:
```bash
sudo docker run --ipc=host -it -v "$(pwd)"/datasets:/usr/src/datasets ultralytics/yolov5:latest

10
docs/en/yolov5/environments/google_cloud_quickstart_tutorial.md
View file

@ -6,13 +6,13 @@ keywords: YOLOv5, Google Cloud Platform, GCP, Deep Learning VM, object detection
# Mastering YOLOv5 🚀 Deployment on Google Cloud Platform (GCP) Deep Learning Virtual Machine (VM) ⭐
Embarking on the journey of artificial intelligence and machine learning can be exhilarating, especially when you leverage the power and flexibility of a cloud platform. Google Cloud Platform (GCP) offers robust tools tailored for machine learning enthusiasts and professionals alike. One such tool is the Deep Learning VM that is preconfigured for data science and ML tasks. In this tutorial, we will navigate through the process of setting up YOLOv5 on a GCP Deep Learning VM. Whether you're taking your first steps in ML or you're a seasoned practitioner, this guide is designed to provide you with a clear pathway to implementing object detection models powered by YOLOv5.
Embarking on the journey of [artificial intelligence](https://www.ultralytics.com/glossary/artificial-intelligence-ai) and machine learning can be exhilarating, especially when you leverage the power and flexibility of a cloud platform. Google Cloud Platform (GCP) offers robust tools tailored for machine learning enthusiasts and professionals alike. One such tool is the Deep Learning VM that is preconfigured for data science and ML tasks. In this tutorial, we will navigate through the process of setting up YOLOv5 on a GCP Deep Learning VM. Whether you're taking your first steps in ML or you're a seasoned practitioner, this guide is designed to provide you with a clear pathway to implementing object detection models powered by YOLOv5.
🆓 Plus, if you're a fresh GCP user, you're in luck with a [$300 free credit offer](https://cloud.google.com/free/docs/free-cloud-features#free-trial) to kickstart your projects.
In addition to GCP, explore other accessible quickstart options for YOLOv5, like our [Colab Notebook](https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb) <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"> for a browser-based experience, or the scalability of [Amazon AWS](./aws_quickstart_tutorial.md). Furthermore, container aficionados can utilize our official Docker image at [Docker Hub](https://hub.docker.com/r/ultralytics/yolov5) <img src="https://img.shields.io/docker/pulls/ultralytics/yolov5?logo=docker" alt="Docker Pulls"> for an encapsulated environment.
## Step 1: Create and Configure Your Deep Learning VM
## Step 1: Create and Configure Your [Deep Learning](https://www.ultralytics.com/glossary/deep-learning-dl) VM
Let's begin by creating a virtual machine that's tuned for deep learning:
@ -42,7 +42,7 @@ cd yolov5
pip install -r requirements.txt
```
This setup process ensures you're working with a Python environment version 3.8.0 or newer and PyTorch 1.8 or above. Our scripts smoothly download [models](https://github.com/ultralytics/yolov5/tree/master/models) and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) rending from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases), making it hassle-free to start model training.
This setup process ensures you're working with a Python environment version 3.8.0 or newer and [PyTorch](https://www.ultralytics.com/glossary/pytorch) 1.8 or above. Our scripts smoothly download [models](https://github.com/ultralytics/yolov5/tree/master/models) and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) rending from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases), making it hassle-free to start model training.
## Step 3: Train and Deploy Your YOLOv5 Models 🌐
@ -62,7 +62,7 @@ python detect.py --weights yolov5s.pt --source path/to/images
python export.py --weights yolov5s.pt --include onnx coreml tflite
```
With just a few commands, YOLOv5 allows you to train custom object detection models tailored to your specific needs or utilize pre-trained weights for quick results on a variety of tasks.
With just a few commands, YOLOv5 allows you to train custom [object detection](https://www.ultralytics.com/glossary/object-detection) models tailored to your specific needs or utilize pre-trained weights for quick results on a variety of tasks.
![Terminal command image illustrating model training on a GCP Deep Learning VM](https://github.com/ultralytics/docs/releases/download/0/terminal-command-model-training.avif)
@ -80,7 +80,7 @@ free -h # confirm the memory increment
### Concluding Thoughts
Congratulations! You are now empowered to harness the capabilities of YOLOv5 with the computational prowess of Google Cloud Platform. This combination provides scalability, efficiency, and versatility for your object detection tasks. Whether for personal projects, academic research, or industrial applications, you have taken a pivotal step into the world of AI and machine learning on the cloud.
Congratulations! You are now empowered to harness the capabilities of YOLOv5 with the computational prowess of Google Cloud Platform. This combination provides scalability, efficiency, and versatility for your object detection tasks. Whether for personal projects, academic research, or industrial applications, you have taken a pivotal step into the world of AI and [machine learning](https://www.ultralytics.com/glossary/machine-learning-ml) on the cloud.
Do remember to document your journey, share insights with the Ultralytics community, and leverage the collaborative arenas such as [GitHub discussions](https://github.com/ultralytics/yolov5/discussions) to grow further. Now, go forth and innovate with YOLOv5 and GCP! 🌟

14
docs/en/yolov5/index.md
View file

@ -22,11 +22,11 @@ keywords: YOLOv5, Ultralytics, object detection, computer vision, deep learning,
<br>
<br>
Welcome to the Ultralytics' <a href="https://github.com/ultralytics/yolov5">YOLOv5</a>🚀 Documentation! YOLOv5, the fifth iteration of the revolutionary "You Only Look Once" object detection model, is designed to deliver high-speed, high-accuracy results in real-time.
Welcome to the Ultralytics' <a href="https://github.com/ultralytics/yolov5">YOLOv5</a>🚀 Documentation! YOLOv5, the fifth iteration of the revolutionary "You Only Look Once" [object detection](https://www.ultralytics.com/glossary/object-detection) model, is designed to deliver high-speed, high-accuracy results in real-time.
<br><br>
Built on PyTorch, this powerful deep learning framework has garnered immense popularity for its versatility, ease of use, and high performance. Our documentation guides you through the installation process, explains the architectural nuances of the model, showcases various use-cases, and provides a series of detailed tutorials. These resources will help you harness the full potential of YOLOv5 for your computer vision projects. Let's get started!
Built on PyTorch, this powerful [deep learning](https://www.ultralytics.com/glossary/deep-learning-dl) framework has garnered immense popularity for its versatility, ease of use, and high performance. Our documentation guides you through the installation process, explains the architectural nuances of the model, showcases various use-cases, and provides a series of detailed tutorials. These resources will help you harness the full potential of YOLOv5 for your [computer vision](https://www.ultralytics.com/glossary/computer-vision-cv) projects. Let's get started!
</div>
@ -42,10 +42,10 @@ Here's a compilation of comprehensive tutorials that will guide you through diff
- [Test-Time Augmentation (TTA)](tutorials/test_time_augmentation.md): Explore how to use TTA to improve your model's prediction accuracy.
- [Model Ensembling](tutorials/model_ensembling.md): Learn the strategy of combining multiple models for improved performance.
- [Model Pruning/Sparsity](tutorials/model_pruning_and_sparsity.md): Understand pruning and sparsity concepts, and how to create a more efficient model.
- [Hyperparameter Evolution](tutorials/hyperparameter_evolution.md): Discover the process of automated hyperparameter tuning for better model performance.
- [Transfer Learning with Frozen Layers](tutorials/transfer_learning_with_frozen_layers.md): Learn how to implement transfer learning by freezing layers in YOLOv5.
- [Hyperparameter Evolution](tutorials/hyperparameter_evolution.md): Discover the process of automated [hyperparameter tuning](https://www.ultralytics.com/glossary/hyperparameter-tuning) for better model performance.
- [Transfer Learning with Frozen Layers](tutorials/transfer_learning_with_frozen_layers.md): Learn how to implement [transfer learning](https://www.ultralytics.com/glossary/transfer-learning) by freezing layers in YOLOv5.
- [Architecture Summary](tutorials/architecture_description.md) 🌟 Delve into the structural details of the YOLOv5 model.
- [Roboflow for Datasets](tutorials/roboflow_datasets_integration.md): Understand how to utilize Roboflow for dataset management, labeling, and active learning.
- [Roboflow for Datasets](tutorials/roboflow_datasets_integration.md): Understand how to utilize Roboflow for dataset management, labeling, and [active learning](https://www.ultralytics.com/glossary/active-learning).
- [ClearML Logging](tutorials/clearml_logging_integration.md) 🌟 Learn how to integrate ClearML for efficient logging during your model training.
- [YOLOv5 with Neural Magic](tutorials/neural_magic_pruning_quantization.md) Discover how to use Neural Magic's Deepsparse to prune and quantize your YOLOv5 model.
- [Comet Logging](tutorials/comet_logging_integration.md) 🌟 NEW: Explore how to utilize Comet for improved model training logging.
@ -95,7 +95,7 @@ We're excited to see the innovative ways you'll use YOLOv5. Dive in, experiment,
### What are the key features of Ultralytics YOLOv5?
Ultralytics YOLOv5 is renowned for its high-speed and high-accuracy object detection capabilities. Built on PyTorch, it is versatile and user-friendly, making it suitable for various computer vision projects. Key features include real-time inference, support for multiple training tricks like Test-Time Augmentation (TTA) and Model Ensembling, and compatibility with export formats such as TFLite, ONNX, CoreML, and TensorRT. To delve deeper into how Ultralytics YOLOv5 can elevate your project, explore our [TFLite, ONNX, CoreML, TensorRT Export guide](tutorials/model_export.md).
Ultralytics YOLOv5 is renowned for its high-speed and high-[accuracy](https://www.ultralytics.com/glossary/accuracy) object detection capabilities. Built on [PyTorch](https://www.ultralytics.com/glossary/pytorch), it is versatile and user-friendly, making it suitable for various computer vision projects. Key features include real-time inference, support for multiple training tricks like Test-Time Augmentation (TTA) and Model Ensembling, and compatibility with export formats such as TFLite, ONNX, CoreML, and TensorRT. To delve deeper into how Ultralytics YOLOv5 can elevate your project, explore our [TFLite, ONNX, CoreML, TensorRT Export guide](tutorials/model_export.md).
### How can I train a custom YOLOv5 model on my dataset?
@ -107,7 +107,7 @@ Ultralytics YOLOv5 is preferred over models like RCNN due to its superior speed
### How can I optimize YOLOv5 model performance during training?
Optimizing YOLOv5 model performance involves tuning various hyperparameters and incorporating techniques like data augmentation and transfer learning. Ultralytics provides comprehensive resources on hyperparameter evolution and pruning/sparsity to improve model efficiency. You can discover practical tips in our [Tips for Best Training Results guide](tutorials/tips_for_best_training_results.md), which offers actionable insights for achieving optimal performance during training.
Optimizing YOLOv5 model performance involves tuning various hyperparameters and incorporating techniques like [data augmentation](https://www.ultralytics.com/glossary/data-augmentation) and transfer learning. Ultralytics provides comprehensive resources on hyperparameter evolution and pruning/sparsity to improve model efficiency. You can discover practical tips in our [Tips for Best Training Results guide](tutorials/tips_for_best_training_results.md), which offers actionable insights for achieving optimal performance during training.
### What environments are supported for running YOLOv5 applications?

8
docs/en/yolov5/quickstart_tutorial.md
View file

@ -6,7 +6,7 @@ keywords: YOLOv5, Quickstart, real-time object detection, AI, ML, PyTorch, infer
# YOLOv5 Quickstart 🚀
Embark on your journey into the dynamic realm of real-time object detection with YOLOv5! This guide is crafted to serve as a comprehensive starting point for AI enthusiasts and professionals aiming to master YOLOv5. From initial setup to advanced training techniques, we've got you covered. By the end of this guide, you'll have the knowledge to implement YOLOv5 into your projects confidently. Let's ignite the engines and soar into YOLOv5!
Embark on your journey into the dynamic realm of real-time [object detection](https://www.ultralytics.com/glossary/object-detection) with YOLOv5! This guide is crafted to serve as a comprehensive starting point for AI enthusiasts and professionals aiming to master YOLOv5. From initial setup to advanced training techniques, we've got you covered. By the end of this guide, you'll have the knowledge to implement YOLOv5 into your projects confidently. Let's ignite the engines and soar into YOLOv5!
## Install
@ -18,7 +18,7 @@ cd yolov5
pip install -r requirements.txt # install dependencies
```
## Inference with PyTorch Hub
## Inference with [PyTorch](https://www.ultralytics.com/glossary/pytorch) Hub
Experience the simplicity of YOLOv5 [PyTorch Hub](./tutorials/pytorch_hub_model_loading.md) inference, where [models](https://github.com/ultralytics/yolov5/tree/master/models) are seamlessly downloaded from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases).
@ -57,7 +57,7 @@ python detect.py --weights yolov5s.pt --source 0 #
## Training
Replicate the YOLOv5 [COCO](https://github.com/ultralytics/yolov5/blob/master/data/scripts/get_coco.sh) benchmarks with the instructions below. The necessary [models](https://github.com/ultralytics/yolov5/tree/master/models) and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) are pulled directly from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). Training YOLOv5n/s/m/l/x on a V100 GPU should typically take 1/2/4/6/8 days respectively (note that [Multi-GPU](./tutorials/multi_gpu_training.md) setups work faster). Maximize performance by using the highest possible `--batch-size` or use `--batch-size -1` for the YOLOv5 [AutoBatch](https://github.com/ultralytics/yolov5/pull/5092) feature. The following batch sizes are ideal for V100-16GB GPUs.
Replicate the YOLOv5 [COCO](https://github.com/ultralytics/yolov5/blob/master/data/scripts/get_coco.sh) benchmarks with the instructions below. The necessary [models](https://github.com/ultralytics/yolov5/tree/master/models) and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) are pulled directly from the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). Training YOLOv5n/s/m/l/x on a V100 GPU should typically take 1/2/4/6/8 days respectively (note that [Multi-GPU](./tutorials/multi_gpu_training.md) setups work faster). Maximize performance by using the highest possible `--batch-size` or use `--batch-size -1` for the YOLOv5 [AutoBatch](https://github.com/ultralytics/yolov5/pull/5092) feature. The following [batch sizes](https://www.ultralytics.com/glossary/batch-size) are ideal for V100-16GB GPUs.
```bash
python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5n.yaml --batch-size 128
@ -69,4 +69,4 @@ python train.py --data coco.yaml --epochs 300 --weights '' --cfg yolov5n.yaml -
<img width="800" src="https://github.com/ultralytics/docs/releases/download/0/yolov5-training-curves.avif" alt="YOLO training curves">
To conclude, YOLOv5 is not only a state-of-the-art tool for object detection but also a testament to the power of machine learning in transforming the way we interact with the world through visual understanding. As you progress through this guide and begin applying YOLOv5 to your projects, remember that you are at the forefront of a technological revolution, capable of achieving remarkable feats. Should you need further insights or support from fellow visionaries, you're invited to our [GitHub repository](https://github.com/ultralytics/yolov5) home to a thriving community of developers and researchers. Keep exploring, keep innovating, and enjoy the marvels of YOLOv5. Happy detecting! 🌠🔍
To conclude, YOLOv5 is not only a state-of-the-art tool for object detection but also a testament to the power of [machine learning](https://www.ultralytics.com/glossary/machine-learning-ml) in transforming the way we interact with the world through visual understanding. As you progress through this guide and begin applying YOLOv5 to your projects, remember that you are at the forefront of a technological revolution, capable of achieving remarkable feats. Should you need further insights or support from fellow visionaries, you're invited to our [GitHub repository](https://github.com/ultralytics/yolov5) home to a thriving community of developers and researchers. Keep exploring, keep innovating, and enjoy the marvels of YOLOv5. Happy detecting! 🌠🔍

16
docs/en/yolov5/tutorials/architecture_description.md
View file

@ -6,7 +6,7 @@ keywords: YOLOv5 architecture, object detection, Ultralytics, YOLO, model struct
# Ultralytics YOLOv5 Architecture
YOLOv5 (v6.0/6.1) is a powerful object detection algorithm developed by Ultralytics. This article dives deep into the YOLOv5 architecture, data augmentation strategies, training methodologies, and loss computation techniques. This comprehensive understanding will help improve your practical application of object detection in various fields, including surveillance, autonomous vehicles, and image recognition.
YOLOv5 (v6.0/6.1) is a powerful object detection algorithm developed by Ultralytics. This article dives deep into the YOLOv5 architecture, [data augmentation](https://www.ultralytics.com/glossary/data-augmentation) strategies, training methodologies, and loss computation techniques. This comprehensive understanding will help improve your practical application of object detection in various fields, including surveillance, autonomous vehicles, and [image recognition](https://www.ultralytics.com/glossary/image-recognition).
## 1. Model Structure
@ -104,9 +104,9 @@ SPPF time: 0.20780706405639648
## 2. Data Augmentation Techniques
YOLOv5 employs various data augmentation techniques to improve the model's ability to generalize and reduce overfitting. These techniques include:
YOLOv5 employs various data augmentation techniques to improve the model's ability to generalize and reduce [overfitting](https://www.ultralytics.com/glossary/overfitting). These techniques include:
- **Mosaic Augmentation**: An image processing technique that combines four training images into one in ways that encourage object detection models to better handle various object scales and translations.
- **Mosaic Augmentation**: An image processing technique that combines four training images into one in ways that encourage [object detection](https://www.ultralytics.com/glossary/object-detection) models to better handle various object scales and translations.
![mosaic](https://github.com/ultralytics/docs/releases/download/0/mosaic-augmentation.avif)
@ -138,9 +138,9 @@ YOLOv5 applies several sophisticated training strategies to enhance the model's
- **Multiscale Training**: The input images are randomly rescaled within a range of 0.5 to 1.5 times their original size during the training process.
- **AutoAnchor**: This strategy optimizes the prior anchor boxes to match the statistical characteristics of the ground truth boxes in your custom data.
- **Warmup and Cosine LR Scheduler**: A method to adjust the learning rate to enhance model performance.
- **Warmup and Cosine LR Scheduler**: A method to adjust the [learning rate](https://www.ultralytics.com/glossary/learning-rate) to enhance model performance.
- **Exponential Moving Average (EMA)**: A strategy that uses the average of parameters over past steps to stabilize the training process and reduce generalization error.
- **Mixed Precision Training**: A method to perform operations in half-precision format, reducing memory usage and enhancing computational speed.
- **[Mixed Precision](https://www.ultralytics.com/glossary/mixed-precision) Training**: A method to perform operations in half-[precision](https://www.ultralytics.com/glossary/precision) format, reducing memory usage and enhancing computational speed.
- **Hyperparameter Evolution**: A strategy to automatically tune hyperparameters to achieve optimal performance.
## 4. Additional Features
@ -153,7 +153,7 @@ The loss in YOLOv5 is computed as a combination of three individual loss compone
- **Objectness Loss (BCE Loss)**: Another Binary Cross-Entropy loss, calculates the error in detecting whether an object is present in a particular grid cell or not.
- **Location Loss (CIoU Loss)**: Complete IoU loss, measures the error in localizing the object within the grid cell.
The overall loss function is depicted by:
The overall [loss function](https://www.ultralytics.com/glossary/loss-function) is depicted by:
![loss](https://latex.codecogs.com/svg.image?Loss=\lambda_1L_{cls}+\lambda_2L_{obj}+\lambda_3L_{loc})
@ -176,7 +176,7 @@ The YOLOv5 architecture makes some important changes to the box prediction strat
However, in YOLOv5, the formula for predicting the box coordinates has been updated to reduce grid sensitivity and prevent the model from predicting unbounded box dimensions.
The revised formulas for calculating the predicted bounding box are as follows:
The revised formulas for calculating the predicted [bounding box](https://www.ultralytics.com/glossary/bounding-box) are as follows:
![bx](<https://latex.codecogs.com/svg.image?b_x=(2\cdot\sigma(t_x)-0.5)+c_x>)
![by](<https://latex.codecogs.com/svg.image?b_y=(2\cdot\sigma(t_y)-0.5)+c_y>)
@ -193,7 +193,7 @@ Compare the height and width scaling ratio(relative to anchor) before and after
### 4.4 Build Targets
The build target process in YOLOv5 is critical for training efficiency and model accuracy. It involves assigning ground truth boxes to the appropriate grid cells in the output map and matching them with the appropriate anchor boxes.
The build target process in YOLOv5 is critical for training efficiency and model [accuracy](https://www.ultralytics.com/glossary/accuracy). It involves assigning ground truth boxes to the appropriate grid cells in the output map and matching them with the appropriate anchor boxes.
This process follows these steps:

6
docs/en/yolov5/tutorials/clearml_logging_integration.md
View file

@ -14,7 +14,7 @@ keywords: ClearML, YOLOv5, machine learning, experiment tracking, data versionin
🔨 Track every YOLOv5 training run in the <b>experiment manager</b>
🔧 Version and easily access your custom training data with the integrated ClearML <b>Data Versioning Tool</b>
🔧 Version and easily access your custom [training data](https://www.ultralytics.com/glossary/training-data) with the integrated ClearML <b>Data Versioning Tool</b>
🔦 <b>Remotely train and monitor</b> your YOLOv5 training runs using ClearML Agent
@ -85,8 +85,8 @@ This will capture:
- Console output
- Scalars (mAP_0.5, mAP_0.5:0.95, precision, recall, losses, learning rates, ...)
- General info such as machine details, runtime, creation date etc.
- All produced plots such as label correlogram and confusion matrix
- Images with bounding boxes per epoch
- All produced plots such as label correlogram and [confusion matrix](https://www.ultralytics.com/glossary/confusion-matrix)
- Images with bounding boxes per [epoch](https://www.ultralytics.com/glossary/epoch)
- Mosaic per epoch
- Validation images per epoch

12
docs/en/yolov5/tutorials/comet_logging_integration.md
View file

@ -12,7 +12,7 @@ This guide will cover how to use YOLOv5 with [Comet](https://bit.ly/yolov5-readm
## About Comet
Comet builds tools that help data scientists, engineers, and team leaders accelerate and optimize machine learning and deep learning models.
Comet builds tools that help data scientists, engineers, and team leaders accelerate and optimize [machine learning](https://www.ultralytics.com/glossary/machine-learning-ml) and [deep learning](https://www.ultralytics.com/glossary/deep-learning-dl) models.
Track and visualize model metrics in real time, save your hyperparameters, datasets, and model checkpoints, and visualize your model predictions with [Comet Custom Panels](https://www.comet.com/docs/v2/guides/comet-dashboard/code-panels/about-panels/?utm_source=yolov5&utm_medium=partner&utm_campaign=partner_yolov5_2022&utm_content=github)! Comet makes sure you never lose track of your work and makes it easy to share results and collaborate across teams of all sizes!
@ -72,7 +72,7 @@ By default, Comet will log the following items
## Metrics
- Box Loss, Object Loss, Classification Loss for the training and validation data
- Box Loss, Object Loss, Classification Loss for the training and [validation data](https://www.ultralytics.com/glossary/validation-data)
- mAP_0.5, mAP_0.5:0.95 metrics for the validation data.
- Precision and Recall for the validation data
@ -83,7 +83,7 @@ By default, Comet will log the following items
## Visualizations
- Confusion Matrix of the model predictions on the validation data
- [Confusion Matrix](https://www.ultralytics.com/glossary/confusion-matrix) of the model predictions on the validation data
- Plots for the PR and F1 curves across all classes
- Correlogram of the Class Labels
@ -120,9 +120,9 @@ python train.py \
By default, model predictions (images, ground truth labels and bounding boxes) will be logged to Comet.
You can control the frequency of logged predictions and the associated images by passing the `bbox_interval` command line argument. Predictions can be visualized using Comet's Object Detection Custom Panel. This frequency corresponds to every Nth batch of data per epoch. In the example below, we are logging every 2nd batch of data for each epoch.
You can control the frequency of logged predictions and the associated images by passing the `bbox_interval` command line argument. Predictions can be visualized using Comet's [Object Detection](https://www.ultralytics.com/glossary/object-detection) Custom Panel. This frequency corresponds to every Nth batch of data per [epoch](https://www.ultralytics.com/glossary/epoch). In the example below, we are logging every 2nd batch of data for each epoch.
**Note:** The YOLOv5 validation dataloader will default to a batch size of 32, so you will have to set the logging frequency accordingly.
**Note:** The YOLOv5 validation dataloader will default to a [batch size](https://www.ultralytics.com/glossary/batch-size) of 32, so you will have to set the logging frequency accordingly.
Here is an [example project using the Panel](https://www.comet.com/examples/comet-example-yolov5?shareable=YcwMiJaZSXfcEXpGOHDD12vA1&utm_source=yolov5&utm_medium=partner&utm_campaign=partner_yolov5_2022&utm_content=github)
@ -152,7 +152,7 @@ env COMET_MAX_IMAGE_UPLOADS=200 python train.py \
### Logging Class Level Metrics
Use the `COMET_LOG_PER_CLASS_METRICS` environment variable to log mAP, precision, recall, f1 for each class.
Use the `COMET_LOG_PER_CLASS_METRICS` environment variable to log mAP, [precision](https://www.ultralytics.com/glossary/precision), [recall](https://www.ultralytics.com/glossary/recall), f1 for each class.
```shell
env COMET_LOG_PER_CLASS_METRICS=true python train.py \

4
docs/en/yolov5/tutorials/hyperparameter_evolution.md
View file

@ -61,7 +61,7 @@ copy_paste: 0.0 # segment copy-paste (probability)
## 2. Define Fitness
Fitness is the value we seek to maximize. In YOLOv5 we define a default fitness function as a weighted combination of metrics: `mAP@0.5` contributes 10% of the weight and `mAP@0.5:0.95` contributes the remaining 90%, with [Precision `P` and Recall `R`](https://en.wikipedia.org/wiki/Precision_and_recall) absent. You may adjust these as you see fit or use the default fitness definition in utils/metrics.py (recommended).
Fitness is the value we seek to maximize. In YOLOv5 we define a default fitness function as a weighted combination of metrics: `mAP@0.5` contributes 10% of the weight and `mAP@0.5:0.95` contributes the remaining 90%, with [Precision `P` and [Recall](https://www.ultralytics.com/glossary/recall) `R`](https://en.wikipedia.org/wiki/Precision_and_recall) absent. You may adjust these as you see fit or use the default fitness definition in utils/metrics.py (recommended).
```python
def fitness(x):
@ -72,7 +72,7 @@ def fitness(x):
## 3. Evolve
Evolution is performed about a base scenario which we seek to improve upon. The base scenario in this example is finetuning COCO128 for 10 epochs using pretrained YOLOv5s. The base scenario training command is:
Evolution is performed about a base scenario which we seek to improve upon. The base scenario in this example is [finetuning](https://www.ultralytics.com/glossary/fine-tuning) COCO128 for 10 [epochs](https://www.ultralytics.com/glossary/epoch) using pretrained YOLOv5s. The base scenario training command is:
```bash
python train.py --epochs 10 --data coco128.yaml --weights yolov5s.pt --cache

4
docs/en/yolov5/tutorials/model_ensembling.md
View file

@ -4,11 +4,11 @@ description: Learn how to use YOLOv5 model ensembling during testing and inferen
keywords: YOLOv5, model ensembling, testing, inference, mAP, Recall, Ultralytics, object detection, PyTorch
---
📚 This guide explains how to use YOLOv5 🚀 **model ensembling** during testing and inference for improved mAP and Recall.
📚 This guide explains how to use YOLOv5 🚀 **model ensembling** during testing and inference for improved mAP and [Recall](https://www.ultralytics.com/glossary/recall).
From [https://en.wikipedia.org/wiki/Ensemble_learning](https://en.wikipedia.org/wiki/Ensemble_learning):
> Ensemble modeling is a process where multiple diverse models are created to predict an outcome, either by using many different modeling algorithms or using different training data sets. The ensemble model then aggregates the prediction of each base model and results in once final prediction for the unseen data. The motivation for using ensemble models is to reduce the generalization error of the prediction. As long as the base models are diverse and independent, the prediction error of the model decreases when the ensemble approach is used. The approach seeks the wisdom of crowds in making a prediction. Even though the ensemble model has multiple base models within the model, it acts and performs as a single model.
> Ensemble modeling is a process where multiple diverse models are created to predict an outcome, either by using many different modeling algorithms or using different [training data](https://www.ultralytics.com/glossary/training-data) sets. The ensemble model then aggregates the prediction of each base model and results in once final prediction for the unseen data. The motivation for using ensemble models is to reduce the generalization error of the prediction. As long as the base models are diverse and independent, the prediction error of the model decreases when the ensemble approach is used. The approach seeks the wisdom of crowds in making a prediction. Even though the ensemble model has multiple base models within the model, it acts and performs as a single model.
## Before You Start

6
docs/en/yolov5/tutorials/model_export.md
View file

@ -6,7 +6,7 @@ keywords: YOLOv5 export, TFLite, ONNX, CoreML, TensorRT, model conversion, YOLOv
# TFLite, ONNX, CoreML, TensorRT Export
📚 This guide explains how to export a trained YOLOv5 🚀 model from PyTorch to ONNX and TorchScript formats.
📚 This guide explains how to export a trained YOLOv5 🚀 model from [PyTorch](https://www.ultralytics.com/glossary/pytorch) to ONNX and TorchScript formats.
## Before You Start
@ -103,7 +103,7 @@ This command exports a pretrained YOLOv5s model to TorchScript and ONNX formats.
python export.py --weights yolov5s.pt --include torchscript onnx
```
💡 ProTip: Add `--half` to export models at FP16 half precision for smaller file sizes
💡 ProTip: Add `--half` to export models at FP16 half [precision](https://www.ultralytics.com/glossary/precision) for smaller file sizes
Output:
@ -205,7 +205,7 @@ results.print() # or .show(), .save(), .crop(), .pandas(), etc.
## OpenCV DNN inference
OpenCV inference with ONNX models:
[OpenCV](https://www.ultralytics.com/glossary/opencv) inference with ONNX models:
```bash
python export.py --weights yolov5s.pt --include onnx

6
docs/en/yolov5/tutorials/multi_gpu_training.md
View file

@ -18,7 +18,7 @@ pip install -r requirements.txt # install
💡 ProTip! **Docker Image** is recommended for all Multi-GPU trainings. See [Docker Quickstart Guide](../environments/docker_image_quickstart_tutorial.md) <a href="https://hub.docker.com/r/ultralytics/yolov5"><img src="https://img.shields.io/docker/pulls/ultralytics/yolov5?logo=docker" alt="Docker Pulls"></a>
💡 ProTip! `torch.distributed.run` replaces `torch.distributed.launch` in **PyTorch>=1.9**. See [docs](https://pytorch.org/docs/stable/distributed.html) for details.
💡 ProTip! `torch.distributed.run` replaces `torch.distributed.launch` in **[PyTorch](https://www.ultralytics.com/glossary/pytorch)>=1.9**. See [docs](https://pytorch.org/docs/stable/distributed.html) for details.
## Training
@ -69,7 +69,7 @@ python -m torch.distributed.run --nproc_per_node 2 train.py --batch 64 --data co
<details>
<summary>Use SyncBatchNorm (click to expand)</summary>
[SyncBatchNorm](https://pytorch.org/docs/master/generated/torch.nn.SyncBatchNorm.html) could increase accuracy for multiple gpu training, however, it will slow down training by a significant factor. It is **only** available for Multiple GPU DistributedDataParallel training.
[SyncBatchNorm](https://pytorch.org/docs/master/generated/torch.nn.SyncBatchNorm.html) could increase [accuracy](https://www.ultralytics.com/glossary/accuracy) for multiple gpu training, however, it will slow down training by a significant factor. It is **only** available for Multiple GPU DistributedDataParallel training.
It is best used when the batch-size on **each** GPU is small (<= 8).
@ -121,7 +121,7 @@ python -m torch.distributed.run --master_port 1234 --nproc_per_node 2 ...
## Results
DDP profiling results on an [AWS EC2 P4d instance](../environments/aws_quickstart_tutorial.md) with 8x A100 SXM4-40GB for YOLOv5l for 1 COCO epoch.
DDP profiling results on an [AWS EC2 P4d instance](../environments/aws_quickstart_tutorial.md) with 8x A100 SXM4-40GB for YOLOv5l for 1 COCO [epoch](https://www.ultralytics.com/glossary/epoch).
<details>
<summary>Profiling code</summary>

6
docs/en/yolov5/tutorials/neural_magic_pruning_quantization.md
View file

@ -30,7 +30,7 @@ DeepSparse is an inference runtime with exceptional performance on CPUs. For ins
<img width="60%" src="https://github.com/ultralytics/docs/releases/download/0/yolov5-speed-improvement.avif" alt="YOLOv5 speed improvement">
</p>
For the first time, your deep learning workloads can meet the performance demands of production without the complexity and costs of hardware accelerators. Put simply, DeepSparse gives you the performance of GPUs and the simplicity of software:
For the first time, your [deep learning](https://www.ultralytics.com/glossary/deep-learning-dl) workloads can meet the performance demands of production without the complexity and costs of hardware accelerators. Put simply, DeepSparse gives you the performance of GPUs and the simplicity of software:
- **Flexible Deployments**: Run consistently across cloud, data center, and edge with any hardware provider from Intel to AMD to ARM
- **Infinite Scalability**: Scale vertically to 100s of cores, out with standard Kubernetes, or fully-abstracted with Serverless
@ -40,7 +40,7 @@ For the first time, your deep learning workloads can meet the performance demand
DeepSparse takes advantage of model sparsity to gain its performance speedup.
Sparsification through pruning and quantization is a broadly studied technique, allowing order-of-magnitude reductions in the size and compute needed to execute a network, while maintaining high accuracy. DeepSparse is sparsity-aware, meaning it skips the zeroed out parameters, shrinking amount of compute in a forward pass. Since the sparse computation is now memory bound, DeepSparse executes the network depth-wise, breaking the problem into Tensor Columns, vertical stripes of computation that fit in cache.
Sparsification through pruning and quantization is a broadly studied technique, allowing order-of-magnitude reductions in the size and compute needed to execute a network, while maintaining high [accuracy](https://www.ultralytics.com/glossary/accuracy). DeepSparse is sparsity-aware, meaning it skips the zeroed out parameters, shrinking amount of compute in a forward pass. Since the sparse computation is now memory bound, DeepSparse executes the network depth-wise, breaking the problem into Tensor Columns, vertical stripes of computation that fit in cache.
<p align="center">
<img width="60%" src="https://github.com/ultralytics/docs/releases/download/0/tensor-columns.avif" alt="YOLO model pruning">
@ -122,7 +122,7 @@ apt-get install libgl1
#### HTTP Server
DeepSparse Server runs on top of the popular FastAPI web framework and Uvicorn web server. With just a single CLI command, you can easily setup a model service endpoint with DeepSparse. The Server supports any Pipeline from DeepSparse, including object detection with YOLOv5, enabling you to send raw images to the endpoint and receive the bounding boxes.
DeepSparse Server runs on top of the popular FastAPI web framework and Uvicorn web server. With just a single CLI command, you can easily setup a model service endpoint with DeepSparse. The Server supports any Pipeline from DeepSparse, including [object detection](https://www.ultralytics.com/glossary/object-detection) with YOLOv5, enabling you to send raw images to the endpoint and receive the bounding boxes.
Spin up the Server with the pruned-quantized YOLOv5s:

4
docs/en/yolov5/tutorials/pytorch_hub_model_loading.md
View file

@ -4,7 +4,7 @@ description: Learn how to load YOLOv5 from PyTorch Hub for seamless model infere
keywords: YOLOv5, PyTorch Hub, model loading, Ultralytics, object detection, machine learning, AI, tutorial, inference
---
📚 This guide explains how to load YOLOv5 🚀 from PyTorch Hub at [https://pytorch.org/hub/ultralytics_yolov5](https://pytorch.org/hub/ultralytics_yolov5/).
📚 This guide explains how to load YOLOv5 🚀 from [PyTorch](https://www.ultralytics.com/glossary/pytorch) Hub at [https://pytorch.org/hub/ultralytics_yolov5](https://pytorch.org/hub/ultralytics_yolov5/).
## Before You Start
@ -44,7 +44,7 @@ results.pandas().xyxy[0]
### Detailed Example
This example shows **batched inference** with **PIL** and **OpenCV** image sources. `results` can be **printed** to console, **saved** to `runs/hub`, **showed** to screen on supported environments, and returned as **tensors** or **pandas** dataframes.
This example shows **batched inference** with **PIL** and **[OpenCV](https://www.ultralytics.com/glossary/opencv)** image sources. `results` can be **printed** to console, **saved** to `runs/hub`, **showed** to screen on supported environments, and returned as **tensors** or **pandas** dataframes.
```python
import cv2

6
docs/en/yolov5/tutorials/roboflow_datasets_integration.md
View file

@ -52,7 +52,7 @@ We have released a custom training tutorial demonstrating all of the above capab
## Active Learning
The real world is messy and your model will invariably encounter situations your dataset didn't anticipate. Using [active learning](https://blog.roboflow.com/what-is-active-learning/?ref=ultralytics) is an important strategy to iteratively improve your dataset and model. With the Roboflow and YOLOv5 integration, you can quickly make improvements on your model deployments by using a battle tested machine learning pipeline.
The real world is messy and your model will invariably encounter situations your dataset didn't anticipate. Using [active learning](https://blog.roboflow.com/what-is-active-learning/?ref=ultralytics) is an important strategy to iteratively improve your dataset and model. With the Roboflow and YOLOv5 integration, you can quickly make improvements on your [model deployments](https://www.ultralytics.com/glossary/model-deployment) by using a battle tested [machine learning](https://www.ultralytics.com/glossary/machine-learning-ml) pipeline.
<p align=""><a href="https://roboflow.com/?ref=ultralytics"><img width="1000" src="https://github.com/ultralytics/docs/releases/download/0/roboflow-active-learning.avif" alt="Roboflow active learning"></a></p>
@ -96,10 +96,10 @@ dataset = project.version("YOUR VERSION").download("yolov5")
This code will download your dataset in a format compatible with YOLOv5, allowing you to quickly begin training your model. For more details, refer to the [Exporting Data](#exporting-data) section.
### What is active learning and how does it work with YOLOv5 and Roboflow?
### What is [active learning](https://www.ultralytics.com/glossary/active-learning) and how does it work with YOLOv5 and Roboflow?
Active learning is a machine learning strategy that iteratively improves a model by intelligently selecting the most informative data points to label. With the Roboflow and YOLOv5 integration, you can implement active learning to continuously enhance your model's performance. This involves deploying a model, capturing new data, using the model to make predictions, and then manually verifying or correcting those predictions to further train the model. For more insights into active learning see the [Active Learning](#active-learning) section above.
### How can I use Ultralytics environments for training YOLOv5 models on different platforms?
Ultralytics provides ready-to-use environments with pre-installed dependencies like CUDA, CUDNN, Python, and PyTorch, making it easier to kickstart your training projects. These environments are available on various platforms such as Google Cloud, AWS, Azure, and Docker. You can also access free GPU notebooks via [Paperspace](https://bit.ly/yolov5-paperspace-notebook), [Google Colab](https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb), and [Kaggle](https://www.kaggle.com/ultralytics/yolov5). For specific setup instructions, visit the [Supported Environments](#supported-environments) section of the documentation.
Ultralytics provides ready-to-use environments with pre-installed dependencies like CUDA, CUDNN, Python, and [PyTorch](https://www.ultralytics.com/glossary/pytorch), making it easier to kickstart your training projects. These environments are available on various platforms such as Google Cloud, AWS, Azure, and Docker. You can also access free GPU notebooks via [Paperspace](https://bit.ly/yolov5-paperspace-notebook), [Google Colab](https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb), and [Kaggle](https://www.kaggle.com/ultralytics/yolov5). For specific setup instructions, visit the [Supported Environments](#supported-environments) section of the documentation.

2
docs/en/yolov5/tutorials/test_time_augmentation.md
View file

@ -6,7 +6,7 @@ keywords: YOLOv5, Test-Time Augmentation, TTA, machine learning, deep learning,
# Test-Time Augmentation (TTA)
📚 This guide explains how to use Test Time Augmentation (TTA) during testing and inference for improved mAP and Recall with YOLOv5 🚀.
📚 This guide explains how to use Test Time Augmentation (TTA) during testing and inference for improved mAP and [Recall](https://www.ultralytics.com/glossary/recall) with YOLOv5 🚀.
## Before You Start

10
docs/en/yolov5/tutorials/tips_for_best_training_results.md
View file

@ -8,7 +8,7 @@ keywords: YOLOv5 training, mAP, dataset best practices, model selection, trainin
Most of the time good results can be obtained with no changes to the models or training settings, **provided your dataset is sufficiently large and well labelled**. If at first you don't get good results, there are steps you might be able to take to improve, but we always recommend users **first train with all default settings** before considering any changes. This helps establish a performance baseline and spot areas for improvement.
If you have questions about your training results **we recommend you provide the maximum amount of information possible** if you expect a helpful response, including results plots (train losses, val losses, P, R, mAP), PR curve, confusion matrix, training mosaics, test results and dataset statistics images such as labels.png. All of these are located in your `project/name` directory, typically `yolov5/runs/train/exp`.
If you have questions about your training results **we recommend you provide the maximum amount of information possible** if you expect a helpful response, including results plots (train losses, val losses, P, R, mAP), PR curve, [confusion matrix](https://www.ultralytics.com/glossary/confusion-matrix), training mosaics, test results and dataset statistics images such as labels.png. All of these are located in your `project/name` directory, typically `yolov5/runs/train/exp`.
We've put together a full guide for users looking to get the best results on their YOLOv5 trainings below.
@ -18,7 +18,7 @@ We've put together a full guide for users looking to get the best results on the
- **Instances per class.** ≥ 10000 instances (labeled objects) per class recommended
- **Image variety.** Must be representative of deployed environment. For real-world use cases we recommend images from different times of day, different seasons, different weather, different lighting, different angles, different sources (scraped online, collected locally, different cameras) etc.
- **Label consistency.** All instances of all classes in all images must be labelled. Partial labelling will not work.
- **Label accuracy.** Labels must closely enclose each object. No space should exist between an object and it's bounding box. No objects should be missing a label.
- **Label [accuracy](https://www.ultralytics.com/glossary/accuracy).** Labels must closely enclose each object. No space should exist between an object and it's [bounding box](https://www.ultralytics.com/glossary/bounding-box). No objects should be missing a label.
- **Label verification.** View `train_batch*.jpg` on train start to verify your labels appear correct, i.e. see [example](./train_custom_data.md#local-logging) mosaic.
- **Background images.** Background images are images with no objects that are added to a dataset to reduce False Positives (FP). We recommend about 0-10% background images to help reduce FPs (COCO has 1000 background images for reference, 1% of the total). No labels are required for background images.
@ -53,13 +53,13 @@ python train.py --data custom.yaml --weights '' --cfg yolov5s.yaml
Before modifying anything, **first train with default settings to establish a performance baseline**. A full list of train.py settings can be found in the [train.py](https://github.com/ultralytics/yolov5/blob/master/train.py) argparser.
- **Epochs.** Start with 300 epochs. If this overfits early then you can reduce epochs. If overfitting does not occur after 300 epochs, train longer, i.e. 600, 1200 etc. epochs.
- **[Epochs](https://www.ultralytics.com/glossary/epoch).** Start with 300 epochs. If this overfits early then you can reduce epochs. If [overfitting](https://www.ultralytics.com/glossary/overfitting) does not occur after 300 epochs, train longer, i.e. 600, 1200 etc. epochs.
- **Image size.** COCO trains at native resolution of `--img 640`, though due to the high amount of small objects in the dataset it can benefit from training at higher resolutions such as `--img 1280`. If there are many small objects then custom datasets will benefit from training at native or higher resolution. Best inference results are obtained at the same `--img` as the training was run at, i.e. if you train at `--img 1280` you should also test and detect at `--img 1280`.
- **Batch size.** Use the largest `--batch-size` that your hardware allows for. Small batch sizes produce poor batchnorm statistics and should be avoided.
- **[Batch size](https://www.ultralytics.com/glossary/batch-size).** Use the largest `--batch-size` that your hardware allows for. Small batch sizes produce poor batchnorm statistics and should be avoided.
- **Hyperparameters.** Default hyperparameters are in [hyp.scratch-low.yaml](https://github.com/ultralytics/yolov5/blob/master/data/hyps/hyp.scratch-low.yaml). We recommend you train with default hyperparameters first before thinking of modifying any. In general, increasing augmentation hyperparameters will reduce and delay overfitting, allowing for longer trainings and higher final mAP. Reduction in loss component gain hyperparameters like `hyp['obj']` will help reduce overfitting in those specific loss components. For an automated method of optimizing these hyperparameters, see our [Hyperparameter Evolution Tutorial](./hyperparameter_evolution.md).
## Further Reading
If you'd like to know more, a good place to start is Karpathy's 'Recipe for Training Neural Networks', which has great ideas for training that apply broadly across all ML domains: [https://karpathy.github.io/2019/04/25/recipe/](https://karpathy.github.io/2019/04/25/recipe/)
If you'd like to know more, a good place to start is Karpathy's 'Recipe for Training [Neural Networks](https://www.ultralytics.com/glossary/neural-network-nn)', which has great ideas for training that apply broadly across all ML domains: [https://karpathy.github.io/2019/04/25/recipe/](https://karpathy.github.io/2019/04/25/recipe/)
Good luck 🍀 and let us know if you have any other questions!

10
docs/en/yolov5/tutorials/train_custom_data.md
View file

@ -77,7 +77,7 @@ Export in `YOLOv5 Pytorch` format, then copy the snippet into your training scri
### 2.1 Create `dataset.yaml`
[COCO128](https://www.kaggle.com/ultralytics/coco128) is an example small tutorial dataset composed of the first 128 images in [COCO](https://cocodataset.org/) train2017. These same 128 images are used for both training and validation to verify our training pipeline is capable of overfitting. [data/coco128.yaml](https://github.com/ultralytics/yolov5/blob/master/data/coco128.yaml), shown below, is the dataset config file that defines 1) the dataset root directory `path` and relative paths to `train` / `val` / `test` image directories (or `*.txt` files with image paths) and 2) a class `names` dictionary:
[COCO128](https://www.kaggle.com/ultralytics/coco128) is an example small tutorial dataset composed of the first 128 images in [COCO](https://cocodataset.org/) train2017. These same 128 images are used for both training and validation to verify our training pipeline is capable of [overfitting](https://www.ultralytics.com/glossary/overfitting). [data/coco128.yaml](https://github.com/ultralytics/yolov5/blob/master/data/coco128.yaml), shown below, is the dataset config file that defines 1) the dataset root directory `path` and relative paths to `train` / `val` / `test` image directories (or `*.txt` files with image paths) and 2) a class `names` dictionary:
```yaml
# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
@ -183,11 +183,11 @@ You can use ClearML Data to version your dataset and then pass it to YOLOv5 simp
Training results are automatically logged with [Tensorboard](https://www.tensorflow.org/tensorboard) and [CSV](https://github.com/ultralytics/yolov5/pull/4148) loggers to `runs/train`, with a new experiment directory created for each new training as `runs/train/exp2`, `runs/train/exp3`, etc.
This directory contains train and val statistics, mosaics, labels, predictions and augmented mosaics, as well as metrics and charts including precision-recall (PR) curves and confusion matrices.
This directory contains train and val statistics, mosaics, labels, predictions and augmented mosaics, as well as metrics and charts including [precision](https://www.ultralytics.com/glossary/precision)-[recall](https://www.ultralytics.com/glossary/recall) (PR) curves and confusion matrices.
<img alt="Local logging results" src="https://github.com/ultralytics/docs/releases/download/0/local-logging-results.avif" width="1280">
Results file `results.csv` is updated after each epoch, and then plotted as `results.png` (below) after training completes. You can also plot any `results.csv` file manually:
Results file `results.csv` is updated after each [epoch](https://www.ultralytics.com/glossary/epoch), and then plotted as `results.png` (below) after training completes. You can also plot any `results.csv` file manually:
```python
from utils.plots import plot_results
@ -202,8 +202,8 @@ plot_results("path/to/results.csv") # plot 'results.csv' as 'results.png'
Once your model is trained you can use your best checkpoint `best.pt` to:
- Run [CLI](https://github.com/ultralytics/yolov5#quick-start-examples) or [Python](./pytorch_hub_model_loading.md) inference on new images and videos
- [Validate](https://github.com/ultralytics/yolov5/blob/master/val.py) accuracy on train, val and test splits
- [Export](./model_export.md) to TensorFlow, Keras, ONNX, TFlite, TF.js, CoreML and TensorRT formats
- [Validate](https://github.com/ultralytics/yolov5/blob/master/val.py) [accuracy](https://www.ultralytics.com/glossary/accuracy) on train, val and test splits
- [Export](./model_export.md) to [TensorFlow](https://www.ultralytics.com/glossary/tensorflow), Keras, ONNX, TFlite, TF.js, CoreML and TensorRT formats
- [Evolve](./hyperparameter_evolution.md) hyperparameters to improve performance
- [Improve](https://docs.roboflow.com/adding-data/upload-api?ref=ultralytics) your model by sampling real-world images and adding them to your dataset

4
docs/en/yolov5/tutorials/transfer_learning_with_frozen_layers.md
View file

@ -4,7 +4,7 @@ description: Learn to freeze YOLOv5 layers for efficient transfer learning, redu
keywords: YOLOv5, transfer learning, freeze layers, machine learning, deep learning, model training, PyTorch, Ultralytics
---
📚 This guide explains how to **freeze** YOLOv5 🚀 layers when **transfer learning**. Transfer learning is a useful way to quickly retrain a model on new data without having to retrain the entire network. Instead, part of the initial weights are frozen in place, and the rest of the weights are used to compute loss and are updated by the optimizer. This requires less resources than normal training and allows for faster training times, though it may also result in reductions to final trained accuracy.
📚 This guide explains how to **freeze** YOLOv5 🚀 layers when **[transfer learning](https://www.ultralytics.com/glossary/transfer-learning)**. Transfer learning is a useful way to quickly retrain a model on new data without having to retrain the entire network. Instead, part of the initial weights are frozen in place, and the rest of the weights are used to compute loss and are updated by the optimizer. This requires less resources than normal training and allows for faster training times, though it may also result in reductions to final trained accuracy.
## Before You Start
@ -121,7 +121,7 @@ train.py --batch 48 --weights yolov5m.pt --data voc.yaml --epochs 50 --cache --i
### Accuracy Comparison
The results show that freezing speeds up training, but reduces final accuracy slightly.
The results show that freezing speeds up training, but reduces final [accuracy](https://www.ultralytics.com/glossary/accuracy) slightly.
![Freezing training mAP50 results](https://github.com/ultralytics/docs/releases/download/0/freezing-training-map50-results.avif)