YOLOv5 🚀 on AzureML

What is Azure?

Azure is Microsoft's cloud computing platform, designed to help organizations move their workloads to the cloud from on-premises data centers. With a full spectrum of cloud services including computing, databases, analytics, machine learning, and networking, users can pick and choose from these services to develop and scale new applications, or run existing applications, in the public cloud.

What is Azure Machine Learning (AzureML)?

Azure Machine Learning, commonly referred to as AzureML, is a fully managed cloud service that enables data scientists and developers to efficiently embed predictive analytics into their applications. AzureML offers a variety of services and capabilities aimed at making machine learning accessible, easy to use, and scalable, providing features like automated machine learning, drag-and-drop model training, and a robust Python SDK.

Prerequisites

Before getting started, you need an AzureML workspace. If you don't have one, you can create a new workspace by following Azure's official documentation.

Create a compute instance

From your AzureML workspace, select Compute > Compute instances > New, and select the instance with the resources you need.

Open a Terminal

From the Notebooks view, open a Terminal and select your compute.

Setup and run YOLOv5

Create a virtual environment

Create a conda environment with your preferred Python version:

conda create --name yolov5env -y python=3.10
conda activate yolov5env
conda install pip -y

Clone YOLOv5 repository

Clone the YOLOv5 repository with its submodules:

git clone https://github.com/ultralytics/yolov5
cd yolov5
git submodule update --init --recursive # You might see a message asking you to add your folder as a safe.directory

Install dependencies

Install the required dependencies:

pip install -r requirements.txt
pip install onnx

Perform YOLOv5 tasks

Train the YOLOv5 model:

python train.py --data coco128.yaml --weights yolov5s.pt --img 640

Validate the model for Precision, Recall, and mAP:

python val.py --weights yolov5s.pt --data coco128.yaml

Run inference on images:

python detect.py --weights yolov5s.pt --source path/to/images

Export models to other formats (like ONNX):

python export.py --weights yolov5s.pt --include onnx

Using a Notebook

If you prefer using a notebook instead of the terminal, you'll need to create a new Kernel and select it at the top of your notebook.

Create a new IPython kernel

From your compute terminal:

conda create --name yolov5env -y python=3.10
conda activate yolov5env
conda install pip ipykernel -y
python -m ipykernel install --user --name yolov5env --display-name "yolov5env"

When creating Python cells in your notebook, they will automatically use your custom environment. For bash cells, you need to activate your environment in each cell:

%%bash
source activate yolov5env
python val.py --weights yolov5s.pt --data coco128.yaml

📅 Created 1 year ago ✏️ Updated 12 days ago

Comments