Tutorial: Yolov8 in PyTorch

Tutorial: YOLO v8 in PyTorch

https://docs.ultralytics.com/quickstart/#install-ultralytics

Ultralytics YOLOv8 is a cutting-edge, state-of-the-art (SOTA) model that builds upon the success of previous YOLO versions and introduces new features and improvements to further boost performance and flexibility. Y

img

Documentation and Github

See the YOLOv8 Docs for full documentation on training, validation, prediction and deployment.

Also, you can visit the github repository: https://github.com/ultralytics/ultralytics

---

Installation of Yolov8

Ultralytics provides various installation methods including pip, conda, and Docker. Install YOLOv8 via the ultralytics pip package for the latest stable release or by cloning the Ultralytics GitHub repository for the most up-to-date version.

Requirement

Install necessary packages such as Python, Numpy, PyTorch, CUDA and more

For installations of requirements, read for more detail instructions

• Python >=3.8

• PyTorch>=1.8

• opencv-python>=4.6.0

• matplotlib>=3.3.0

• and more. See requirements

1)Install Yolov8 via pip package

First, create a new environment for YOLOv8 in Anaconda Prompt.

• e.g. $myENV$ = yolov8

You can also make an exact copy of the existing environment by creating a clone

• If you already have an environment named py39, clone it as yolov8

conda create --name yolov8 --clone py39

Activate the environment and Install YOLOv8 with pip to get stable packages.

Also, install the latest ONNX

conda activate yolov8
pip install ultralytics
pip install onnx

2)Install Yolov8 via pip package(numpy 1.26 ver.)

If There is an OpenCV DLL issue, follow the setup intructions below to configure your environment

(The DLL issue might be caused by an incompatible Numpy version[your numpy ver. numpy 2.x])

conda create -n yolov8 python=3.9.12
conda activate yolov8

conda install -c anaconda numpy==1.26
pip install opencv-python matplotlib

conda install pytorch=2.1 torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia

pip install torchsummary
pip install ultralytics
pip install onnx  

Check for YOLO Installation

After the installation, you can check the saved source code and libs of YOLOv8 in the local folder :

\USER\anaconda3\envs\yolov8\Lib\site-packages\ultralytics

Now, lets run simple prediction examples to check the YOLO installation.

In Anaconda Prompt, activate yolov8 environment.

Then, move directory to the working directory. Here, the result images will be saved.

• Example: C:\Users\ykkim\source\repos\DLIP\yolov8\

conda activate yolov8

cd C:\Users\ykkim\source\repos\DLIP\yolov8

Run a Detection Example

In the Anaconda prompt, type the following command to predict a simple image.

yolo predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg'

The result will be saved in the project folder \runs\detect\predict\

Example: C:\Users\ykkim\source\repos\DLIP\yolov8\runs\detect\predict\

Run a Segmentation Example

Predict a YouTube video using a pretrained segmentation model at image size 320:

yolo predict model=yolov8n-seg.pt source='https://youtu.be/LNwODJXcvt4' imgsz=320

The result will be saved in the project folder \runs\segment\predict\

---

Using YOLOv8 with Python : Example Codes

In the project folder, create a new python code file

• Project Folder: \source\repos\DLIP\yolov8\

• Activate yolov8 environment in Anaconda Prompt

A list of useful commands for YOLOv8

from ultralytics import YOLO

# Create a new YOLO model from scratch
model = YOLO('yolov8n.yaml')

# Load a pretrained YOLO model (recommended for training)
model = YOLO('yolov8n.pt')

# Train the model using the 'coco8.yaml' dataset for 3 epochs
results = model.train(data='coco8.yaml', epochs=3)

# Evaluate the model's performance on the validation set
results = model.val()

# Perform object detection on an image using the model
results = model('https://ultralytics.com/images/bus.jpg')

Example: Detection Inference

Read Doc of Prediction with YOLO for more examples

Download the dataset file and save in the project folder

• bus.jpg

You can download the COCO pretrained models such as YOLOv8n and more.

https://docs.ultralytics.com/datasets/detect/coco/

Inference one image

Create a new python source file in the project folder

• Yolo8-Inference-Ex1.py

#########################################################
# YOLO v8  Tutorial : Prediction  Ex1
#
# Load Pretrained Model and Display the Annoted Results
#
#########################################################

from ultralytics import YOLO
import cv2
from matplotlib import pyplot as plt



# Load a pretrained YOLO model (recommended for training)
model = YOLO('yolov8n.pt')

# Inference Source - a single source
src = cv2.imread("bus.jpg")


# Perform object detection on an image using the model
result = model.predict(source=src, save=True, save_txt=True)  # save predictions as labels


# View result   
for r in result:

    # print the Boxes object containing the detection bounding boxes        
    print(r.boxes)  
    
    # Show results to screen (not recommended)
    print("result.show()")
    r.show()    

   
    # Plot results image    
    print("result.plot()") 
    dst = r.plot()      # return BGR-order numpy array
    cv2.imshow("result plot",dst)         
    

    # Plot the original image (NParray)
    print("result.orig_img")
    cv2.imshow("result orig",r.orig_img)   


# Save results to disk
r.save(filename='result.jpg')
cv2.waitKey(0)



##########################################################################################

Inference of multiple images

Create a new python source file in the project folder

• Yolo8-Inference-Ex2.py

For multiple input source images, you can copy bus.jpg as bus2.jpg.

#########################################################
# YOLO v8  Tutorial : Prediction  Ex2
#
# Load Pretrained Model and Display the Annoted Results (multiple images)
#
#########################################################


from ultralytics import YOLO
import cv2
from matplotlib import pyplot as plt


# Load a pretrained YOLO model (recommended for training)
model = YOLO('yolov8n.pt')


# Inference Source - multiple images
# Perform object detection on images using the model
results = model(['bus.jpg', 'bus2.jpg'])  # return a list of Results objects


# View results
for i, r in enumerate(results):

    # Plot results image    
    dst = r.plot()      # return BGR-order numpy array
    cv2.imshow("r.plot",dst)       
   

    # Save results to disk
    r.save(filename=f'results{i}.jpg')
    
cv2.waitKey(0)

Inference on Webcam stream

Create a new python source file in the project folder

• Yolo8-Inference-Webcam-Ex3.py

#########################################################
# YOLO v8  Tutorial : Prediction  Ex3
#
# Stream Video Prediction 
#
# This script will run predictions on each frame of the video
# visualize the results, and display them in a window. 
# The loop can be exited by pressing 'q'.
#########################################################


import cv2 as cv
from ultralytics import YOLO

# Load the YOLOv8 model
model = YOLO('yolov8n.pt')

# Open the video camera no.0
cap = cv.VideoCapture(0)

# If not success, exit the program
if not cap.isOpened():
    print('Cannot open camera')


# Loop through the video frames
while cap.isOpened():
    # Read a frame from the video
    success, frame = cap.read()

    if success:
        # Run YOLOv8 inference on the frame
        results = model(frame)

        # Visualize the results on the frame
        annotated_frame = results[0].plot()

        # Display the annotated frame
        cv.imshow("YOLOv8 Inference", annotated_frame)

        # Break the loop if 'q' is pressed
        if cv.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        # Break the loop if the end of the video is reached
        break

# Release the video capture object and close the display window
cap.release()
cv.destroyAllWindows()

Example: Train