Confusion Matrix gets me confused

[pfcouto]

Search before asking

• I have searched the YOLOv8 issues and discussions and found no similar questions.

Question

So, I know this question as been asked before, but I want to add something to those questions. As shows in the image below, I have 317 "Background" being predicted as "Tomato", which is incorrect and should not be happening. Is there a way to know, in which image, which label is misplaced and therefore adding up to these 317? Thanks!

Additional

No response

[github-actions]

👋 Hello @pfcouto, thank you for your interest in Ultralytics YOLOv8 🚀! We recommend a visit to the Docs for new users where you can find many Python and CLI usage examples and where many of the most common questions may already be answered.

If this is a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset image examples and training logs, and verify you are following our Tips for Best Training Results.

Join the vibrant Ultralytics Discord 🎧 community for real-time conversations and collaborations. This platform offers a perfect space to inquire, showcase your work, and connect with fellow Ultralytics users.

Install

Pip install the ultralytics package including all requirements in a Python>=3.8 environment with PyTorch>=1.8.

pip install ultralytics

Environments

YOLOv8 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all Ultralytics CI tests are currently passing. CI tests verify correct operation of all YOLOv8 Modes and Tasks on macOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@pfcouto hi there! It looks like you're aiming to identify the specific images associated with those misclassified labels. You can accomplish this by examining the predictions during validation and recording those that are incorrectly identified as "Tomato" when they should be "Background".

Here’s a quick code snippet on how you can do this using Python, assuming you're using YOLOv8 for predictions:

from ultralytics import YOLO

# Load your model
model = YOLO('path/to/your/model.pt')

# Perform validation 
results = model.val()

# Inspect misclassifications
for img, output in zip(results.imgs, results.pred):
    for box in output.boxes:
        if box.cls == 'Tomato' and results.labels[results.imgs.index(img)].cls != 'Tomato':
            print(f'Misclassified Tomato in image: {img}, true label: {results.labels[results.imgs.index(img)].cls}')

This snippet runs a validation step with your trained model and checks for any 'Tomato' classifications where the true label is anything but 'Tomato'. Important images are then printed for further examination. Make sure to adjust the classes as per your setup. Use this as a starting point and tailor it as needed. Hope this helps clear things up! 🍅🔍