What is the correct way to Pseudo-label?

[samf1986]

❔Question

I'm trying to implement pseudo-labeling with yolov5 by predicting labels for unlabeled data at a high confidence level (e.g. 0.9). However most of my images have multiple true positives, so I'm concerned that if I add this "student" data as is, I will be adding many false negatives to the training set. I haven't found any resources addressing this issue, so my provision solution was to black out all regions of the images not within a pseudo-label bounding box before adding the images to the training set.

Is this correct? Do false negatives in the training set actually cause problems in Yolov5?

[github-actions[bot]]

👋 Hello @samf1986, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Google Colab and Kaggle 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 YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@samf1986 I don't understand your question, but in general yes, incorrect labels will negatively affect your training results.

[ShirleyHe2020]

❔Question

I'm trying to implement pseudo-labeling with yolov5 by predicting labels for unlabeled data at a high confidence level (e.g. 0.9). However most of my images have multiple true positives, so I'm concerned that if I add this "student" data as is, I will be adding many false negatives to the training set. I haven't found any resources addressing this issue, so my provision solution was to black out all regions of the images not within a pseudo-label bounding box before adding the images to the training set.

Is this correct? Do false negatives in the training set actually cause problems in Yolov5?

I have same interest in pseudo-labeling, do you have any progress up to now?

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.