shuffle index

[Gorazor]

🐛 Bug

A bug exsits in func create_dataloader in utils/dataset.py. If training not in DDP mode, no shuffle operation in dataloader. if you training data configed as list: list: [path1/images/, path2/images/] and each path's images are different classes or sources. This could be a problem. (this may not be a big problem if you use mosaic since for mosaic augment other image pathes are random selected)

  ......
  sampler = torch.utils.data.distributed.DistributedSampler(dataset) if rank != -1 else None
  loader = torch.utils.data.DataLoader if image_weights else InfiniteDataLoader
  dataloader = loader(dataset,
                      batch_size=batch_size,
                      num_workers=nw,
                      sampler=sampler,
                      pin_memory=True,
                      collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn)

 ......

solution

add an argument ''is_train' for shuffle

def create_dataloader(path, imgsz, batch_size, stride, opt, hyp=None, augment=False, cache=False, pad=0.0, rect=False,
                      rank=-1, world_size=1, workers=8, image_weights=False, quad=False, prefix='', **is_train=True**):
    ......
    sampler = torch.utils.data.distributed.DistributedSampler(dataset) if rank != -1 else None
    loader = torch.utils.data.DataLoader if image_weights else InfiniteDataLoader
    shuffle = True if is_train and sampler is None else False
    dataloader = loader(dataset,
                        batch_size=batch_size,
                        num_workers=nw,
                        shuffle=shuffle,
                        sampler=sampler,
                        pin_memory=True,
                        collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn)
   ......

[github-actions]

👋 Hello @Gorazor, 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]

@Gorazor yes this is a good point, which has also been brought up before. Though as you mention if you use mosaic=1.0 in your hyperparameters then 75% of the images will still be shuffled.

@Gorazor have you tried to train some baseline datasets against the default code with your proposed change? A quick test might be to finetune VOC with YOLOv5s using the existing code and your proposed code and compare the result. In Colab this usually only takes a few hours, i.e.:
https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb#scrollTo=BSgFCAcMbk1R&line=1&uniqifier=1

# VOC
for b, m in zip([64], ['yolov5s']):  # zip(batch_size, model)
  !python train.py --batch {b} --weights {m}.pt --data voc.yaml --epochs 50 --cache --img 512 --nosave --hyp hyp.finetune.yaml --project VOC --name {m}

[github-actions]

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.

[glenn-jocher]

@Gorazor good news 😃! Your original issue may now be fixed ✅ in PR #5623 by @werner-duvaud. This PR turns on shuffling in the YOLOv5 training DataLoader by default, which was missing until now. This works for all training formats: CPU, Single-GPU, Multi-GPU DDP.

train_loader, dataset = create_dataloader(train_path, imgsz, batch_size // WORLD_SIZE, gs, single_cls,
                                          hyp=hyp, augment=True, cache=opt.cache, rect=opt.rect, rank=LOCAL_RANK,
                                          workers=workers, image_weights=opt.image_weights, quad=opt.quad,
                                          prefix=colorstr('train: '), shuffle=True)  # <--- NEW

I evaluated this PR against master on VOC finetuning for 50 epochs, and the results show a slight improvement in most metrics and losses, particularly in objectness loss and mAP@0.5, perhaps indicating that the shuffle addition may help delay overtraining.

https://wandb.ai/glenn-jocher/VOC

To receive this update:

• Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
• PyTorch Hub – Force-reload model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• Notebooks – View updated notebooks
• Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!