Caching images problem

[Bilgee]

I am trying to train a detection model using my custom dataset.
When I set image-size at 640, it works. But when I set mage-size at 1280, it shows the following error. I already tried --workers 0, but there was no difference.
yolov5: v4.0
env: google colab

Transferred 500/506 items from yolov5m.pt
Scaled weight_decay = 0.0005
Optimizer groups: 86 .bias, 86 conv.weight, 83 other
Scanning '/content/fish/labels/train.cache' for images and labels... 2368 found, 0 missing, 0 empty, 0 corrupted: 100% 2368/2368 [00:00<00:00, 26000292.86it/s]
Caching images (6.5GB): 100% 2368/2368 [04:12<00:00, 9.37it/s]
Scanning '/content/fish/labels/val.cache' for images and labels... 591 found, 0 missing, 0 empty, 0 corrupted: 100% 591/591 [00:00<00:00, 4063661.74it/s]
Caching images (1.3GB): 77% 458/591 [00:56<00:14, 9.14it/s]Traceback (most recent call last):
File "/usr/lib/python3.6/multiprocessing/pool.py", line 720, in next
item = self._items.popleft()
IndexError: pop from an empty deque

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "train.py", line 520, in
train(hyp, opt, device, tb_writer, wandb)
File "train.py", line 202, in train
rank=-1, world_size=opt.world_size, workers=opt.workers, pad=0.5)[0]
File "/content/yolov5/utils/datasets.py", line 71, in create_dataloader
image_weights=image_weights)
File "/content/yolov5/utils/datasets.py", line 432, in init
for i, x in pbar:
File "/usr/local/lib/python3.6/dist-packages/tqdm/std.py", line 1104, in iter
for obj in iterable:
File "/usr/lib/python3.6/multiprocessing/pool.py", line 724, in next
self._cond.wait(timeout)
File "/usr/lib/python3.6/threading.py", line 295, in wait
waiter.acquire()
File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/signal_handling.py", line 66, in handler
_error_if_any_worker_fails()
RuntimeError: DataLoader worker (pid 1192) is killed by signal: Killed.
Caching images (1.3GB): 77% 458/591 [00:58<00:17, 7.77it/s]

[github-actions]

👋 Hello @Bilgee, 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 Notebook with free GPU:
• Kaggle Notebook with free GPU: https://www.kaggle.com/ultralytics/yolov5
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Docker Image https://hub.docker.com/r/ultralytics/yolov5. 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]

@Bilgee don't --cache if your system doesn't have the RAM to support it.

[Bilgee]

@glenn-jocher thank you for reply. Yolov5 caching is memory caching? I think that caching makes special files like tfrecords, right?

Scanning '/content/fish/labels/train.cache' for images and labels... 2368 found, 0 missing, 0 empty, 0 corrupted: 100% 2368/2368 [00:00<00:00, 26000292.86it/s]
Caching images (6.5GB): 100% 2368/2368 [04:12<00:00, 9.37it/s]
Scanning '/content/fish/labels/val.cache' for images and labels... 591 found, 0 missing, 0 empty, 0 corrupted: 100% 591/591 [00:00<00:00, 4063661.74it/s]
Caching images (1.3GB): 77% 458/591 [00:56<00:14, 9.14it/s]Traceback (most recent call last):

it failed here. Google Colab RAM is almost 13GB.

[Bilgee]

I think that after creating train.cache or val.cache, the code does not release memory, so that it fails to allocate the more memory.

[Bilgee]

@glenn-jocher

[phantom-cloud]

Try setting the image size between 640 and 1000 seems you are low on memory