WARNING: Dataset not found.

[danielsoy]

Hello ppl, This yolov5 notebook, gives me this errror:

WARNING: Dataset not found, nonexistent paths: ['/content/valid/images']

I can see the images and its labels, here:/content/datasets/template_yolov5_API-1/valid/images

But the execution of the notebook, is stuck at step 3, because it is unzipping the valid images and labels in one directory, but then tries to retrieve them, from another nonexistent location.

https://colab.research.google.com/github/roboflow-ai/yolov5-custom-training-tutorial/blob/main/yolov5-custom-training.ipynb

Additional context

[github-actions[bot]]

👋 Hello @danielsoy, 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://ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python>=3.6.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

$ git clone https://github.com/ultralytics/yolov5
$ cd yolov5
$ 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), validation (val.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

@danielsoy hi, thanks for raising this issue! Is there any additional information you can provide to help us reproduce the problem?

I'll loop in our Roboflow colleague @Jacobsolawetz who might be able to help.

[danielsoy]

Hi Glenn, thanks for your answer.

The issue will pop running the jupyter netbook that is linked in this youtube video: Custom Training Tutorial and Notebook https://www.youtube.com/watch?v=x0ThXHbtqCQ&t=1s&ab_channel=Roboflow.

On the comments section, you can read the comments from other guys, with the same problem.

El sáb, 9 oct 2021 a las 14:07, Glenn Jocher @.***>) escribió:

@danielsoy https://github.com/danielsoy hi, thanks for raising this issue! is there any additional information you can provide to help us reproduce the problem?

I'll loop in our Roboflow colleague @Jacobsolawetz https://github.com/Jacobsolawetz who might be able to help.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ultralytics/yolov5/issues/5107#issuecomment-939328834, or unsubscribe https://github.com/notifications/unsubscribe-auth/ALVI3RLPI6DZPQCNBR6DFELUGBZFBANCNFSM5FUW2TJA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[danielsoy]

I could make it work by editing the data.yaml file to make it point to the path where the needed files are located.

In my case it is: train: ../datasets/anotado-2/train/images val: ../datasets/anotado-2/valid/images

nc: 2 names: ['bad', 'ok']

It is not a professional solution, because i need to edit the file, for every new dataset with a different name than anotado-2....

At this moment, the model is being trained, lets see what happens...

[glenn-jocher]

@danielsoy great, thanks for the feedback! I'll pass it on to Roboflow to see if there's some fixes that can be made.

[Jacobsolawetz]

@danielsoy thanks for the info, this will help us debug speedily. I'll post here when the fix is alive.

[Jacobsolawetz]

@danielsoy @glenn-jocher the fix for this is live - long story short, I had forgotten to push some fixes to the roboflow pip package after I had been adapting dataset download to some more models

[glenn-jocher]

@Jacobsolawetz thanks!

@danielsoy can you try again after a new pip install -U roboflow and see if this solves the problem for you?

[danielsoy]

Its well known that Robooflow crew delivers!

I will try pip i... n see how it goes, thanks again.

[danielsoy]

Yes, it works! All the way, down to the end of the notebook now.

Kudos to Jacobsolawetz and glenn-jocher !🤓

[ghost]

Hi guys, same problem happened to me and pip install -U roboflow is not working.

[Emyyr]

I think it is about roboflow train-valid-test split. pip install -U roboflow not working. Maybe you can try create new dataset folder after download on roboflow.

[salehghulamqasim]

The problem with me was i didnt have any validation images so roboflow maybe didnt create path for it somehow but it would show inside data.yaml and the notebook would say dataset not found.

what i did is i tried different dataset which had validation images and it worked like a charm

so in short the problem was i didnt have the folder valid/images since i didnt even have any validation images in roboflow at first place, fix that , add some validation images then export and try

GOODLUCK

[glenn-jocher]

@salehghulamqasim thanks for sharing your experience and workaround! It's great to hear that you were able to identify the issue and resolve it. If you have any further questions or need assistance, feel free to ask. Good luck with your continued work on YOLOv5!

[notanadrian]

i had the same problem dataset not found but when i check the directory in the google collab, the images file are there. btw im using yolov7

[glenn-jocher]

@notanadrian it sounds like there might be a mismatch between the expected directory path in your code and the actual location where your dataset resides. This is a common issue that can occur for various reasons, such as incorrect dataset path specifications or changes in the directory structure.

For YOLOv7, which is a different repository and not directly related to the Ultralytics YOLOv5 project, I recommend double-checking the dataset path specified in your training script or notebook. Ensure that it exactly matches the location of your dataset in Google Colab. You might also need to adjust the dataset path in any configuration files or scripts you are using.

If you're still encountering issues, consider reaching out to the YOLOv7 repository maintainers or community for more specific guidance related to YOLOv7. Good luck with your project!