Where to place the downloaded dataset, and how should I reference it without getting FileNotFoundError?

[weeix]

Hi all,

I downloaded a dataset from Roboflow, which I stored here:

datasets/
--mydataset/ <-- here
----train/...
----valid/...
----data.yaml
----README.dataset.txt
----README.roboflow.txt
train.ipynb
...

I ran the following block of code in my Jupyter notebook (train.ipynb):

from ultralytics import YOLO

model = YOLO("yolov8n.yaml")
model.train(data="mydataset/data.yaml", epochs=5)
model.val()

But I encountered the following errors:

FileNotFoundError                         Traceback (most recent call last)
File c:\Users\myusername\.virtualenvs\myproject-aimgoxY-\lib\site-packages\ultralytics\yolo\engine\trainer.py:127, in BaseTrainer.__init__(self, cfg, overrides)
    126 elif self.args.data.endswith('.yaml') or self.args.task in ('detect', 'segment'):
--> 127     self.data = check_det_dataset(self.args.data)
    128     if 'yaml_file' in self.data:

File c:\Users\myusername\.virtualenvs\myproject-aimgoxY-\lib\site-packages\ultralytics\yolo\data\utils.py:192, in check_det_dataset(dataset, autodownload)
    190 def check_det_dataset(dataset, autodownload=True):
    191     # Download, check and/or unzip dataset if not found locally
--> 192     data = check_file(dataset)
    194     # Download (optional)

File c:\Users\myusername\.virtualenvs\myproject-aimgoxY-\lib\site-packages\ultralytics\yolo\utils\checks.py:260, in check_file(file, suffix, download)
    259 if not files:
--> 260     raise FileNotFoundError(f"'{file}' does not exist")
    261 elif len(files) > 1:

FileNotFoundError: 'mydataset/data.yaml' does not exist

The above exception was the direct cause of the following exception:

FileNotFoundError                         Traceback (most recent call last)
Cell In[10], line 4
      1 from ultralytics import YOLO
...
--> 131     raise FileNotFoundError(emojis(f"Dataset '{self.args.data}' error ❌ {e}")) from e
    133 self.trainset, self.testset = self.get_dataset(self.data)
    134 self.ema = None

FileNotFoundError: Dataset 'mydataset/data.yaml' error  'mydataset/data.yaml' does not exist

Please advise me on where to place and reference datasets, or provide a link to the relevant documentation.

Thank you.

[weeix]

I was able to solve the issue by converting the relative path into an absolute path:

from ultralytics import YOLO

model = YOLO("yolov8n.yaml")
model.train(data="/fullpath/to/my/datasets/mydataset/data.yaml", epochs=5)

[glenn-jocher]

@weeix great to hear that you've resolved the issue! Using absolute paths can indeed prevent FileNotFoundError issues by explicitly specifying the location of your dataset. Remember that the path should be accessible from the environment where your code is running. If you encounter any further issues or have questions about using YOLOv8, feel free to consult our documentation or reach out again. Happy training! 🚀