Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Task]: Drift between code and quick-start-toturial #984

Closed
TorAP opened this issue Mar 27, 2023 · 4 comments · Fixed by #997
Closed

[Task]: Drift between code and quick-start-toturial #984

TorAP opened this issue Mar 27, 2023 · 4 comments · Fixed by #997

Comments

@TorAP
Copy link

TorAP commented Mar 27, 2023

What is the motivation for this task?

When I follow this Training with Custom Data -guide: https://openvinotoolkit.github.io/anomalib/how_to_guides/train_custom_data.html I cannot run the training: python tools/train.py --config custom_padim.yaml. Where the config is as in the tutorial. I get the this error: AttributeError: 'NoneType' object has no attribute 'keys'. When I guess is most likely becuase there is no data, but the I checked the paths and the dataset hazelnut_toy.

Describe the solution you'd like

Updated guide..

Additional context

No response

@Ravindu987
Copy link
Contributor

Hi
I also encountered the same error and resolved it this way.

First make sure all configs under dataset is indented. If you copy and paste from the guide they might not be indented.

Then I had to add these configs as I got Missing key errors for these
normalization: imagenet
test_split_mode: from_dir
val_split_ratio: 0.5

Also if you downloaded the dataset from the provided link, the extracted folder is named hazelnut_toy and in the given config it is named Hazelnut_toy. Change one of them to match.

Let me know if it works after making these changes.

@Ravindu987
Copy link
Contributor

@samet-akcay shall I make these changes and do a PR or is someone already on it

@samet-akcay
Copy link
Contributor

@Ravindu987, if you could make the changes, that'd be great. Currently there is no one working on it

@Ravindu987
Copy link
Contributor

Will do the necessary.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants