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

How to create 'Evaluation Result' directly from 'Table' data? #6813

Open
xianglun918 opened this issue May 30, 2024 · 4 comments · May be fixed by #6852
Open

How to create 'Evaluation Result' directly from 'Table' data? #6813

xianglun918 opened this issue May 30, 2024 · 4 comments · May be fixed by #6852

Comments

@xianglun918
Copy link

xianglun918 commented May 30, 2024

Many analysis widgets in Orange are expecting 'Evaluation Results' as inputs.
I need to analyze on results of a complicated DNN, and find that it is hard to generate 'Evaluation Results' based on model's output csv files.

How can I transform 'Table' directly to 'Evaluation Results', say 'Table' already has categorical features including 'true lable' and 'predictions'?

Something Like:
image

@janezd
Copy link
Contributor

janezd commented May 30, 2024

You can't. Evaluation results is not just a table but a more complicated structure, which also contains other data like subset indices, models etc.

@xianglun918
Copy link
Author

Understood. Maybe we can have a new widget which mocks 'Evaluation Results' with only 'Table data'?
It would be great if Orange can be used to analyze on Pytorch or tensorflow model's results. Many evaluation widgets are verfy useful.

@xianglun918
Copy link
Author

Something like this :)
image

@janezd
Copy link
Contributor

janezd commented May 31, 2024

@BlazZupan suggested that we could have a classifier that "predicts" based on values from a column. If the column is numeric, it can also apply logistic function.

@janezd janezd linked a pull request Jul 13, 2024 that will close this issue
3 tasks
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.

2 participants