Multi-run confusion matrix is missing class

[susbrock]

I am working on a 3 classes classification problem and my confusion matrix is getting messed up. It does not show class 1.

The test accuracy for all three classes is 95% and all three classes are actually in the dataset.

        # log data for the confusion matrix
        wandb.log({"conf_mat" : wandb.plot.confusion_matrix(probs=None,
                        y_true=y_test_true, preds=y_test_pred,
                        class_names=labels)})

Any idea how to solve this issue?

[artsiom]

Hi Susanne!

Could I see a link to your workspace?

Cheers!
Artsiom

[artsiom]

Hi Susanne,

We wanted to follow up with you regarding your support request as we have not heard back from you. Please let us know if we can be of further assistance or if your issue has been resolved.

Best,
Weights & Biases

[artsiom]

Hi Susanne, since we have not heard back from you we are going to close this request. If you would like to re-open the conversation, please let us know!