Yakau Bubnou comments

Results 29 comments of


                                            Yakau Bubnou

Discrepancy between keras-metrics and scikit-learn

@david-b-6, thank you for the issue. In the code above I don't see how you print the metrics from `keras-metrics` package, there is only evaluation through `sklearn`.

Discrepancy between keras-metrics and scikit-learn

I've extended unit test to perform cross-validation with sklearn metrics: #46

Discrepancy between keras-metrics and scikit-learn

It seems I understand your confusion now, let me explain. `keras-metrics` are implemented as regular layers of the model, so they are part of the model's execution graph. So whenever...

Precision and Recall and F1-Score > 1 ?

Hi, @hachreak, thank you for posting the issue. Are the returned values also bigger than 1?

Precision and Recall and F1-Score > 1 ?

@hachreak, I see, if it is possible could you show runnable sample of code and data you feed to the model, this will much help troubleshooting. Most common thing why...

Precision and Recall and F1-Score > 1 ?

The are two possible ways to get negative _false positive_ counter: 1. Feed training data where the output (actual Y value) is not binary, so this code strikes: ```py class...

Upgrading keras gives decimal values for true_positive, false_positive, true_negative, false_negative

@amolgharat, true, seems builds (and tests) are failing for Keras `2.3.0` *and tensorflow 2.0.0*.

Upgrading keras gives decimal values for true_positive, false_positive, true_negative, false_negative

I've added in #49 a deprecation warning for this library that support will be dropped since Keras `2.3.0` (as they already have a replacement for all metrics).

Metrics for multi-label..

@Sathibhu, thank you for the issue! Currently ```keras-metrics``` supports only averaging for recall. You can use one of: * binary_average_recall * categorical_average_recall * sparse_categorical_average_recall I hope there will be averaging...

Non-Sensical Print Values For Metrics

Hi @onurdanaci, thank you for posting an issue. Is it possible to put the model and data with which I could reproduce the issue? From my previous investigations it's most...