Eduardo Fonseca

Documentation of iterative_train_test_split incomplete

5

`iterative_train_test_split` is briefly documented [here](http://scikit.ml/api/skmultilearn.model_selection.iterative_stratification.html#module-skmultilearn.model_selection.iterative_stratification) (at the bottom), but the input params `X`, `y` are not explained. I tried passing `y`as a list of lists, encoding the labels as categorical...

Set the order parameter in iterative_train_test_split

1

Thanks for the effort on buliding this library. I have 2 questions: **a)** The `IterativeStratification` method has a parameter `order`: _the order of label relationship to take into account when...

Make response type meanings of validation task more accessible

1

In order to gather high quality annotations, users must clearly understand the meaning of the response types (PP, PNP, NP, U). Currently they are explained in: 1. Task instructions 2....

Explore FSD release & report wrong annotations

4

Apart from exploring the content of a current state of a dataset and/or of its available releases, we should also allow users to report faulty audio samples or wrong annotations....

Considering the unconstrained nature of crowd-sourcing, the PP and PNP definitions in the modal could be further clarified in the modal, with the help of examples. There have been 2...

Currently, lists of candidates for sound categories are the result of a tag-matching process. Freesound tags were manually assigned to the target categories, which were automatically populated with the Freesound...

Annotation tasks brainstorming for FSD

7

Currently, we have only one task available for FSD: a **validation task**, which consists of validating annotations that were automatically generated through a mapping. More specifically, for a given sound...

Hi, thanks a lot for the theme. I like the design, simple and clean. Although the theme works very well in mobile phones, I just realized that the main picture...