Gilles Vandewiele
Gilles Vandewiele
It seems that the Contributing.MD file has been updated, but no longer includes a section that explains how to run the tests locally. I think that is an important section...
On MacOS, the `tslearn.datasets` does not work out-of-the-box. In order to make it work, you need to apply the following steps: 1) Go to your finder 2) run "/Apps/Python/Install Certificates.command"....
**Describe the bug** When passing a list to silhouette_score as such: ```python3 silhouette_score(X, tsk.labels_, metric='euclidean') ``` with `X` a pure python list. It crashes: ``` --------------------------------------------------------------------------- AttributeError Traceback (most recent...
I was browsing through the code and stumbled on this line: https://github.com/tslearn-team/tslearn/blob/7ead159338c91112341cdb2e4f8d2d59c75129a8/tslearn/shapelets.py#L402 Shouldn't this become self.max_size? Is there a unit test in place already that specifically tests for the behaviour...
It seems that the `decision_function` and `predict_proba` of `TimeSeriesSVC` do not produce perfect rank correlation all the time (see log below). One possible fix would be to patch the `make_blobs`...
[pyts](https://github.com/johannfaouzi/pyts) contains some cool feature extractors such as WEASEL & BOSS that are quite unique to that package. A wrapper around pyts would be a nice addition! [The BOSS is...
[sktime](https://github.com/alan-turing-institute/sktime) contains an efficient implementation of ROCKET, which achieves state-of-the-art performances on many time series classification datasets! A wrapper would be very cool! [ROCKET: exceptionally fast and accurate time series...
First of all, big props for this project! A big help in constructing big stacking models. It would maybe be interesting to get some visualizations in the tool, like e.g....
Small to-do note for my own: fix inverse walking & hash ALL nodes in walks and then at transform-time, hash the provided entities before going to the lookup-table.
## 🚀 Feature Currently, the sampling techniques are rather slow. The depth-first-search (DFS) algorithm can potentially be improved by making use of smarter data structures and techniques such as caching....