aeon
aeon copied to clipboard
A toolkit for machine learning from time series
### Describe the feature or idea you want to propose following dev discussion about codecov, looking at lack of testing coverage for networks. I think its because it lacks tests...
### Describe the bug For PAM-LCSS and PAM-EDR sometimes the correct amount of clusters are not assigned. Specifically LCSS on Adiac and EDR on OliveOil assign incorrect number of clusters....
### Describe the issue Some of the estimators that use joblib for parallelization use process-based backend, while other use threads-based backend. Ideally, we want this to be a parameter tunable...
#### Reference Issues/PRs Fixes #1777 #### What does this implement/fix? Explain your changes. Add class_weight parameter to classifiers that use sklearn based estimators ### PR checklist - [x] The PR...
### Describe the feature or idea you want to propose as I understand it, deep learning estimators save a model by default, as is standard practice in the field. This...
### Describe the issue The `tf.keras.layers.Reshape` was introduced in `clustering/deep_learnin` to make the input shape of the encoder equal to the output shape of the decoder which is already being...
This is PR refactors TDE to use SFA_FAST. - Added a new class `_tde_fast.py` with `FastTemporalDictionaryEnsemble` - Removed Multivariate support for now - Runtime still not optimal
#### Reference Issues/PRs fixes #1724 #### What does this implement/fix? Explain your changes. Removes the version bound on `keras` (if CI passes).
Updates the version bound on numpy to include 2.0.0.
### Describe the issue linked to the documentation just having a play with similarity search, some examples in the docstring would be helpful. ### Suggest a potential alternative/fix _No response_