Sagar Mishra
Sagar Mishra
**Describe the bug** As per the discussion in #2697, it was [found](https://github.com/alan-turing-institute/sktime/issues/2697#issuecomment-1196290311) that `_load_provided_dataset()` must throw an error if `return_type` is not a reference to `pd.DataFrame` and `return_X_y=False`. This happens...
#### Reference Issues/PRs Fixes #3169 #### What does this implement/fix? Explain your changes. - Modify `_load_provided_dataset()` under `datasets/_data_io.py` to inhibit failures from data-type conversions. - Add `return_type` parameter to `load_UCR_UEA_dataset()`...
#### Reference Issues/PRs Fixes #3412. #### What does this implement/fix? Explain your changes. Add an `elif` block to read for `@targetlabel` in two places - `_read_header()` and `load_from_tsfile_to_dataframe()`. The implementation...
#### Reference Issues/PRs Fixes #3022 #### What does this implement/fix? Explain your changes. The Save/Load function for classical estimators is from #3336 and has been expanded for DL models along...
#### Reference Issues/PRs #### What does this implement/fix? Explain your changes. Initially, `test_persistence_via_pickle()` of `test_all_estimators.py` was skipped for all DL models since there were problems correctly serializing the objects. So,...
**Is your feature request related to a problem? Please describe.** Regression datasets from [tseregression](http://tseregression.org/) have `.ts` files that makes use of `@targetlabel` as opposed to `@classlabel` to check whether the...
#### Reference Issues/PRs Fixes #2043 #### What does this implement/fix? Explain your changes. Adding a reStructuredText file that will enable users to create their own `.ts` files, given a dataset....
#### Reference Issues/PRs Fixes #3372 #### What does this implement/fix? Explain your changes. Migrate TapNet Network and Classifier from sktime-dl into sktime. Changes are kept to a minimum. Remove a...
**Is your feature request related to a problem? Please describe.** Currently, we're using `sphinx-panel` as an [extension](https://github.com/alan-turing-institute/sktime/blob/3c56f535dc1d060c0053ab6c7f1a1a7199133539/docs/source/conf.py#L51) to build documentation. However, in the current state, `sphinx-panel` is not being actively...
**Is your feature request related to a problem? Please describe.** Migrating TapNet Classifier model from sktime-dl to sktime.