[BUG] dealing with sklearn 1.2 deprecation warnings and solving the first part

[hmtbgc]

Reference Issues/PRs

Fixes the first part of #2143 : replacing float variable names by str in pandas.DataFrame passed.

What does this implement/fix? Explain your changes.

I trace the warning and find out in summarize.py, quantiles are presented as "float" while other features are represented as "str", and that is the root of this warning. So I transform these float numbers to string.

Does your contribution introduce a new dependency? If yes, which one?

No.

What should a reviewer concentrate their feedback on?

All unit tests in 'classification' directory have passed. And here is the output:

======================================== test session starts ========================================
platform linux -- Python 3.8.12, pytest-7.0.1, pluggy-1.0.0
rootdir: /root/sktime_dev/sktime, configfile: setup.cfg
plugins: cov-3.0.0, xdist-2.5.0, forked-1.4.0
collected 1 item                                                                                    

classification/feature_based/_summary_classifier.py .                                         [100%]

========================================= 1 passed in 4.14s ======================================                       


(sktime-dev) root@x86_64-conda-linux-gnu:sktime/sktime ‹main*›# 
(sktime-dev) root@x86_64-conda-linux-gnu:sktime/sktime ‹main*›# 
(sktime-dev) root@x86_64-conda-linux-gnu:sktime/sktime ‹main*›# pytest classification/feature_based/tests/test_summary_classifier.py
================================================== test session starts ===================================================
platform linux -- Python 3.8.12, pytest-7.0.1, pluggy-1.0.0
rootdir: /root/sktime_dev/sktime, configfile: setup.cfg
plugins: cov-3.0.0, xdist-2.5.0, forked-1.4.0
collected 2 items                                                                                                        

classification/feature_based/tests/test_summary_classifier.py ..                                                   [100%]

=================================================== 2 passed in 4.80s ====================================================
(sktime-dev) root@x86_64-conda-linux-gnu:sktime/sktime ‹main*›# pytest classification/tests/test_all_classifiers.py
================================================== test session starts ===================================================
platform linux -- Python 3.8.12, pytest-7.0.1, pluggy-1.0.0
rootdir: /root/sktime_dev/sktime, configfile: setup.cfg
plugins: cov-3.0.0, xdist-2.5.0, forked-1.4.0
collected 90 items                                                                                                       

classification/tests/test_all_classifiers.py ...........
.......................................................... [ 76%]
.....................                                                                                              [100%]

================================================== 90 passed in 44.68s ===================================================

For tests/test_all_estimators.py, here is the output:

==================================================== warnings summary ====================================================
sktime/tests/test_all_estimators.py: 14 warnings
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/scipy/stats/morestats.py:897: RuntimeWarning: invalid value encountered in log
    logdata = np.log(data)

sktime/tests/test_all_estimators.py: 14 warnings
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/scipy/stats/morestats.py:906: RuntimeWarning: invalid value encountered in power
    variance = np.var(data**lmb / lmb, axis=0)

sktime/tests/test_all_estimators.py: 64 warnings
  /root/sktime_dev/sktime/sktime/performance_metrics/forecasting/_functions.py:1543: FutureWarning: In the percentage error metric functions the default argument symmetric=True is changing to symmetric=False in v0.12.0.
    warn(

sktime/tests/test_all_estimators.py: 16 warnings
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/sklearn/decomposition/_pca.py:499: RuntimeWarning: invalid value encountered in true_divide
    explained_variance_ = (S ** 2) / (n_samples - 1)

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[Prophet-ForecasterFitPredictUnivariateNoX]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/pyximport/pyximport.py:51: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
    import imp

sktime/tests/test_all_estimators.py: 28 warnings
  /root/sktime_dev/sktime/sktime/transformations/series/boxcox.py:377: RuntimeWarning: invalid value encountered in power
    x_ratio = x_std / x_mean ** (1 - lmb)

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[TimeSeriesKMeans-ClustererFitPredict]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/utils/utils.py:156: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    def to_time_series_dataset(dataset, dtype=numpy.float):

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[TimeSeriesKMeans-ClustererFitPredict]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/utils/cast.py:15: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    def to_sklearn_dataset(dataset, dtype=numpy.float, return_dim=False):

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[TimeSeriesKMeans-ClustererFitPredict]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/metrics/utils.py:9: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    compute_diagonal=True, dtype=numpy.float, *args, **kwargs):

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[TimeSeriesKMeans-ClustererFitPredict]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/metrics/sax.py:2: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    from .cysax import cydist_sax

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[TimeSeriesKMeans-ClustererFitPredict]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/metrics/sax.py:2: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    from .cysax import cydist_sax

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_updates_state[TimeSeriesKMeans-ClustererFitPredict]
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/metrics/__init__.py:24: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    from .cycc import cdist_normalized_cc, y_shifted_sbd_vec

sktime/tests/test_all_estimators.py: 50137 warnings
  /root/miniconda/envs/sktime-dev/lib/python3.8/site-packages/tslearn/utils/utils.py:149: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
  Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
    if ts_out.dtype != numpy.float:

sktime/tests/test_all_estimators.py::TestAllEstimators::test_fit_idempotent[LogTransformer-TransformerFitTransformPanelUnivariateWithClassY]
sktime/tests/test_all_estimators.py::TestAllEstimators::test_methods_do_not_change_state[LogTransformer-TransformerFitTransformPanelUnivariateWithClassY]
sktime/tests/test_all_estimators.py::TestAllEstimators::test_methods_have_no_side_effects[LogTransformer-TransformerFitTransformPanelUnivariateWithClassY]
sktime/tests/test_all_estimators.py::TestAllEstimators::test_persistence_via_pickle[LogTransformer-TransformerFitTransformPanelUnivariateWithClassY]
  /root/sktime_dev/sktime/sktime/transformations/series/boxcox.py:250: RuntimeWarning: invalid value encountered in log
    Xt = np.log(X)

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
============================= 3679 passed, 243 skipped, 50284 warnings in 402.83s (0:06:42) ==============================

There is no warning about "FutureWarning: Feature names only support names that are all strings. Got feature names with dtypes: ['float', 'str']. An error will be raised in 1.2."

Any other comments?

The above warnings need to be solved too.

PR checklist

For all contributions

• I've added myself to the list of contributors.
• Optionally, I've updated sktime's CODEOWNERS to receive notifications about future changes to these files.
• I've added unit tests and made sure they pass locally.

For new estimators

• I've added the estimator to the online documentation.
• I've updated the existing example notebooks or provided a new one to showcase how my estimator works.

[TonyBagnall]

looks good to me, thanks.