drop python 3.9 support

.github/CONTRIBUTING.md

@@ -103,12 +103,12 @@ review and accept your changes.
 If your contribution requires a new library dependency:
 
 * Double-check that the new dependency is easy to install via pip and Anaconda.
-* The library should support Python 3.9 and higher.
+* The library should support Python 3.10 and 3.11.
 * Make sure the code works with the latest version of the library.
 * Update the dependencies in the documentation.
 * Add the library with the minimum required version to `pyproject.toml`.
 
 After submitting your pull request, GitHub will automatically run the tests
 on your changes and make sure that the updated code builds successfully.
-The checks run on Python 3.9, 3.10 and 3.11, on Ubuntu and Windows. We also
+The checks run on Python 3.10 and 3.11, on Ubuntu and Windows. We also
 use services that automatically check code style and test coverage.

.github/workflows/config.yml

@@ -88,7 +88,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest, windows-latest]
-        python-version: ["3.9", "3.10", "3.11"]
+        python-version: ["3.10", "3.11"]
     steps:
       - name: Check out source repository
         uses: actions/checkout@v3

README.md

@@ -30,9 +30,9 @@
 --- | ---
 **Repository** | [![Project Status: Active](https://www.repostatus.org/badges/latest/active.svg)](https://www.repostatus.org/#active) [![Conda Recipe](https://img.shields.io/badge/recipe-atom--ml-green.svg)](https://anaconda.org/conda-forge/atom-ml) [![License: MIT](https://img.shields.io/github/license/tvdboom/ATOM)](https://opensource.org/licenses/MIT) [![Downloads](https://static.pepy.tech/badge/atom-ml)](https://pepy.tech/project/atom-ml)
 **Release** | [![pdm-managed](https://img.shields.io/badge/pdm-managed-blueviolet)](https://pdm.fming.dev) [![PyPI version](https://img.shields.io/pypi/v/atom-ml)](https://pypi.org/project/atom-ml/) [![Conda Version](https://img.shields.io/conda/vn/conda-forge/atom-ml.svg)](https://anaconda.org/conda-forge/atom-ml) [![DOI](https://zenodo.org/badge/195069958.svg)](https://zenodo.org/badge/latestdoi/195069958)
-**Compatibility** | [![Python 3.9\|3.10\|3.11](https://img.shields.io/badge/python-3.9%20%7C%203.10%20%7C%203.11-blue?logo=python)](https://www.python.org) [![Conda Platforms](https://img.shields.io/conda/pn/conda-forge/atom-ml.svg)](https://anaconda.org/conda-forge/atom-ml)
+**Compatibility** | [![Python 3.10\|3.11](https://img.shields.io/badge/python-3.10%20%7C%203.11-blue?logo=python)](https://www.python.org) [![Conda Platforms](https://img.shields.io/conda/pn/conda-forge/atom-ml.svg)](https://anaconda.org/conda-forge/atom-ml)
 **Build status** | [![Build Status](https://github.com/tvdboom/ATOM/workflows/ATOM/badge.svg)](https://github.com/tvdboom/ATOM/actions) [![Azure Pipelines](https://dev.azure.com/conda-forge/feedstock-builds/_apis/build/status/atom-ml-feedstock?branchName=master)](https://dev.azure.com/conda-forge/feedstock-builds/_build/latest?definitionId=10822&branchName=master) [![codecov](https://codecov.io/gh/tvdboom/ATOM/branch/master/graph/badge.svg)](https://codecov.io/gh/tvdboom/ATOM)
-**Code analysis** | [![PEP8](https://img.shields.io/badge/code%20style-pep8-orange.svg)](https://www.python.org/dev/peps/pep-0008/) [![Imports: isort](https://img.shields.io/badge/%20imports-isort-%231674b1?style=flat&labelColor=ef8336)](https://pycqa.github.io/isort/)
+**Code analysis** | [![PEP8](https://img.shields.io/badge/code%20style-pep8-orange.svg)](https://www.python.org/dev/peps/pep-0008/) [![Imports: isort](https://img.shields.io/badge/%20imports-isort-%231674b1?style=flat&labelColor=ef8336)](https://pycqa.github.io/isort/) [![flake8](https://img.shields.io/badge/flake8-checked-blue)](https://flake8.pycqa.org/en/latest/) [![mypy](http://www.mypy-lang.org/static/mypy_badge.svg)](http://www.mypy-lang.org/)
 
 
 <br><br>

atom/atom.py

@@ -22,7 +22,7 @@
 import numpy as np
 import pandas as pd
 from beartype import beartype
-from beartype.typing import Any, Callable, Iterator, Literal, TypeVar
+from beartype.typing import Any, Callable, Iterator, Literal, Sequence, TypeVar
 from joblib.memory import Memory
 from pandas._typing import DtypeObj
 from scipy import stats
@@ -55,9 +55,8 @@
     FloatZeroToOneInc, Index, IndexSelector, Int, IntLargerEqualZero,
     IntLargerTwo, IntLargerZero, MetricConstructor, ModelsConstructor, NItems,
     NJobs, NormalizerStrats, NumericalStrats, Operators, Pandas, PrunerStrats,
-    RowSelector, Scalar, ScalerStrats, Sequence, SequenceTypes, Series,
-    TargetSelector, Transformer, TSIndexTypes, VectorizerStarts, Verbose,
-    Warnings, XSelector, YSelector,
+    RowSelector, Scalar, ScalerStrats, Series, TargetSelector, Transformer,
+    TSIndex, VectorizerStarts, Verbose, Warnings, XSelector, YSelector,
 )
 from atom.utils.utils import (
     ClassMap, DataConfig, DataContainer, Goal, adjust_verbosity, bk,
@@ -541,7 +540,7 @@ def eda(
 
         if isinstance(rows, str):
             rows_c = [(self.branch._get_rows(rows), rows)]
-        elif isinstance(rows, SequenceTypes):
+        elif isinstance(rows, Sequence):
             rows_c = [(self.branch._get_rows(r), r) for r in rows]
         elif isinstance(rows, dict):
             rows_c = [(self.branch._get_rows(v), k) for k, v in rows.items()]
@@ -938,7 +937,7 @@ def stats(self, _vb: Int = -2, /):
         for set_ in ("train", "test", "holdout"):
             if (data := getattr(self, set_)) is not None:
                 self._log(f"{set_.capitalize()} set size: {len(data)}", _vb)
-                if isinstance(self.branch.train.index, TSIndexTypes):
+                if isinstance(self.branch.train.index, TSIndex):
                     self._log(f" --> From: {min(data.index)}  To: {max(data.index)}", _vb)
 
         self._log("-" * 37, _vb)

atom/basemodel.py

@@ -67,11 +67,10 @@
 from atom.plots import RunnerPlot
 from atom.utils.constants import DF_ATTRS
 from atom.utils.types import (
-    HT, Backend, Bool, DataFrame, DataFrameTypes, Engine, FHSelector, Float,
-    FloatTypes, FloatZeroToOneExc, Int, IntLargerEqualZero, IntTypes,
-    MetricConstructor, NJobs, Pandas, PredictionMethod, Predictor, RowSelector,
-    Scalar, Scorer, Sequence, Stages, TargetSelector, Verbose, Warnings,
-    XSelector, YSelector,
+    HT, Backend, Bool, DataFrame, Engine, FHSelector, Float, FloatZeroToOneExc,
+    Int, IntLargerEqualZero, MetricConstructor, NJobs, Pandas,
+    PredictionMethod, Predictor, RowSelector, Scalar, Scorer, Sequence, Stages,
+    TargetSelector, Verbose, Warnings, XSelector, YSelector,
 )
 from atom.utils.utils import (
     ClassMap, DataConfig, Goal, PlotCallback, ShapExplanation, Task,
@@ -282,7 +281,7 @@ def __contains__(self, item: str) -> bool:
         return item in self.dataset
 
     def __getitem__(self, item: Int | str | list) -> Pandas:
-        if isinstance(item, IntTypes):
+        if isinstance(item, Int):
             return self.dataset[self.columns[item]]
         else:
             return self.dataset[item]  # Get a subset of the dataset
@@ -412,7 +411,7 @@ def _trial_to_est(self, params: dict[str, Any]) -> dict[str, Any]:
         """
         return deepcopy(params)
 
-    def _get_est(self, **params) -> Predictor:
+    def _get_est(self, params: dict[str, Any]) -> Predictor:
         """Get the estimator instance.
 
         Use the multioutput meta-estimator if the estimator has
@@ -423,8 +422,8 @@ def _get_est(self, **params) -> Predictor:
 
         Parameters
         ----------
-        **params
-            Unpacked hyperparameters for the estimator.
+        params: dict
+            Hyperparameters for the estimator.
 
         Returns
         -------
@@ -819,7 +818,7 @@ def _get_score(
         else:
             if threshold and self.task.is_binary and hasattr(self, "predict_proba"):
                 y_true, y_pred = self._get_pred(rows, attr="predict_proba")
-                if isinstance(y_pred, DataFrameTypes):
+                if isinstance(y_pred, DataFrame):
                     # Update every target column with its corresponding threshold
                     for i, value in enumerate(threshold):
                         y_pred.iloc[:, i] = (y_pred.iloc[:, i] > value).astype("int")
@@ -962,9 +961,7 @@ def fit_model(
                 trial.set_user_attr(key, value)
 
             # Create estimator instance with trial-specific hyperparameters
-            estimator = self._get_est(
-                **{**self._est_params, **self._trial_to_est(params)}
-            )
+            estimator = self._get_est(self._est_params | self._trial_to_est(params))
 
             # Check if the same parameters have already been evaluated
             for t in trial.study.get_trials(False, states=(TrialState.COMPLETE,))[::-1]:
@@ -977,13 +974,13 @@ def fit_model(
                 # Follow the same stratification strategy as atom
                 cols = self._config.get_stratify_columns(self.og.train, self.og.y_train)
 
-                if isinstance(cv := self._ht["cv"], IntTypes):
+                if isinstance(cv := self._ht["cv"], Int):
                     if self.task.is_forecast:
                         if cv == 1:
                             splitter = SingleWindowSplitter(range(1, len(self.og.test)))
                         else:
                             splitter = TimeSeriesSplit(n_splits=cv)
-                    elif isinstance(self._ht["cv"], IntTypes):
+                    elif isinstance(self._ht["cv"], Int):
                         # We use ShuffleSplit instead of K-fold because it
                         # works with n_splits=1 and multioutput stratification
                         if cols is None:
@@ -1169,7 +1166,7 @@ def fit(self, X: DataFrame | None = None, y: Pandas | None = None):
         # Assign estimator if not done already
         if self._estimator is None:
             self._check_est_params()
-            self._estimator = self._get_est(**{**self._est_params, **self.best_params})
+            self._estimator = self._get_est(self._est_params | self.best_params)
 
         self._estimator = self._fit_estimator(
             estimator=self.estimator,
@@ -1808,7 +1805,7 @@ def inference(*X) -> Scalar | str | list[Scalar | str]:
             conv = lambda elem: elem.item() if hasattr(elem, "item") else elem
 
             y_pred = self.inverse_transform(y=self.predict([X], verbose=0), verbose=0)
-            if isinstance(y_pred, DataFrameTypes):
+            if isinstance(y_pred, DataFrame):
                 return [conv(elem) for elem in y_pred.iloc[0, :]]
             else:
                 return conv(y_pred[0])
@@ -2031,7 +2028,7 @@ def evaluate(
             Scores of the model.
 
         """
-        if isinstance(threshold, FloatTypes):
+        if isinstance(threshold, Float):
             threshold_c = [threshold] * self.branch._data.n_cols  # Length=n_targets
         elif len(threshold) != self.branch._data.n_cols:
             raise ValueError(

atom/baserunner.py

@@ -32,10 +32,9 @@
 from atom.pipeline import Pipeline
 from atom.utils.constants import DF_ATTRS
 from atom.utils.types import (
-    Bool, DataFrame, DataFrameTypes, FloatZeroToOneExc, Int, IntTypes,
-    MetricConstructor, Model, ModelSelector, ModelsSelector, Pandas,
-    RowSelector, Scalar, Segment, SegmentTypes, SequenceTypes, Series,
-    YSelector,
+    Bool, DataFrame, FloatZeroToOneExc, Int, MetricConstructor, Model,
+    ModelSelector, ModelsSelector, Pandas, RowSelector, Scalar, Segment,
+    Series, YSelector,
 )
 from atom.utils.utils import (
     ClassMap, DataContainer, Task, bk, check_is_fitted, composed, crash,
@@ -53,7 +52,7 @@ class BaseRunner(BaseTracker, metaclass=ABCMeta):
 
     def __getstate__(self) -> dict[str, Any]:
         # Store an extra attribute with the package versions
-        return {**self.__dict__, "_versions": get_versions(self._models)}
+        return self.__dict__ | {"_versions": get_versions(self._models)}
 
     def __setstate__(self, state: dict[str, Any]):
         versions = state.pop("_versions", None)
@@ -109,7 +108,7 @@ def __getitem__(self, item: Int | str | list) -> Any:
                 "This instance has no dataset annexed to it. "
                 "Use the run method before calling __getitem__."
             )
-        elif isinstance(item, IntTypes):
+        elif isinstance(item, Int):
             return self.dataset[self.columns[item]]
         elif isinstance(item, str):
             if item in self._branches:
@@ -289,7 +288,7 @@ def _set_index(self, df: DataFrame, y: Pandas | None) -> DataFrame:
             pass
         elif self._config.index is False:
             df = df.reset_index(drop=True)
-        elif isinstance(self._config.index, IntTypes):
+        elif isinstance(self._config.index, Int):
             if -df.shape[1] <= self._config.index <= df.shape[1]:
                 df = df.set_index(df.columns[int(self._config.index)], drop=True)
             else:
@@ -415,7 +414,7 @@ def _no_data_sets(
                 )
             data = _subsample(data)
 
-            if isinstance(self._config.index, SequenceTypes):
+            if isinstance(self._config.index, Sequence):
                 if len(self._config.index) != len(data):
                     raise IndexError(
                         "Invalid value for the index parameter. Length of "
@@ -486,7 +485,7 @@ def _no_data_sets(
 
             except ValueError as ex:
                 # Clarify common error with stratification for multioutput tasks
-                if "least populated class" in str(ex) and isinstance(y, DataFrameTypes):
+                if "least populated class" in str(ex) and isinstance(y, DataFrame):
                     raise ValueError(
                         "Stratification for multioutput tasks is applied over all target "
                         "columns, which results in a least populated class that has only "
@@ -572,7 +571,7 @@ def _has_data_sets(
                 )
 
             # If the index is a sequence, assign it before shuffling
-            if isinstance(self._config.index, SequenceTypes):
+            if isinstance(self._config.index, Sequence):
                 len_data = len(train) + len(test)
                 if holdout is not None:
                     len_data += len(holdout)
@@ -605,7 +604,7 @@ def _has_data_sets(
         # Process input arrays ===================================== >>
 
         if len(arrays) == 0:
-            if self._goal.name == "forecast" and not isinstance(y, (IntTypes, str)):
+            if self._goal.name == "forecast" and not isinstance(y, Int | str):
                 # arrays=() and y=y for forecasting
                 sets = _no_data_sets(*self._check_input(y=y))
             elif not self.branch._container:
@@ -626,7 +625,7 @@ def _has_data_sets(
                 X_train, y_train = self._check_input(arrays[0][0], arrays[0][1])
                 X_test, y_test = self._check_input(arrays[1][0], arrays[1][1])
                 sets = _has_data_sets(X_train, y_train, X_test, y_test)
-            elif isinstance(arrays[1], (*IntTypes, str)) or n_cols(arrays[1]) == 1:
+            elif isinstance(arrays[1], Int | str) or n_cols(arrays[1]) == 1:
                 if not self._goal.name == "forecast":
                     # arrays=(X, y)
                     sets = _no_data_sets(*self._check_input(arrays[0], arrays[1]))
@@ -730,11 +729,11 @@ def _get_models(
         exc: list[Model] = []
         if models is None:
             inc = self._models.values()
-        elif isinstance(models, SegmentTypes):
+        elif isinstance(models, Segment):
             inc = get_segment(self._models, models)
         else:
             for model in lst(models):
-                if isinstance(model, IntTypes):
+                if isinstance(model, Int):
                     try:
                         inc.append(self._models[model])
                     except KeyError:

atom/basetrainer.py

@@ -26,7 +26,7 @@
 from atom.data_cleaning import BaseTransformer
 from atom.models import MODELS, CustomModel
 from atom.plots import RunnerPlot
-from atom.utils.types import Model, SequenceTypes
+from atom.utils.types import Model, Sequence
 from atom.utils.utils import (
     ClassMap, DataConfig, Goal, Task, check_dependency, get_custom_scorer, lst,
     sign, time_to_str,
@@ -104,7 +104,7 @@ def _check_param(self, param: str, value: Any) -> dict:
             Parameter with model names as keys.
 
         """
-        if isinstance(value, SequenceTypes):
+        if isinstance(value, Sequence):
             if len(value) != len(self._models):
                 raise ValueError(
                     f"Invalid value for the {param} parameter. The length "
@@ -296,7 +296,7 @@ def _prepare_parameters(self):
                                     )
                             elif k not in self._models:
                                 self._ht_params[key][name][k] = v
-            elif key in {**sign(create_study), **sign(Study.optimize)}:
+            elif key in sign(create_study) | sign(Study.optimize):
                 self._ht_params[key] = {k: value for k in self._models.keys()}
             else:
                 raise ValueError(

atom/basetransformer.py

@@ -36,9 +36,8 @@
 from sklearn.utils.validation import check_memory
 
 from atom.utils.types import (
-    Backend, Bool, BoolTypes, DataFrame, DataFrameTypes, Engine, Estimator,
-    Int, IntLargerEqualZero, IntTypes, Pandas, SequenceTypes, Severity,
-    Verbose, Warnings, XSelector, YSelector,
+    Backend, Bool, DataFrame, Engine, Estimator, Int, IntLargerEqualZero,
+    Pandas, Seq1dim, Severity, Verbose, Warnings, XSelector, YSelector,
 )
 from atom.utils.utils import crash, flt, n_cols, sign, to_df, to_pandas
 
@@ -209,7 +208,7 @@ def warnings(self) -> Warnings:
     @warnings.setter
     @beartype
     def warnings(self, value: Bool | Warnings):
-        if isinstance(value, BoolTypes):
+        if isinstance(value, Bool):
             self._warnings: Warnings = "once" if value else "ignore"
         else:
             self._warnings = value
@@ -502,7 +501,7 @@ def _check_input(
                         )
 
         # Prepare target column
-        if isinstance(y, (dict, *SequenceTypes, *DataFrameTypes)):
+        if isinstance(y, dict | Seq1dim | DataFrame):
             if isinstance(y, dict):
                 yt = to_df(deepcopy(y), index=getattr(Xt, "index", None))
                 if n_cols(yt) == 1:
@@ -516,7 +515,7 @@ def _check_input(
                         for col in y:
                             if col in Xt.columns:
                                 targets.append(col)
-                            elif isinstance(col, IntTypes):
+                            elif isinstance(col, Int):
                                 if -Xt.shape[1] <= col < Xt.shape[1]:
                                     targets.append(Xt.columns[int(col)])
                                 else:
@@ -541,7 +540,7 @@ def _check_input(
                     data=deepcopy(yt),
                     index=getattr(Xt, "index", None),
                     name=flt(name) if name is not None else "target",
-                    columns=name if isinstance(name, SequenceTypes) else default_cols,
+                    columns=name if isinstance(name, Sequence) else default_cols,
                 )
 
             # Check X and y have the same indices
@@ -558,7 +557,7 @@ def _check_input(
             else:
                 raise ValueError("X can't be None when y is a string.")
 
-        elif isinstance(y, IntTypes):
+        elif isinstance(y, Int):
             if Xt is None:
                 raise ValueError("X can't be None when y is an int.")