before forecasting 3

atom/__init__.py

@@ -7,5 +7,5 @@
 
 """
 
-from atom.api import ATOMClassifier, ATOMModel, ATOMRegressor
+from atom.api import ATOMClassifier, ATOMForecaster, ATOMModel, ATOMRegressor
 from atom.utils import __version__

atom/api.py

@@ -520,7 +520,8 @@ class ATOMForecaster(BaseTransformer, ATOM):
 
     n_rows: int or float, default=1
         Subsample of the dataset to use. The cut is made from the head
-        of the dataset. The default value selects all rows.
+        of the dataset (older entries are dropped when sorted by date
+        ascending). The default value selects all rows.
 
         - If <=1: Fraction of the dataset to select.
         - If >1: Exact number of rows to select. Only if `arrays` is X

atom/basetransformer.py

@@ -636,6 +636,8 @@ def _subsample(df: DATAFRAME) -> DATAFRAME:
 
             if self.shuffle:
                 return df.iloc[random.sample(range(len(df)), k=n_rows)]
+            elif self.goal == "fc":
+                return df.iloc[-n_rows:]  # For time series, select from tail
             else:
                 return df.iloc[sorted(random.sample(range(len(df)), k=n_rows))]
 
@@ -847,7 +849,7 @@ def _has_data_sets(
                 return self.branch._data, self.branch._idx, self.branch._holdout
 
         elif len(arrays) == 1:
-            # arrays=(X,)
+            # arrays=(X,) or arrays=(y,) for forecasting
             sets = _no_data_sets(*self._prepare_input(arrays[0], y=y))
 
         elif len(arrays) == 2:

atom/utils.py

@@ -1972,9 +1972,6 @@ def get_custom_scorer(metric: str | Callable | Scorer) -> Scorer:
 def infer_task(y: PANDAS, goal: str = "class") -> str:
     """Infer the task corresponding to a target column.
 
-    If goal is provided, only look at number of unique values to
-    determine the classification task.
-
     Parameters
     ----------
     y: series or dataframe
@@ -1994,6 +1991,8 @@ def infer_task(y: PANDAS, goal: str = "class") -> str:
             return "regression"
         else:
             return "multioutput regression"
+    elif goal == "fc":
+        return "forecasting"
 
     if y.ndim > 1:
         if all(y[col].nunique() == 2 for col in y):
@@ -2467,7 +2466,6 @@ def prepare_df(out: FEATURES, og: DATAFRAME) -> DATAFRAME:
         )
         if isinstance(y, DATAFRAME_TYPES):
             y_new = prepare_df(y_new, y)
-
     elif "X" in params and X is not None and any(c in X for c in inc):
         # X in -> X out
         X_new = prepare_df(out, X)

docs_sources/dependencies.md

@@ -25,9 +25,9 @@ And operating systems:
 ATOM is built on top of several existing Python libraries. These
 packages are necessary for its correct functioning.
 
-* **[category-encoders](https://contrib.scikit-learn.org/categorical-encoding/index.html)** (>=2.5.1)
+* **[category-encoders](https://contrib.scikit-learn.org/categorical-encoding/index.html)** (>=2.6.1)
 * **[dill](https://pypi.org/project/dill/)** (>=0.3.6)
-* **[dagshub](https://github.com/DagsHub/client)** (<=0.2.10)
+* **[dagshub](https://github.com/DagsHub/client)** (>=0.2.18)
 * **[gplearn](https://gplearn.readthedocs.io/en/stable/index.html)** (>=0.4.2)
 * **[imbalanced-learn](https://imbalanced-learn.readthedocs.io/en/stable/api.html)** (>=0.10.1)
 * **[ipython](https://ipython.readthedocs.io/en/stable/)** (>=8.11.0)
@@ -87,8 +87,8 @@ using `pip install atom-ml[dev]`.
 * **[pytest](https://docs.pytest.org/en/latest/)** (>=7.2.1)
 * **[pytest-cov](https://pytest-cov.readthedocs.io/en/latest/)** (>=4.0.0)
 * **[pytest-xdist](https://github.com/pytest-dev/pytest-xdist)** (>=3.2.0)
-* **[scikeras](https://github.com/adriangb/scikeras)** (>=0.10.0)
-* **[tensorflow](https://www.tensorflow.org/learn)** (>=2.11.0)
+* **[scikeras](https://github.com/adriangb/scikeras)** (>=0.11.0)
+* **[tensorflow](https://www.tensorflow.org/learn)** (>=2.13.0)
 
 **Documentation**
 

docs_sources/user_guide/nomenclature.md

@@ -140,14 +140,15 @@ an estimator's fit method.
 
 <div id="task"><strong>task</strong></div>
 <div markdown style="margin: -1em 0 0 1.2em">
-One of the six supervised machine learning approaches that ATOM supports:
+One of the supervised machine learning approaches that ATOM supports:
 <ul style="line-height:1.2em;margin-top:-10px">
 <li><a href="https://en.wikipedia.org/wiki/Binary_classification">binary classification</a></li>
 <li><a href="https://en.wikipedia.org/wiki/Multiclass_classification">multiclass classification</a></li>
 <li><a href="https://scikit-learn.org/stable/modules/multiclass.html#multilabel-classification">multilabel classification</a></li>
 <li><a href="https://scikit-learn.org/stable/modules/multiclass.html#multiclass-multioutput-classification">multiclass-multioutput classification</a></li>
 <li><a href="https://en.wikipedia.org/wiki/Regression_analysis">regression</a></li>
 <li><a href="https://scikit-learn.org/stable/modules/multiclass.html#multioutput-regression">multioutput regression</a></li>
+<li><a href="https://en.wikipedia.org/wiki/Time_series">forecasting</a></li>
 </ul>
 </div>