TSCV: plot - finalize

src/pytimetk/crossvalidation/time_series_cv.py

@@ -255,17 +255,17 @@ def plot(
         y: pd.Series, 
         time_series: pd.Series = None,
         color_palette: Optional[Union[dict, list, str]] = None,
-        bar_height: float = 0.6,  # Height of each bar (adjust as needed)
+        bar_height: float = 0.3,  
         title: str = "Time Series Cross-Validation Plot",
         x_lab: str = "",
         y_lab: str = "Fold",
-        x_axis_date_labels: str = None,
+        x_axis_date_labels: str = None, 
         base_size: float = 11,
         width: Optional[int] = None,
         height: Optional[int] = None,
         engine: str = "plotly"
     ):
-        """Plots the cross-validation folds on a single plot with folds on the y-axis and dates on the x-axis using rectangle shapes.
+        """Plots the cross-validation folds on a single plot with folds on the y-axis and dates on the x-axis using filled Scatter traces.
 
         Arguments:
             y: The Pandas series of target values to plot.
@@ -307,7 +307,7 @@ def plot(
         # Calculate the vertical positions for each fold
         fold_positions = list(range(1, num_folds + 1))
 
-        # Enumerate through the splits and add rectangle shapes
+        # Enumerate through the splits and add filled Scatter traces
         for fold, (train_forecast, split_state) in enumerate(splits, start=1):
             train_indices, forecast_indices = train_forecast
 
@@ -338,36 +338,45 @@ def plot(
             y0 = fold - bar_height / 2
             y1 = fold + bar_height / 2
 
-            # Add rectangle for the training period
-            fig.add_shape(
-                type="rect",
-                x0=ts_date,
-                y0=y0,
-                x1=te_date,
-                y1=y1,
-                line=dict(width=0),
+            # Create coordinates for the training period rectangle
+            x_train = [ts_date, te_date, te_date, ts_date, ts_date]
+            y_train = [y0, y0, y1, y1, y0]
+
+            # Add Scatter trace for the training period
+            fig.add_trace(go.Scatter(
+                x=x_train,
+                y=y_train,
+                mode='lines',
+                fill='toself',
                 fillcolor=color_palette[0],
-                opacity=0.8,
-                layer='below',
-            )
-
-            # Add rectangle for the forecast period
-            fig.add_shape(
-                type="rect",
-                x0=fs_date,
-                y0=y0,
-                x1=fe_date,
-                y1=y1,
                 line=dict(width=0),
+                hoverinfo='text',
+                hoverlabel=dict(font_size=base_size * 0.8),
+                text=f"Fold {fold}<br>Train Period<br>{ts_date.date()} to {te_date.date()}",
+                showlegend=False,
+            ))
+
+            # Create coordinates for the forecast period rectangle
+            x_forecast = [fs_date, fe_date, fe_date, fs_date, fs_date]
+            y_forecast = [y0, y0, y1, y1, y0]
+
+            # Add Scatter trace for the forecast period
+            fig.add_trace(go.Scatter(
+                x=x_forecast,
+                y=y_forecast,
+                mode='lines',
+                fill='toself',
                 fillcolor=color_palette[1],
-                opacity=0.8,
-                layer='below',
-            )
+                line=dict(width=0),
+                hoverinfo='text',
+                hoverlabel=dict(font_size=base_size * 0.8),
+                text=f"Fold {fold}<br>Forecast Period<br>{fs_date.date()} to {fe_date.date()}",
+                showlegend=False,
+            ))
 
-            # Calculate midpoint for annotation
+            # Optionally, add text annotations for each fold
             train_midpoint = ts_date + (te_date - ts_date) / 2
 
-            # Optionally, add text annotations for each fold
             fig.add_trace(go.Scatter(
                 x=[train_midpoint],
                 y=[fold],
@@ -419,6 +428,7 @@ def plot(
 
 
 
+
 # class TimeSeriesCV:
 #     """Generates tuples of train_idx, test_idx pairs
 #     Assumes the MultiIndex contains levels 'symbol' and 'date'