ODSC-46457/refactor_forecast.csv (#349)

oracle · Sep 26, 2023 · 02c1f00 · 02c1f00
1 parent e106bd9
commit 02c1f00
Show file tree

Hide file tree

Showing 4 changed files with 90 additions and 14 deletions.
diff --git a/ads/opctl/operator/lowcode/forecast/model/arima.py b/ads/opctl/operator/lowcode/forecast/model/arima.py
@@ -111,6 +111,8 @@ def _build_model(self) -> pd.DataFrame:
 
             output_i["Date"] = outputs[f"{col}_{cat}"].index
             output_i["Series"] = cat
+            output_i["input_value"] = float("nan")
+            output_i[f"fitted_value"] = float("nan")
             output_i[f"forecast_value"] = outputs[f"{col}_{cat}"]["yhat"].values
             output_i[yhat_upper_name] = outputs[f"{col}_{cat}"]["yhat_upper"].values
             output_i[yhat_lower_name] = outputs[f"{col}_{cat}"]["yhat_lower"].values

diff --git a/ads/opctl/operator/lowcode/forecast/model/automlx.py b/ads/opctl/operator/lowcode/forecast/model/automlx.py
@@ -104,14 +104,22 @@ def _build_model(self) -> pd.DataFrame:
         # Merge the outputs from each model into 1 df with all outputs by target and category
         col = self.original_target_column
         output_col = pd.DataFrame()
+        yhat_lower_percentage = (
+            100 - (self.spec.confidence_interval_width or 0.5) * 100
+        ) // 2
+        yhat_upper_name = "p" + str(int(100 - yhat_lower_percentage))
+        yhat_lower_name = "p" + str(int(yhat_lower_percentage))
         for cat in self.categories:  # Note: add [:2] to restrict
             output_i = pd.DataFrame()
-            output_i[self.spec.datetime_column.name] = outputs[f"{col}_{cat}"]["ds"]
+            output_i["Date"] = outputs[f"{col}_{cat}"]["ds"]
             output_i["Series"] = cat
-            output_i[f"{col}_forecast"] = outputs[f"{col}_{cat}"]["yhat"]
-            output_i[f"{col}_forecast_upper"] = outputs[f"{col}_{cat}"]["yhat_upper"]
-            output_i[f"{col}_forecast_lower"] = outputs[f"{col}_{cat}"]["yhat_lower"]
+            output_i["input_value"] = float("nan")
+            output_i[f"fitted_value"] = float("nan")
+            output_i[f"forecast_value"] = outputs[f"{col}_{cat}"]["yhat"]
+            output_i[yhat_upper_name] = outputs[f"{col}_{cat}"]["yhat_upper"]
+            output_i[yhat_lower_name] = outputs[f"{col}_{cat}"]["yhat_lower"]
             output_col = pd.concat([output_col, output_i])
+
         # output_col = output_col.sort_values(self.spec.datetime_column.name).reset_index(drop=True)
         output_col = output_col.reset_index(drop=True)
         outputs_merged = pd.concat([outputs_merged, output_col], axis=1)

diff --git a/ads/opctl/operator/lowcode/forecast/model/neuralprophet.py b/ads/opctl/operator/lowcode/forecast/model/neuralprophet.py
@@ -219,14 +219,46 @@ def objective(trial):
 
             output_i["Date"] = outputs[f"{col}_{cat}"]["ds"]
             output_i["Series"] = cat
-            output_i[f"forecast_value"] = outputs[f"{col}_{cat}"]["yhat1"]
-            output_i[f"p{int(quantiles[1]*100)}"] = outputs[f"{col}_{cat}"][
-                f"yhat1 {quantiles[1]*100}%"
-            ]
-            output_i[f"p{int(quantiles[0]*100)}"] = outputs[f"{col}_{cat}"][
-                f"yhat1 {quantiles[0]*100}%"
-            ]
+            output_i[f"input_value"] = full_data_dict[f"{col}_{cat}"][f"{col}_{cat}"]
+
+            output_i[f"fitted_value"] = float("nan")
+            output_i[f"forecast_value"] = float("nan")
+            output_i[f"p{int(quantiles[1]*100)}"] = float("nan")
+            output_i[f"p{int(quantiles[0]*100)}"] = float("nan")
+
+            output_i.iloc[
+                : -self.spec.horizon.periods, output_i.columns.get_loc(f"fitted_value")
+            ] = (
+                outputs[f"{col}_{cat}"]["yhat1"]
+                .iloc[: -self.spec.horizon.periods]
+                .values
+            )
+            output_i.iloc[
+                -self.spec.horizon.periods :,
+                output_i.columns.get_loc(f"forecast_value"),
+            ] = (
+                outputs[f"{col}_{cat}"]["yhat1"]
+                .iloc[-self.spec.horizon.periods :]
+                .values
+            )
+            output_i.iloc[
+                -self.spec.horizon.periods :,
+                output_i.columns.get_loc(f"p{int(quantiles[1]*100)}"),
+            ] = (
+                outputs[f"{col}_{cat}"][f"yhat1 {quantiles[1]*100}%"]
+                .iloc[-self.spec.horizon.periods :]
+                .values
+            )
+            output_i.iloc[
+                -self.spec.horizon.periods :,
+                output_i.columns.get_loc(f"p{int(quantiles[0]*100)}"),
+            ] = (
+                outputs[f"{col}_{cat}"][f"yhat1 {quantiles[0]*100}%"]
+                .iloc[-self.spec.horizon.periods :]
+                .values
+            )
             output_col = pd.concat([output_col, output_i])
+
         # output_col = output_col.sort_values(operator.ds_column).reset_index(drop=True)
         output_col = output_col.reset_index(drop=True)
         outputs_merged = pd.concat([outputs_merged, output_col], axis=1)

diff --git a/ads/opctl/operator/lowcode/forecast/model/prophet.py b/ads/opctl/operator/lowcode/forecast/model/prophet.py
@@ -194,10 +194,44 @@ def objective(trial):
 
             output_i["Date"] = outputs[f"{col}_{cat}"]["ds"]
             output_i["Series"] = cat
-            output_i[f"forecast_value"] = outputs[f"{col}_{cat}"]["yhat"]
-            output_i[yhat_upper_name] = outputs[f"{col}_{cat}"]["yhat_upper"]
-            output_i[yhat_lower_name] = outputs[f"{col}_{cat}"]["yhat_lower"]
+            output_i["input_value"] = full_data_dict[f"{col}_{cat}"][f"{col}_{cat}"]
+
+            output_i[f"fitted_value"] = float("nan")
+            output_i[f"forecast_value"] = float("nan")
+            output_i[yhat_upper_name] = float("nan")
+            output_i[yhat_lower_name] = float("nan")
+
+            output_i.iloc[
+                : -self.spec.horizon.periods, output_i.columns.get_loc(f"fitted_value")
+            ] = (
+                outputs[f"{col}_{cat}"]["yhat"]
+                .iloc[: -self.spec.horizon.periods]
+                .values
+            )
+            output_i.iloc[
+                -self.spec.horizon.periods :,
+                output_i.columns.get_loc(f"forecast_value"),
+            ] = (
+                outputs[f"{col}_{cat}"]["yhat"]
+                .iloc[-self.spec.horizon.periods :]
+                .values
+            )
+            output_i.iloc[
+                -self.spec.horizon.periods :, output_i.columns.get_loc(yhat_upper_name)
+            ] = (
+                outputs[f"{col}_{cat}"]["yhat_upper"]
+                .iloc[-self.spec.horizon.periods :]
+                .values
+            )
+            output_i.iloc[
+                -self.spec.horizon.periods :, output_i.columns.get_loc(yhat_lower_name)
+            ] = (
+                outputs[f"{col}_{cat}"]["yhat_lower"]
+                .iloc[-self.spec.horizon.periods :]
+                .values
+            )
             output_col = pd.concat([output_col, output_i])
+
         # output_col = output_col.sort_values(self.spec.datetime_column.name).reset_index(drop=True)
         output_col = output_col.reset_index(drop=True)
         outputs_merged = pd.concat([outputs_merged, output_col], axis=1)