Remove use of deprecated start_index from Categorify (#1134)

Co-authored-by: rnyak <16246900+rnyak@users.noreply.github.com>

Author: oliverholworthy

examples/usecases/ecommerce-session-based-next-item-prediction-for-fashion.ipynb

@@ -468,7 +468,7 @@
    "source": [
     "%%time\n",
     "item_features_names = ['f_' + str(col) for col in [47, 68]]\n",
-    "cat_features = [['item_id', 'purchase_id']] + item_features_names >> nvt.ops.Categorify(start_index=1)\n",
+    "cat_features = [['item_id', 'purchase_id']] + item_features_names >> nvt.ops.Categorify()\n",
     "\n",
     "features = ['session_id', 'timestamp'] + cat_features"
    ]

examples/usecases/transformers-next-item-prediction.ipynb

@@ -352,7 +352,7 @@
     "\n",
     "categorical_features = (['city_id', 'booker_country', 'hotel_country'] +\n",
     "                         weekday_checkin + weekday_checkout\n",
-    "                       ) >> ops.Categorify(start_index=1)  \n",
+    "                       ) >> ops.Categorify()\n",
     "\n",
     "groupby_features = categorical_features + ['utrip_id', 'checkin'] >> ops.Groupby(\n",
     "    groupby_cols=['utrip_id'],\n",

merlin/datasets/ecommerce/booking/dataset.py

@@ -183,7 +183,7 @@ def default_booking_transformation(**kwargs):
     Returns:
         Workflow: NVTabular workflow
     """
-    cat = lambda: nvt.ops.Categorify(start_index=1)  # noqa: E731
+    cat = lambda: nvt.ops.Categorify()  # noqa: E731
 
     df_season = get_lib().DataFrame(
         {"month": range(1, 13), "season": ([0] * 3) + ([1] * 3) + ([2] * 3) + ([3] * 3)}