comments from review

Author: isabelizimm

vetiver/handlers/base.py

@@ -20,14 +20,14 @@ def __init__(
 
 
 @singledispatch
-def create_handler(model, ptype_data):
+def create_handler(model, prototype_data):
     """check for model type to handle prediction
 
     Parameters
     ----------
     model: object
         Description of parameter `x`.
-    ptype_data : object
+    prototype_data : object
         An object with information (data) whose layout is to be determined.
 
     Returns
@@ -63,7 +63,7 @@ class BaseHandler:
     ----------
     model :
         a trained model
-    ptype_data :
+    prototype_data :
         An object with information (data) whose layout is to be determined.
     """
 
@@ -73,9 +73,9 @@ def __init_subclass__(cls, **kwargs):
         with suppress(AttributeError, NameError):
             create_handler.register(cls.model_class(), cls)
 
-    def __init__(self, model, ptype_data):
+    def __init__(self, model, prototype_data):
         self.model = model
-        self.ptype_data = ptype_data
+        self.prototype_data = prototype_data
 
     def describe(self):
         """Create description for model"""
@@ -93,21 +93,21 @@ def create_meta(
 
         return meta
 
-    def construct_ptype(self):
+    def construct_prototype(self):
         """Create data prototype for a model
 
         Parameters
         ----------
-        ptype_data : pd.DataFrame, np.ndarray, or None
-            Training data to create ptype
+        prototype_data : pd.DataFrame, np.ndarray, or None
+            Training data to create prototype
 
         Returns
         -------
-        ptype : pd.DataFrame or None
+        prototype : pd.DataFrame or None
             Zero-row DataFrame for storing data types
         """
-        ptype = vetiver_create_ptype(self.ptype_data)
-        return ptype
+        prototype = vetiver_create_ptype(self.prototype_data)
+        return prototype
 
     def handler_startup():
         """Include required packages for prediction
@@ -117,7 +117,7 @@ def handler_startup():
         """
         ...
 
-    def handler_predict(self, input_data, check_ptype):
+    def handler_predict(self, input_data, check_prototype):
         """Generates method for /predict endpoint in VetiverAPI
 
         The `handler_predict` function executes at each API call. Use this
@@ -128,8 +128,8 @@ def handler_predict(self, input_data, check_ptype):
         ----------
         input_data:
             Data used to generate prediction
-        check_ptype:
-            If type should be checked against `ptype` or not
+        check_prototype:
+            If type should be checked against `prototype` or not
 
         Returns
         -------
@@ -144,8 +144,8 @@ def handler_predict(self, input_data, check_ptype):
 
 
 @create_handler.register
-def _(model: base.BaseHandler, ptype_data):
-    if model.ptype_data is None and ptype_data is not None:
-        model.ptype_data = ptype_data
+def _(model: base.BaseHandler, prototype_data):
+    if model.prototype_data is None and prototype_data is not None:
+        model.prototype_data = prototype_data
 
     return model

vetiver/handlers/torch.py

@@ -58,7 +58,7 @@ def handler_predict(self, input_data, check_prototype):
         if not torch_exists:
             raise ImportError("Cannot import `torch`.")
         if check_prototype:
-            input_data = np.array(input_data, dtype=np.array(self.ptype_data).dtype)
+            input_data = np.array(input_data, dtype=np.array(self.prototype_data).dtype)
             prediction = self.model(torch.from_numpy(input_data))
 
         # do not check ptype

vetiver/server.py

@@ -43,21 +43,22 @@ def __init__(
         self,
         model: VetiverModel,
         check_prototype: bool = True,
-        check_ptype: bool = None,
         app_factory=FastAPI,
+        **kwargs,
     ) -> None:
         self.model = model
         self.app_factory = app_factory
         self.app = app_factory()
 
-        if check_ptype is not None:
-            check_prototype = check_ptype
+        if "check_ptype" in kwargs:
+            check_prototype = kwargs.pop("check_ptype")
             warn(
                 "argument for checking input data prototype has changed to "
                 "check_prototype, from check_ptype",
                 DeprecationWarning,
                 stacklevel=2,
             )
+
         self.check_prototype = check_prototype
 
         self._init_app()
@@ -151,9 +152,6 @@ def vetiver_post(self, endpoint_fx: Callable, endpoint_name: str = None, **kw):
         if not endpoint_name:
             endpoint_name = endpoint_fx.__name__
 
-        if hasattr(self.model, "ptype"):
-            self.model.prototype = self.model.ptype
-
         if self.check_prototype is True:
 
             @self.app.post(urljoin("/", endpoint_name), name=endpoint_name)

vetiver/tests/test_build_vetiver_model.py

@@ -13,8 +13,7 @@
 model = get_mock_model().fit(X_df, y)
 
 
-def test_vetiver_model_array_ptype():
-    # build VetiverModel, no ptype
+def test_vetiver_model_array_prototype():
     vt1 = vt.VetiverModel(
         model=model,
         prototype_data=X_array,
@@ -29,8 +28,7 @@ def test_vetiver_model_array_ptype():
     assert list(vt1.prototype.__fields__.values())[0].type_ == int
 
 
-def test_vetiver_model_df_ptype():
-    # build VetiverModel, df ptype_data
+def test_vetiver_model_df_prototype():
     vt2 = vt.VetiverModel(
         model=model,
         prototype_data=X_df,
@@ -45,7 +43,7 @@ def test_vetiver_model_df_ptype():
     assert list(vt2.prototype.__fields__.values())[0].type_ == int
 
 
-def test_vetiver_model_dict_ptype():
+def test_vetiver_model_dict_prototype():
     dict_data = {"B": 0, "C": 0, "D": 0}
     vt3 = vt.VetiverModel(
         model=model,
@@ -61,8 +59,7 @@ def test_vetiver_model_dict_ptype():
     assert list(vt3.prototype.__fields__.values())[0].type_ == int
 
 
-def test_vetiver_model_no_ptype():
-    # build VetiverModel, no ptype
+def test_vetiver_model_no_prototype():
     vt4 = vt.VetiverModel(
         model=model,
         prototype_data=None,
@@ -76,6 +73,21 @@ def test_vetiver_model_no_ptype():
     assert vt4.prototype is None
 
 
+def test_vetiver_model_use_ptype():
+    vt5 = vt.VetiverModel(
+        model=model,
+        ptype_data=X_df,
+        model_name="model",
+        versioned=None,
+        description=None,
+        metadata=None,
+    )
+
+    assert vt5.model == model
+    assert isinstance(vt5.prototype.construct(), pydantic.BaseModel)
+    assert list(vt5.prototype.__fields__.values())[0].type_ == int
+
+
 def test_vetiver_model_from_pin():
 
     v = vt.VetiverModel(

vetiver/vetiver_model.py

@@ -66,25 +66,24 @@ def __init__(
         model,
         model_name: str,
         prototype_data=None,
-        ptype_data=None,
         versioned=None,
         description: str = None,
         metadata: dict = None,
         **kwargs
     ):
-        if ptype_data is not None:
-            prototype_data = ptype_data
+        if "ptype_data" in kwargs:
+            prototype_data = kwargs.pop("ptype_data")
             warn(
                 "argument for saving input data prototype has changed to "
-                "save_prototype, from save_ptype",
+                "prototype_data, from ptype_data",
                 DeprecationWarning,
                 stacklevel=2,
             )
 
         translator = create_handler(model, prototype_data)
 
         self.model = translator.model
-        self.prototype = translator.construct_ptype()
+        self.prototype = translator.construct_prototype()
         self.model_name = model_name
         self.description = description if description else translator.describe()
         self.versioned = versioned