Merge pull request #109 from rstudio/docs-update

add examples to docstrings

docs/source/index.rst

@@ -70,11 +70,12 @@ Monitor
 
 Model Handlers
 ==================
+
 .. autosummary::
    :toctree: reference/
-   :caption: Monitor
+   :caption: Model Handlers
 
-   ~BaseHandler
+   ~handlers.BaseHandler
    ~SKLearnHandler
    ~TorchHandler
    ~StatsmodelsHandler

vetiver/pin_read_write.py

@@ -30,6 +30,16 @@ def vetiver_pin_write(board, model: VetiverModel, versioned: bool = True):
         VetiverModel to be written to board
     versioned: bool
         Whether or not the pin should be versioned
+
+    Example
+    -------
+    >>> import vetiver
+    >>> from pins import board_temp
+    >>> model_board = board_temp(versioned = True, allow_pickle_read = True)
+    >>> X, y = vetiver.get_mock_data()
+    >>> model = vetiver.get_mock_model().fit(X, y)
+    >>> v = vetiver.VetiverModel(model = model, model_name = "my_model", ptype_data = X)
+    >>> vetiver.vetiver_pin_write(model_board, v)
     """
     if not board.allow_pickle_read:
         raise NotImplementedError  # must be pickle-able

vetiver/rsconnect.py

@@ -26,7 +26,7 @@ def deploy_rsconnect(
 
     Parameters
     ----------
-        connect_server:
+        connect_server: rsconnect.api.RSConnectServer
             RSConnect Server
         board:
             Pins board
@@ -52,6 +52,26 @@ def deploy_rsconnect(
             Callback to use to write the log to
         image: str
             Docker image to be specified for off-host execution
+
+        Example
+        -------
+        >>> import vetiver
+        >>> import pins
+        >>> import rsconnect
+        >>> board = pins.board_temp(allow_pickle_read=True)
+        >>> connect_server = rsconnect.api.RSConnectServer(
+        ...    url = url,
+        ...    api_key = api_key)      # doctest: +SKIP
+        >>> X, y = vetiver.get_mock_data()
+        >>> model = vetiver.get_mock_model().fit(X, y)
+        >>> v = vetiver.VetiverModel(model = model,
+        ...    model_name = "my_model",
+        ...    ptype_data = X)
+        >>> vetiver.deploy_rsconnect(
+        ...    connect_server = connect_server,
+        ...    board = board,
+        ...    pin_name = "my_model"
+        ... )      # doctest: +SKIP
     """
     if not title:
         title = pin_name + "_vetiver"

vetiver/server.py

@@ -15,16 +15,22 @@
 class VetiverAPI:
     """Create model aware API
 
-    Attributes
+    Parameters
     ----------
     model :  VetiverModel
         Model to be deployed in API
     check_ptype : bool
         Determine if data prototype should be enforced
     app_factory :
         Type of API to be deployed
-    app :
-        API that is deployed
+
+    Example
+    -------
+    >>> import vetiver
+    >>> X, y = vetiver.get_mock_data()
+    >>> model = vetiver.get_mock_model().fit(X, y)
+    >>> v = vetiver.VetiverModel(model = model, model_name = "my_model", ptype_data = X)
+    >>> v_api = vetiver.VetiverAPI(model = v, check_ptype = True)
     """
 
     app = None
@@ -133,7 +139,7 @@ async def rapidoc():
     def vetiver_post(
         self, endpoint_fx: Callable, endpoint_name: str = "custom_endpoint"
     ):
-        """Create new POST endpoint
+        """Create new POST endpoint that is aware of model input data
 
         Parameters
         ----------
@@ -142,25 +148,31 @@ def vetiver_post(
         endpoint_name : str
             Name of endpoint
 
-        Returns
+        Example
         -------
-        dict
-            Key: endpoint_name Value: Output of endpoint_fx, in list format
+        >>> import vetiver
+        >>> X, y = vetiver.get_mock_data()
+        >>> model = vetiver.get_mock_model().fit(X, y)
+        >>> v = vetiver.VetiverModel(model = model, model_name = "model", ptype_data = X)
+        >>> v_api = vetiver.VetiverAPI(model = v, check_ptype = True)
+        >>> def sum_values(x):
+        ...     return x.sum()
+        >>> v_api.vetiver_post(sum_values, "sums")
         """
         if self.check_ptype is True:
 
             @self.app.post("/" + endpoint_name)
             async def custom_endpoint(input_data: self.model.ptype):
                 y = _prepare_data(input_data)
-                new = endpoint_fx(pd.Series(y))
+                new = endpoint_fx(pd.DataFrame(y))
                 return {endpoint_name: new.tolist()}
 
         else:
 
             @self.app.post("/" + endpoint_name)
             async def custom_endpoint(input_data: Request):
                 y = await input_data.json()
-                new = endpoint_fx(pd.Series(y))
+                new = endpoint_fx(pd.DataFrame(y))
 
                 return {endpoint_name: new.tolist()}
 
@@ -174,6 +186,15 @@ def run(self, port: int = 8000, host: str = "127.0.0.1", **kw):
             An integer that indicates the server port that should be listened on.
         host : str
             A valid IPv4 or IPv6 address, which the application will listen on.
+
+        Example
+        -------
+        >>> import vetiver
+        >>> X, y = vetiver.get_mock_data()
+        >>> model = vetiver.get_mock_model().fit(X, y)
+        >>> v = vetiver.VetiverModel(model = model, model_name = "model", ptype_data = X)
+        >>> v_api = vetiver.VetiverAPI(model = v, check_ptype = True)
+        >>> v_api.run()     # doctest: +SKIP
         """
         _jupyter_nb()
         uvicorn.run(self.app, port=port, host=host, **kw)
@@ -209,6 +230,13 @@ def predict(endpoint, data: Union[dict, pd.DataFrame, pd.Series], **kw):
     -------
     dict
         Endpoint_name and list of endpoint_fx output
+
+    Example
+    -------
+    >>> import vetiver
+    >>> X, y = vetiver.get_mock_data()
+    >>> endpoint = vetiver.vetiver_endpoint(url='http://127.0.0.1:8000/predict')
+    >>> vetiver.predict(endpoint, X)     # doctest: +SKIP
     """
     if isinstance(endpoint, testclient.TestClient):
         requester = endpoint
@@ -273,5 +301,10 @@ def vetiver_endpoint(url="http://127.0.0.1:8000/predict"):
     -------
     url : str
         URI path to endpoint
+
+    Example
+    -------
+    >>> import vetiver
+    >>> endpoint = vetiver.vetiver_endpoint(url='http://127.0.0.1:8000/predict')
     """
     return url

vetiver/tests/test_add_endpoint.py

@@ -30,7 +30,7 @@ def test_endpoint_adds_ptype():
     data = {"B": 0, "C": 0, "D": 0}
     response = client.post("/sum", json=data)
     assert response.status_code == 200, response.text
-    assert response.json() == {"sum": 0}, response.json()
+    assert response.json() == {"sum": [0]}, response.json()
 
 
 def test_endpoint_adds_no_ptype():
@@ -40,4 +40,4 @@ def test_endpoint_adds_no_ptype():
     data = [0, 0, 0]
     response = client.post("/sum", json=data)
     assert response.status_code == 200, response.text
-    assert response.json() == {"sum": 0}, response.json()
+    assert response.json() == {"sum": [0]}, response.json()

vetiver/vetiver_model.py

@@ -25,7 +25,7 @@ class VetiverModel:
     ----------
     model :
         A trained model, such as an sklearn or torch model
-    name : string
+    model_name : string
         Model name or ID
     ptype_data : pd.DataFrame, np.array
         Sample of data model should expect when it is being served
@@ -49,6 +49,14 @@ class VetiverModel:
     VetiverModel can also take an initialized custom VetiverHandler
     as a model, for advanced use cases or non-supported model types.
 
+    Example
+    -------
+    >>> from vetiver import mock, VetiverModel
+    >>> X, y = mock.get_mock_data()
+    >>> model = mock.get_mock_model().fit(X, y)
+    >>> v = VetiverModel(model = model, model_name = "my_model", ptype_data = X)
+    >>> v.description
+    "Scikit-learn <class 'sklearn.dummy.DummyRegressor'> model"
     """
 
     def __init__(

vetiver/write_docker.py

@@ -29,8 +29,8 @@ def write_docker(
 ):
     """Writes a Dockerfile to run VetiverAPI in a container
 
-    Args
-    ----
+    Parameters
+    ----------
         app_file: str
             File containing VetiverAPI to be deployed into container
         path: str
@@ -41,6 +41,22 @@ def write_docker(
             Host address to run VetiverAPI from Dockerfile
         port: str
             Port to run VetiverAPI from Dockerfile
+
+    Example
+    -------
+    >>> import vetiver
+    >>> import tempfile
+    >>> import pins
+    >>> tmp = tempfile.TemporaryDirectory()
+    >>> board = pins.board_temp(allow_pickle_read=True)
+    >>> X, y = vetiver.get_mock_data()
+    >>> model = vetiver.get_mock_model().fit(X, y)
+    >>> v = vetiver.VetiverModel(model = model, model_name = "my_model", ptype_data = X)
+    >>> vetiver.vetiver_pin_write(board, v)
+    >>> vetiver.write_app(board,
+    ...     "my_model",
+    ...     file = tmp.name + "/app.py") # need file for model
+    >>> vetiver.write_docker(app_file = "app.py", path = tmp.name)
     """
     py_version = str(sys.version_info.major) + "." + str(sys.version_info.minor)
 

vetiver/write_fastapi.py

@@ -56,16 +56,31 @@ def write_app(
 ):
     """Write VetiverAPI app to a file
 
-    Args
-    ----
-    board :  pinse.BaseBoard
+    Parameters
+    ----------
+    board :
         API to be written
     pin_name : string
         Name of pin containing VetiverModel
     version :
         Pins version of VetiverModel
     file :
         Name of file
+
+    Example
+    -------
+    >>> import vetiver
+    >>> import tempfile
+    >>> import pins
+    >>> tmp = tempfile.TemporaryDirectory()
+    >>> board = pins.board_temp(allow_pickle_read=True)
+    >>> X, y = vetiver.get_mock_data()
+    >>> model = vetiver.get_mock_model().fit(X, y)
+    >>> v = vetiver.VetiverModel(model = model, model_name = "my_model", ptype_data = X)
+    >>> vetiver.vetiver_pin_write(board, v)
+    >>> vetiver.write_app(board,
+    ...     "my_model",
+    ...     file = tmp.name + "/app.py")
     """
 
     if board.versioned: