move api_data_to_frame to server

Author: isabelizimm

vetiver/handlers/sklearn.py

@@ -2,7 +2,6 @@
 import sklearn
 
 from .base import BaseHandler
-from ..helpers import api_data_to_frame
 
 
 class SKLearnHandler(BaseHandler):
@@ -34,12 +33,14 @@ def handler_predict(self, input_data, check_prototype):
         prediction
             Prediction from model
         """
-        if check_prototype:
-            input_data = api_data_to_frame(input_data)
+        # if check_prototype:
+        #     input_data = api_data_to_frame(input_data)
 
         if not check_prototype or isinstance(input_data, pd.DataFrame):
             prediction = self.model.predict(input_data)
         else:
             prediction = self.model.predict([input_data])
 
+        # some sort of post-prediction/pre send back to user hook
+
         return prediction.tolist()

vetiver/handlers/statsmodels.py

@@ -1,7 +1,6 @@
 import pandas as pd
 
 from .base import BaseHandler
-from ..helpers import api_data_to_frame
 
 sm_exists = True
 try:
@@ -43,8 +42,8 @@ def handler_predict(self, input_data, check_prototype):
         if not sm_exists:
             raise ImportError("Cannot import `statsmodels`")
 
-        if check_prototype:
-            input_data = api_data_to_frame(input_data)
+        # if check_prototype:
+        #     input_data = api_data_to_frame(input_data)
 
         if isinstance(input_data, (list, pd.DataFrame)):
             prediction = self.model.predict(input_data)

vetiver/handlers/torch.py

@@ -1,7 +1,6 @@
 import numpy as np
 
 from .base import BaseHandler
-from ..helpers import api_data_to_frame
 
 torch_exists = True
 try:
@@ -44,7 +43,7 @@ def handler_predict(self, input_data, check_prototype):
             raise ImportError("Cannot import `torch`.")
 
         if check_prototype:
-            input_data = api_data_to_frame(input_data)
+            # input_data = api_data_to_frame(input_data)
             input_data = np.array(input_data, dtype=np.array(self.prototype_data).dtype)
             prediction = self.model(torch.from_numpy(input_data))
 

vetiver/handlers/xgboost.py

@@ -1,7 +1,6 @@
 import pandas as pd
 
 from .base import BaseHandler
-from ..helpers import api_data_to_frame
 
 xgb_exists = True
 try:
@@ -44,8 +43,8 @@ def handler_predict(self, input_data, check_prototype):
         if not xgb_exists:
             raise ImportError("Cannot import `xgboost`")
 
-        if check_prototype:
-            input_data = api_data_to_frame(input_data)
+        # if check_prototype:
+        #     input_data = api_data_to_frame(input_data)
 
         if not isinstance(input_data, pd.DataFrame):
             try:

vetiver/server.py

@@ -16,6 +16,7 @@
 from .utils import _jupyter_nb
 from .vetiver_model import VetiverModel
 from .meta import VetiverMeta
+from .helpers import api_data_to_frame
 
 
 class VetiverAPI:
@@ -175,7 +176,8 @@ def vetiver_post(self, endpoint_fx: Callable, endpoint_name: str = None, **kw):
 
             @self.app.post(urljoin("/", endpoint_name), name=endpoint_name)
             async def custom_endpoint(input_data: List[self.model.prototype]):
-                predictions = endpoint_fx(input_data, **kw)
+                _to_frame = api_data_to_frame(input_data)
+                predictions = endpoint_fx(_to_frame, **kw)
                 return {endpoint_name: predictions}
 
         else:

vetiver/tests/test_add_endpoint.py

@@ -26,7 +26,11 @@ def vetiver_model():
 
 
 def sum_values(x):
-    return api_data_to_frame(x).sum()
+    return x.sum().to_list()
+
+
+def sum_values_no_prototype(x):
+    return api_data_to_frame(x).sum().to_list()
 
 
 @pytest.fixture
@@ -45,7 +49,7 @@ def vetiver_client(vetiver_model):  # With check_prototype=True
 def vetiver_client_check_ptype_false(vetiver_model):  # With check_prototype=False
 
     app = VetiverAPI(vetiver_model, check_prototype=False)
-    app.vetiver_post(sum_values, "sum")
+    app.vetiver_post(sum_values_no_prototype, "sum")
 
     app.app.root_path = "/sum"
     client = TestClient(app.app)