WIP: Allow predict for sklearn standalone server as wll as predict_proba

doc/source/servers/sklearn.md

@@ -27,5 +27,29 @@ spec:
 
 ```
 
+## Sklearn Method
+
+By default the server will call `predict_proba` on your loaded model/pipeline. If you wish for it to call `predict` instead you can pass a parameter `method` and set it to `predict`. For example:
+
+```
+apiVersion: machinelearning.seldon.io/v1alpha2
+kind: SeldonDeployment
+metadata:
+  name: sklearn
+spec:
+  name: iris-predict
+  predictors:
+  - graph:
+      children: []
+      implementation: SKLEARN_SERVER
+      modelUri: gs://seldon-models/sklearn/iris
+      name: classifier
+      parameters:
+        - name: method
+          type: STRING
+          value: predict
+    name: default
+    replicas: 1
+```
 
 Try out a [worked notebook](../examples/server_examples.html)

servers/sklearnserver/samples/iris_predict.yaml

@@ -0,0 +1,18 @@
+apiVersion: machinelearning.seldon.io/v1alpha2
+kind: SeldonDeployment
+metadata:
+  name: sklearn
+spec:
+  name: iris-predict
+  predictors:
+  - graph:
+      children: []
+      implementation: SKLEARN_SERVER
+      modelUri: gs://seldon-models/sklearn/iris
+      name: classifier
+      parameters:
+        - name: method
+          type: STRING
+          value: predict
+    name: default
+    replicas: 1

servers/sklearnserver/sklearnserver/SKLearnServer.py

@@ -4,25 +4,41 @@
 from seldon_core.user_model import SeldonComponent
 from typing import Dict, List, Union, Iterable
 import os
+import logging
+
+logger = logging.getLogger(__name__)
+
 
 JOBLIB_FILE = "model.joblib"
 
+
 class SKLearnServer(SeldonComponent):
-    def __init__(self, model_uri: str):
+    def __init__(self, model_uri: str = None,  method: str = "predict_proba"):
         super().__init__()
         self.model_uri = model_uri
+        self.method = method
         self.ready = False
+        print("Model uri:",self.model_uri)
+        print("method:",self.method)
+        self.load()
 
     def load(self):
         print("load")
         model_file = os.path.join(seldon_core.Storage.download(self.model_uri), JOBLIB_FILE)
-        print("model file",model_file)
+        print("model file", model_file)
         self._joblib = joblib.load(model_file)
         self.ready = True
 
     def predict(self, X: np.ndarray, names: Iterable[str], meta: Dict = None) -> Union[np.ndarray, List, str, bytes]:
-        print("predict")
-        if not self.ready:
-            self.load()
-        result = self._joblib.predict(X)
-        return result
+        try:
+            if not self.ready:
+                self.load()
+            if self.method == "predict_proba":
+                logger.info("Calling predict_proba")
+                result = self._joblib.predict_proba(X)
+            else:
+                logger.info("Calling predict")
+                result = self._joblib.predict(X)
+            return result
+        except Exception as ex:
+            logging.exception("Exception during predict")