Update Keras model example and docs (#109)

Signed-off-by: ted chang <htchang@us.ibm.com>
kserve · Jan 19, 2022 · 5ca8841 · 5ca8841
1 parent a77e2a9
commit 5ca8841
Show file tree

Hide file tree

Showing 5 changed files with 38 additions and 10 deletions.
diff --git a/config/example-predictors/example-keras-mnist-predictor.yaml b/config/example-predictors/example-keras-mnist-predictor.yaml
@@ -0,0 +1,24 @@
+# Copyright 2022 IBM Corporation
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+apiVersion: serving.kserve.io/v1alpha1
+kind: Predictor
+metadata:
+  name: example-keras-mnist
+spec:
+  modelType:
+    name: keras
+  path: keras/mnist.h5
+  storage:
+    s3:
+      secretKey: localMinIO
diff --git a/config/runtimes/triton-2.x.yaml b/config/runtimes/triton-2.x.yaml
@@ -21,6 +21,15 @@ metadata:
     maxLoadingConcurrency: "2"
 spec:
   supportedModelFormats:
+    - name: keras
+      version: "2" # 2.6.0
+      autoSelect: true
+    - name: onnx
+      version: "1" # 1.5.3
+      autoSelect: true
+    - name: pytorch
+      version: "1" # 1.8.0a0+17f8c32
+      autoSelect: true
     - name: tensorflow
       version: "1" # 1.15.4
       autoSelect: true
@@ -30,12 +39,6 @@ spec:
     - name: tensorrt
       version: "7" # 7.2.1
       autoSelect: true
-    - name: pytorch
-      version: "1" # 1.8.0a0+17f8c32
-      autoSelect: true
-    - name: onnx
-      version: "1" # 1.5.3
-      autoSelect: true
 
   multiModel: true
 

diff --git a/controllers/testdata/servingruntime_controller.golden b/controllers/testdata/servingruntime_controller.golden
@@ -850,7 +850,7 @@ spec:
         - name: MM_DEFAULT_VMODEL_OWNER
           value: ksp
         - name: MM_LABELS
-          value: mt:onnx,mt:onnx:1,mt:pytorch,mt:pytorch:1,mt:tensorflow,mt:tensorflow:1,mt:tensorflow:2,mt:tensorrt,mt:tensorrt:7,rt:triton-2.x
+          value: mt:keras,mt:keras:2,mt:onnx,mt:onnx:1,mt:pytorch,mt:pytorch:1,mt:tensorflow,mt:tensorflow:1,mt:tensorflow:2,mt:tensorrt,mt:tensorrt:7,rt:triton-2.x
         - name: MM_TYPE_CONSTRAINTS_PATH
           value: /etc/watson/mmesh/config/type_constraints
         - name: MM_DATAPLANE_CONFIG_PATH

diff --git a/docs/model-types/README.md b/docs/model-types/README.md
@@ -2,6 +2,7 @@
 
 By leveraging existing third-party model servers, we support a number of standard ML model formats out-of-the box, with more to follow. Currently supported model types:
 
+- [Keras](keras.md)
 - [LightGBM](lightgbm.md)
 - [ONNX](onnx.md)
 - [PyTorch ScriptModule](pytorch.md)

diff --git a/docs/model-types/keras.md b/docs/model-types/keras.md
@@ -21,8 +21,8 @@ The following example is using the HDF5 file format, which typically uses a `.h5
 **Storage Layout**
 
 ```
-s3://modelmesh-serving-examples/tensorflow-model/keras-mnist
-└── mnist.h5
+s3://modelmesh-serving-examples/
+└── keras-models/mnist.h5
 ```
 
 **Predictor**
@@ -35,7 +35,7 @@ metadata:
 spec:
   modelType:
     name: keras
-  path: tensorflow-model/keras-mnist/mnist.h5
+  path: keras-models/mnist.h5
   storage:
     s3:
       secretKey: modelStorage