Merge pull request #1466 from shrinath-suresh/kservev2-regression-tes…

…t-fix

Kservev2 regression test fix

kubernetes/kserve/image_transformer/README.md

@@ -134,7 +134,7 @@ The request first comes to the image transformer at port 8080 and in turn reques
 - The curl request for inference is as below:
 
 ```
-curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/mnist.json http://0.0.0.0:8080/v1/models/mnist:predict
+curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/v1/mnist.json http://0.0.0.0:8080/v1/models/mnist:predict
 ```
 
 output:

kubernetes/kserve/kserve_wrapper/README.md

@@ -122,7 +122,7 @@ For v1 protocol
 The curl request for inference is as below:
 
 ```bash
-curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/mnist.json http://0.0.0.0:8080/v1/models/mnist:predict
+curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/v1/mnist.json http://0.0.0.0:8080/v1/models/mnist:predict
 ```
 
 Output:
@@ -134,7 +134,7 @@ Output:
 The curl request for explain is as below:
 
 ```
-curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/mnist.json http://0.0.0.0:8080/v1/models/mnist:explain
+curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/v1/mnist.json http://0.0.0.0:8080/v1/models/mnist:explain
 ```
 
 Output:
@@ -176,7 +176,7 @@ Response:
 The curl request for explain is as below:
 
 ```
-curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/mnist.json http://0.0.0.0:8080/v2/models/mnist/explain
+curl -H "Content-Type: application/json" --data @serve/kubernetes/kserve/kf_request_json/v1/mnist.json http://0.0.0.0:8080/v2/models/mnist/explain
 ```
 
 Response:
@@ -238,25 +238,25 @@ For v1 protocol
 The curl request for inference is as below:
 
 ```bash
-curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/bert.json http://0.0.0.0:8080/v1/models/bert:predict
+curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/v1/bert.json http://0.0.0.0:8080/v1/models/bert:predict
 ```
 
 The curl request for Explain is as below:
 
 ```bash
-curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/bert.json http://0.0.0.0:8080/v1/models/bert:explain
+curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/v1/bert.json http://0.0.0.0:8080/v1/models/bert:explain
 ```
 
 For v2 protocol
 
 The curl request for inference is as below:
 
 ```bash
-curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/bert.json http://0.0.0.0:8080/v2/models/bert/infer
+curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/v1/bert.json http://0.0.0.0:8080/v2/models/bert/infer
 ```
 
 The curl request for Explain is as below:
 
 ```bash
-curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/bert.json http://0.0.0.0:8080/v2/models/bert/explain
+curl -H "Content-Type: application/json" --data @kubernetes/kserve/kf_request_json/v1/bert.json http://0.0.0.0:8080/v2/models/bert/explain
 ```

test/postman/kf_https_test_collection.json

@@ -303,7 +303,7 @@
 				"body": {
 					"mode": "file",
 					"file": {
-						"src": "../kubernetes/kserve/kf_request_json/mnist.json"
+						"src": "../kubernetes/kserve/kf_request_json/v1/mnist.json"
 					},
 					"options": {
 						"raw": {
@@ -355,7 +355,7 @@
 				"body": {
 					"mode": "file",
 					"file": {
-						"src": "../kubernetes/kserve/kf_request_json/mnist.json"
+						"src": "../kubernetes/kserve/kf_request_json/v1/mnist.json"
 					},
 					"options": {
 						"raw": {

test/postman/kf_inference_data.json

@@ -4,7 +4,7 @@
     "model_name":"mnist",
     "worker":1,
     "synchronous":"true",
-    "file":"../kubernetes/kserve/kf_request_json/mnist.json",
+    "file":"../kubernetes/kserve/kf_request_json/v1/mnist.json",
     "content-type":"application/json",
     "expected":{
         "predictions": [

test/postman/kfv2_https_test_collection.json

@@ -295,7 +295,7 @@
 				"body": {
 					"mode": "file",
 					"file": {
-						"src": "../kubernetes/kserve/kf_request_json/mnist_v2.json"
+						"src": "../kubernetes/kserve/kf_request_json/v2/mnist/mnist_v2_tensor.json"
 					},
 					"options": {
 						"raw": {
@@ -347,7 +347,7 @@
 				"body": {
 					"mode": "file",
 					"file": {
-						"src": "../kubernetes/kserve/kf_request_json/mnist_v2.json"
+						"src": "../kubernetes/kserve/kf_request_json/v2/mnist/mnist_v2_tensor.json"
 					},
 					"options": {
 						"raw": {

test/postman/kfv2_inference_data.json

@@ -3,7 +3,7 @@
    "model_name": "mnist",
    "worker": 1,
    "synchronous": "true",
-   "file": "../kubernetes/kserve/kf_request_json/mnist_v2.json",
+   "file": "../kubernetes/kserve/kf_request_json/v2/mnist/mnist_v2_tensor.json",
    "content-type": "application/json",
    "expected": {
       "id":"d3b15cad-50a2-4eaf-80ce-8b0a428bd298","model_name":"mnist","model_version":"1.0","outputs":[{"name":"predict","shape":[],"datatype":"INT64","data":[1]}]

test/pytest/test_handler.py

@@ -9,7 +9,7 @@
 snapshot_file_kf = os.path.join(REPO_ROOT,"test/config_kf.properties")
 snapshot_file_tf = os.path.join(REPO_ROOT,"test/config_ts.properties")
 data_file_mnist = os.path.join(REPO_ROOT, 'examples/image_classifier/mnist/test_data/1.png')
-input_json_mnist = os.path.join(REPO_ROOT, "kubernetes/kserve/kf_request_json/mnist.json")
+input_json_mnist = os.path.join(REPO_ROOT, "kubernetes/kserve/kf_request_json/v1/mnist.json")
 input_json_mmf = os.path.join(REPO_ROOT, "examples/MMF-activity-recognition/372CC.info.json")
 
 def getAPIS(snapshot_file):
@@ -202,7 +202,7 @@ def test_kserve_mnist_model_register_scale_inference_with_non_existent_handler(
         data = json.loads(s)
 
     response = run_inference_using_url_with_data_json(KF_INFERENCE_API + '/v1/models/mnist:predict', data)
-    
+
     if response is None:
         assert True, "Inference failed as the handler is non existent"
     else:
@@ -240,7 +240,7 @@ def test_huggingface_bert_batch_inference():
     test_utils.start_torchserve(no_config_snapshots=True)
     test_utils.register_model_with_params(params)
     input_text = os.path.join(REPO_ROOT, 'examples/Huggingface_Transformers/Seq_classification_artifacts/sample_text.txt')
-    
+
     # Make 2 curl requests in parallel with &
     # curl --header \"X-Forwarded-For: 1.2.3.4\" won't work since you can't access local host anymore
     response = os.popen(f"curl http://127.0.0.1:8080/predictions/BERTSeqClassification -T {input_text} & curl http://127.0.0.1:8080/predictions/BERTSeqClassification -T {input_text}")
@@ -253,7 +253,7 @@ def test_huggingface_bert_batch_inference():
 
 @pytest.mark.skip(reason="MMF doesn't support PT 1.10 yet")
 def test_MMF_activity_recognition_model_register_and_inference_on_valid_model():
-  
+
     test_utils.start_torchserve(snapshot_file = snapshot_file_tf)
     test_utils.register_model('MMF_activity_recognition_v2', 'https://torchserve.pytorch.org/mar_files/MMF_activity_recognition_v2.mar')
     os.system('wget https://mmfartifacts.s3-us-west-2.amazonaws.com/372CC.mp4 -P ../../examples/MMF-activity-recognition')