PyTorch Serve: Custom handler not saving inference results about serve

[hungtrieu07]

Can you try curl http://127.0.0.1:8080/ping ? It will return Status equals Healthy if your service run smoothly, else return Unhealthy

[hungtrieu07]

This is my config.properties file, you can refer to it:

inference_address=http://0.0.0.0:8080
management_address=http://0.0.0.0:8081
metrics_address=http://0.0.0.0:8082
load_models=all
install_py_dep_per_model=true
model_store=model-store
models={\
  "FaceDetection": {\
    "1.0": {\
        "defaultVersion": true,\
        "marName": "FaceDetection.mar",\
        "minWorkers": 1,\
        "maxWorkers": 1,\
        "batchSize": 256,\
        "maxBatchDelay": 100,\
        "responseTimeout": 120\
    }\
  },\
  "FaceExpression": {\
    "1.0": {\
        "defaultVersion": true,\
        "marName": "FaceExpression.mar",\
        "minWorkers": 1,\
        "maxWorkers": 4,\
        "batchSize": 256,\
        "maxBatchDelay": 100,\
        "responseTimeout": 120\
    }\
  },\
  "FaceRecognition": {\
    "1.0": {\
        "defaultVersion": true,\
        "marName": "FaceRecognition.mar",\
        "minWorkers": 1,\
        "maxWorkers": 4,\
        "batchSize": 256,\
        "maxBatchDelay": 100,\
        "responseTimeout": 120\
    }\
  },\
    "HumanPose": {\
    "1.0": {\
        "defaultVersion": true,\
        "marName": "HumanPose.mar",\
        "minWorkers": 1,\
        "maxWorkers": 4,\
        "batchSize": 256,\
        "maxBatchDelay": 100,\
        "responseTimeout": 120\
    }\
  },\
  "ActionRecognition": {\
    "1.0": {\
        "defaultVersion": true,\
        "marName": "ActionRecognition.mar",\
        "minWorkers": 1,\
        "maxWorkers": 4,\
        "batchSize": 256,\
        "maxBatchDelay": 100,\
        "responseTimeout": 120\
    }\
  }\
}

[namannandan]

@danial880 Once you've started torchserve and loaded the model could you please try the following API to check that the model has been successfully loaded and the workers have started successfully:
curl http://127.0.0.1:8081/models/facex

Also, when you run curl -X POST http://127.0.0.1:8080/predictions/facex -T 0294.png what happens?
Does curl hang or show any error?

[danial880]

Status

Model Info

Upon posting image: curl command get executed with no error