Files
Dan SunandGitHub 45d774139c Add Unified Inference Type and refactor REST/gRPC server code (#2629)
* Refactor encoding/decoding

Signed-off-by: Dan Sun <[email protected]>

* grpc multi-processing

Signed-off-by: Dan Sun <[email protected]>

* Add normalized infer types for handlers

Signed-off-by: Dan Sun <[email protected]>

* Fix lint errors

Signed-off-by: Dan Sun <[email protected]>

* Add gRPC content converter

Signed-off-by: Dan Sun <[email protected]>

* Add from_grpc flag on InferRequest

Signed-off-by: Dan Sun <[email protected]>

* Rename to infer_type

Signed-off-by: Dan Sun <[email protected]>

* Add infer output

Signed-off-by: Dan Sun <[email protected]>

* Add timing_asgi

Signed-off-by: Dan Sun <[email protected]>

* Return InferResponse in custom model example

Signed-off-by: Dan Sun <[email protected]>

* Add back ModelReady v2 endpoint

Signed-off-by: Dan Sun <[email protected]>

* Fix gRPC server

Signed-off-by: Dan Sun <[email protected]>

* pin fastapi to 0.88 as 0.89 breaks raw response

Signed-off-by: Dan Sun <[email protected]>

* Update model handlers doc

Signed-off-by: Dan Sun <[email protected]>

* Convert gRPC response to dict

Signed-off-by: Dan Sun <[email protected]>

* Remove outdated example code files

Signed-off-by: Dan Sun <[email protected]>

* Add from_grpc methods

Signed-off-by: Dan Sun <[email protected]>

* Address comments

Signed-off-by: Dan Sun <[email protected]>

* Refactor np dtype converter func

Signed-off-by: Dan Sun <[email protected]>

Signed-off-by: Dan Sun <[email protected]>
2023-01-20 12:31:57 -05:00
..
2020-11-17 14:48:05 -08:00
2020-11-17 14:48:05 -08:00
2020-11-17 14:48:05 -08:00

Predict on a Triton InferenceService with TorchScript model

The guidelines are described here.