model-backend

README

Model Backend

The service serve for uploading AI model into Instill platform and retrieving AI model info.

Table of Contents:

• Prerequisite
• Quick start
• Community support
• Documentation
  • API reference
  • Build docker
• License

Prerequisite

• sample models example CPU models running in Triton server To download the sample model, you could run quick-download.sh

$ ./scripts/quick-download.sh

Quick start

$ make all
$ go run ./examples-go/grpc_client.go upload --file sample-models/yolov4-onnx-cpu.zip --name yolov4 --task TASK_DETECTION  # upload a YOLOv4 model for object detection; note --task is optional and could be specified as TASK_DETECTION, TASK_CLASSIFICATION, without specifying task will default TASK_UNSPECIFIED
$ go run ./examples-go/grpc_client.go load -n yolov4 --version 1  # deploy the ensemble model
$ go run ./examples-go/grpc_client.go predict -n yolov4 --version 1 -f sample-models/dog.jpg # make inference

Create a your own model to run in Triton server

Community support

For general help using VDP, you can use one of these channels:

• GitHub (bug reports, feature requests, project discussions and contributions)
• Discord (live discussion with the community and the Instill AI Team)

Documentation

API reference

Build docker

You can build a development Docker image using:

$ docker build -t {tag} .

License

See the LICENSE file for licensing information.