Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

docs(triton): update triton adapter docs to make it run #5

Merged
merged 1 commit into from
Oct 19, 2021
Merged

docs(triton): update triton adapter docs to make it run #5

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 14 additions & 7 deletions model-mesh-triton-adapter/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,26 +4,33 @@ This is an adapter which implements the internal model-mesh model management API

## How to

1. Pull Triton Serving Docker Image
1. Clone the repository

$ git clone https://github.com/kserve/modelmesh-runtime-adapter.git
$ cd modelmesh-runtime-adapter/model-mesh-triton-adapter

2. Pull Triton Serving Docker Image

$ docker pull nvcr.io/nvidia/tritonserver:20.09-py3

2. Run Triton Serving Container with model data mounted
3. Run Triton Serving Container with model data mounted

By default, Triton Serving Docker expose Port `8000` for HTTP and Port `8001` for gRPC.

Using following command to forward container's `8000` to your workstation's `8000` and container's `8001` to your workstation's `8001`.

$ docker run -p 8000:8000 -p 8001:8001 -v /Users/tnarayan/AI/model-mesh-triton-adapter/examples/models/:/models nvcr.io/nvidia/tritonserver:20.09-py3 tritonserver --model-store=/models --model-control-mode=explicit --strict-model-config=false --strict-readiness=false
$ docker run -p 8000:8000 -p 8001:8001 -v $(pwd)/examples/models:/models nvcr.io/nvidia/tritonserver:20.09-py3 tritonserver --model-store=/models --model-control-mode=explicit --strict-model-config=false --strict-readiness=false

3. Setup your Golang, gRPC and Protobuff Development Environment locally
4. Setup your Golang, gRPC and Protobuff Development Environment locally

Follow this [gRPC Go Quick Start Guide](https://grpc.io/docs/quickstart/go/)

4. Run Triton adapter with:
5. Run Triton adapter with:

$ export ROOT_MODEL_DIR=$(pwd)/examples/models
$ export CONTAINER_MEM_REQ_BYTES=268435456 # 256MB
$ go run main.go

5. Test adapter with this client from another terminal:
6. Test adapter with this client from another terminal:

$ go run triton/client/client.go
$ go run triton/adapter_client/adapter_client.go