ChatQnA - refactoring README.md for deploy application on ROCm (#1857)
Signed-off-by: Chingis Yundunov <YundunovCN@sibedge.com> Signed-off-by: Chingis Yundunov <c.yundunov@datamonsters.com> Co-authored-by: Chingis Yundunov <YundunovCN@sibedge.com> Co-authored-by: Artem Astafev <a.astafev@datamonsters.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
This commit is contained in:
chyundunovDatamonsters
committed by
GitHub
GitHub
parent
f90a6d2a8e
commit
bb7a675665
@@ -1,163 +1,90 @@
|
||||
# Build and Deploy ChatQnA Application on AMD GPU (ROCm)
|
||||
# Deploying ChatQnA on AMD ROCm GPU
|
||||
|
||||
## Build Docker Images
|
||||
This document outlines the single node deployment process for a ChatQnA application utilizing the [GenAIComps](https://github.com/opea-project/GenAIComps.git) microservices on Intel Xeon server and AMD GPU. The steps include pulling Docker images, container deployment via Docker Compose, and service execution using microservices `llm`.
|
||||
|
||||
### 1. Build Docker Image
|
||||
Note: The default LLM is `meta-llama/Meta-Llama-3-8B-Instruct`. Before deploying the application, please make sure either you've requested and been granted the access to it on [Huggingface](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) or you've downloaded the model locally from [ModelScope](https://www.modelscope.cn/models).
|
||||
|
||||
- #### Create application install directory and go to it:
|
||||
## Table of Contents
|
||||
|
||||
```bash
|
||||
mkdir ~/chatqna-install && cd chatqna-install
|
||||
```
|
||||
1. [ChatQnA Quick Start Deployment](#chatqna-quick-start-deployment)
|
||||
2. [ChatQnA Docker Compose Files](#chatqna-docker-compose-files)
|
||||
3. [Validate Microservices](#validate-microservices)
|
||||
4. [Conclusion](#conclusion)
|
||||
|
||||
- #### Clone the repository GenAIExamples (the default repository branch "main" is used here):
|
||||
## ChatQnA Quick Start Deployment
|
||||
|
||||
```bash
|
||||
git clone https://github.com/opea-project/GenAIExamples.git
|
||||
```
|
||||
This section describes how to quickly deploy and test the ChatQnA service manually on an AMD ROCm GPU. The basic steps are:
|
||||
|
||||
If you need to use a specific branch/tag of the GenAIExamples repository, then (v1.3 replace with its own value):
|
||||
1. [Access the Code](#access-the-code)
|
||||
2. [Configure the Deployment Environment](#configure-the-deployment-environment)
|
||||
3. [Deploy the Services Using Docker Compose](#deploy-the-services-using-docker-compose)
|
||||
4. [Check the Deployment Status](#check-the-deployment-status)
|
||||
5. [Validate the Pipeline](#validate-the-pipeline)
|
||||
6. [Cleanup the Deployment](#cleanup-the-deployment)
|
||||
|
||||
```bash
|
||||
git clone https://github.com/opea-project/GenAIExamples.git && cd GenAIExamples && git checkout v1.3
|
||||
```
|
||||
### Access the Code
|
||||
|
||||
We remind you that when using a specific version of the code, you need to use the README from this version:
|
||||
Clone the GenAIExample repository and access the ChatQnA AMD ROCm GPU platform Docker Compose files and supporting scripts:
|
||||
|
||||
- #### Go to build directory:
|
||||
```bash
|
||||
git clone https://github.com/opea-project/GenAIExamples.git
|
||||
cd GenAIExamples/ChatQnA
|
||||
```
|
||||
|
||||
```bash
|
||||
cd ~/chatqna-install/GenAIExamples/ChatQnA/docker_image_build
|
||||
```
|
||||
Then checkout a released version, such as v1.3:
|
||||
|
||||
- Cleaning up the GenAIComps repository if it was previously cloned in this directory.
|
||||
This is necessary if the build was performed earlier and the GenAIComps folder exists and is not empty:
|
||||
```bash
|
||||
git checkout v1.3
|
||||
```
|
||||
|
||||
```bash
|
||||
echo Y | rm -R GenAIComps
|
||||
```
|
||||
### Configure the Deployment Environment
|
||||
|
||||
- #### Clone the repository GenAIComps (the default repository branch "main" is used here):
|
||||
To set up environment variables for deploying ChatQnA services, set up some parameters specific to the deployment environment and source the `set_env_*.sh` script in this directory:
|
||||
|
||||
```bash
|
||||
git clone https://github.com/opea-project/GenAIComps.git
|
||||
```
|
||||
- if used vLLM - set_env_vllm.sh
|
||||
- if used vLLM with FaqGen - set_env_faqgen_vllm.sh
|
||||
- if used TGI - set_env.sh
|
||||
- if used TGI with FaqGen - set_env_faqgen.sh
|
||||
|
||||
If you use a specific tag of the GenAIExamples repository,
|
||||
then you should also use the corresponding tag for GenAIComps. (v1.3 replace with its own value):
|
||||
Set the values of the variables:
|
||||
|
||||
```bash
|
||||
git clone https://github.com/opea-project/GenAIComps.git && cd GenAIComps && git checkout v1.3
|
||||
```
|
||||
- **HOST_IP, HOST_IP_EXTERNAL** - These variables are used to configure the name/address of the service in the operating system environment for the application services to interact with each other and with the outside world.
|
||||
|
||||
We remind you that when using a specific version of the code, you need to use the README from this version.
|
||||
If your server uses only an internal address and is not accessible from the Internet, then the values for these two variables will be the same and the value will be equal to the server's internal name/address.
|
||||
|
||||
- #### Setting the list of images for the build (from the build file.yaml)
|
||||
If your server uses only an external, Internet-accessible address, then the values for these two variables will be the same and the value will be equal to the server's external name/address.
|
||||
|
||||
If you want to deploy a vLLM-based or TGI-based application, then the set of services is installed as follows:
|
||||
If your server is located on an internal network, has an internal address, but is accessible from the Internet via a proxy/firewall/load balancer, then the HOST_IP variable will have a value equal to the internal name/address of the server, and the EXTERNAL_HOST_IP variable will have a value equal to the external name/address of the proxy/firewall/load balancer behind which the server is located.
|
||||
|
||||
#### vLLM-based application
|
||||
We set these values in the file set_env\*\*\*\*.sh
|
||||
|
||||
```bash
|
||||
service_list="dataprep retriever vllm-rocm chatqna chatqna-ui nginx"
|
||||
```
|
||||
- **Variables with names like "**\*\*\*\*\*\*\_PORT"\*\* - These variables set the IP port numbers for establishing network connections to the application services.
|
||||
The values shown in the file set_env.sh or set_env_vllm.sh they are the values used for the development and testing of the application, as well as configured for the environment in which the development is performed. These values must be configured in accordance with the rules of network access to your environment's server, and must not overlap with the IP ports of other applications that are already in use.
|
||||
|
||||
#### vLLM-based application with FaqGen
|
||||
Setting variables in the operating system environment:
|
||||
|
||||
```bash
|
||||
service_list="dataprep retriever vllm-rocm llm-faqgen chatqna chatqna-ui nginx"
|
||||
```
|
||||
```bash
|
||||
export HUGGINGFACEHUB_API_TOKEN="Your_HuggingFace_API_Token"
|
||||
source ./set_env_*.sh # replace the script name with the appropriate one
|
||||
```
|
||||
|
||||
#### TGI-based application
|
||||
Consult the section on [ChatQnA Service configuration](#chatqna-configuration) for information on how service specific configuration parameters affect deployments.
|
||||
|
||||
```bash
|
||||
service_list="dataprep retriever chatqna chatqna-ui nginx"
|
||||
```
|
||||
### Deploy the Services Using Docker Compose
|
||||
|
||||
#### TGI-based application with FaqGen
|
||||
To deploy the ChatQnA services, execute the `docker compose up` command with the appropriate arguments. For a default deployment with TGI, execute the command below. It uses the 'compose.yaml' file.
|
||||
|
||||
```bash
|
||||
service_list="dataprep retriever llm-faqgen chatqna chatqna-ui nginx"
|
||||
```
|
||||
|
||||
- #### Pull Docker Images
|
||||
|
||||
```bash
|
||||
docker pull redis/redis-stack:7.2.0-v9
|
||||
docker pull ghcr.io/huggingface/text-embeddings-inference:cpu-1.5
|
||||
```
|
||||
|
||||
- #### Optional. Pull TGI Docker Image (Do this if you want to use TGI)
|
||||
|
||||
```bash
|
||||
docker pull ghcr.io/huggingface/text-generation-inference:2.3.1-rocm
|
||||
```
|
||||
|
||||
- #### Build Docker Images
|
||||
|
||||
```bash
|
||||
docker compose -f build.yaml build ${service_list} --no-cache
|
||||
```
|
||||
|
||||
After the build, we check the list of images with the command:
|
||||
|
||||
```bash
|
||||
docker image ls
|
||||
```
|
||||
|
||||
The list of images should include:
|
||||
|
||||
##### vLLM-based application:
|
||||
|
||||
- redis/redis-stack:7.2.0-v9
|
||||
- opea/dataprep:latest
|
||||
- ghcr.io/huggingface/text-embeddings-inference:cpu-1.5
|
||||
- opea/retriever:latest
|
||||
- opea/vllm-rocm:latest
|
||||
- opea/chatqna:latest
|
||||
- opea/chatqna-ui:latest
|
||||
- opea/nginx:latest
|
||||
|
||||
##### vLLM-based application with FaqGen:
|
||||
|
||||
- redis/redis-stack:7.2.0-v9
|
||||
- opea/dataprep:latest
|
||||
- ghcr.io/huggingface/text-embeddings-inference:cpu-1.5
|
||||
- opea/retriever:latest
|
||||
- opea/vllm-rocm:latest
|
||||
- opea/llm-faqgen:latest
|
||||
- opea/chatqna:latest
|
||||
- opea/chatqna-ui:latest
|
||||
- opea/nginx:latest
|
||||
|
||||
##### TGI-based application:
|
||||
|
||||
- redis/redis-stack:7.2.0-v9
|
||||
- opea/dataprep:latest
|
||||
- ghcr.io/huggingface/text-embeddings-inference:cpu-1.5
|
||||
- opea/retriever:latest
|
||||
- ghcr.io/huggingface/text-generation-inference:2.3.1-rocm
|
||||
- opea/chatqna:latest
|
||||
- opea/chatqna-ui:latest
|
||||
- opea/nginx:latest
|
||||
|
||||
##### TGI-based application with FaqGen:
|
||||
|
||||
- redis/redis-stack:7.2.0-v9
|
||||
- opea/dataprep:latest
|
||||
- ghcr.io/huggingface/text-embeddings-inference:cpu-1.5
|
||||
- opea/retriever:latest
|
||||
- ghcr.io/huggingface/text-generation-inference:2.3.1-rocm
|
||||
- opea/llm-faqgen:latest
|
||||
- opea/chatqna:latest
|
||||
- opea/chatqna-ui:latest
|
||||
- opea/nginx:latest
|
||||
|
||||
---
|
||||
|
||||
## Deploy the ChatQnA Application
|
||||
|
||||
### Docker Compose Configuration for AMD GPUs
|
||||
```bash
|
||||
cd docker_compose/amd/gpu/rocm
|
||||
# if used TGI
|
||||
docker compose -f compose.yaml up -d
|
||||
# if used TGI with FaqGen
|
||||
# docker compose -f compose_faqgen.yaml up -d
|
||||
# if used vLLM
|
||||
# docker compose -f compose_vllm.yaml up -d
|
||||
# if used vLLM with FaqGen
|
||||
# docker compose -f compose_faqgen_vllm.yaml up -d
|
||||
```
|
||||
|
||||
To enable GPU support for AMD GPUs, the following configuration is added to the Docker Compose file:
|
||||
|
||||
@@ -198,332 +125,103 @@ security_opt:
|
||||
**How to Identify GPU Device IDs:**
|
||||
Use AMD GPU driver utilities to determine the correct `cardN` and `renderN` IDs for your GPU.
|
||||
|
||||
### Set deploy environment variables
|
||||
> **Note**: developers should build docker image from source when:
|
||||
>
|
||||
> - Developing off the git main branch (as the container's ports in the repo may be different > from the published docker image).
|
||||
> - Unable to download the docker image.
|
||||
> - Use a specific version of Docker image.
|
||||
|
||||
#### Setting variables in the operating system environment:
|
||||
Please refer to the table below to build different microservices from source:
|
||||
|
||||
##### Set variable HUGGINGFACEHUB_API_TOKEN:
|
||||
| Microservice | Deployment Guide |
|
||||
| --------------- | ------------------------------------------------------------------------------------------------------------------ |
|
||||
| vLLM | [vLLM build guide](https://github.com/opea-project/GenAIComps/tree/main/comps/third_parties/vllm#build-docker) |
|
||||
| TGI | [TGI project](https://github.com/huggingface/text-generation-inference.git) |
|
||||
| LLM | [LLM build guide](https://github.com/opea-project/GenAIComps/tree/main/comps/llms) |
|
||||
| Redis Vector DB | [Redis](https://github.com/redis/redis.git) |
|
||||
| Dataprep | [Dataprep build guide](https://github.com/opea-project/GenAIComps/tree/main/comps/dataprep/src/README_redis.md) |
|
||||
| TEI Embedding | [TEI guide](https://github.com/huggingface/text-embeddings-inference.git) |
|
||||
| Retriever | [Retriever build guide](https://github.com/opea-project/GenAIComps/tree/main/comps/retrievers/src/README_redis.md) |
|
||||
| TEI Reranking | [TEI guide](https://github.com/huggingface/text-embeddings-inference.git) |
|
||||
| MegaService | [MegaService guide](../../../../README.md) |
|
||||
| UI | [UI guide](../../../../ui/react/README.md) |
|
||||
| Nginx | [Nginx guide](https://github.com/opea-project/GenAIComps/tree/main/comps/third_parties/nginx) |
|
||||
|
||||
### Check the Deployment Status
|
||||
|
||||
After running docker compose, check if all the containers launched via docker compose have started:
|
||||
|
||||
```bash
|
||||
### Replace the string 'your_huggingfacehub_token' with your HuggingFacehub repository access token.
|
||||
export HUGGINGFACEHUB_API_TOKEN='your_huggingfacehub_token'
|
||||
docker ps -a
|
||||
```
|
||||
|
||||
#### Set variables value in set_env\*\*\*\*.sh file:
|
||||
For the default deployment with TGI, the following 9 containers should have started:
|
||||
|
||||
Go to Docker Compose directory:
|
||||
|
||||
```bash
|
||||
cd ~/chatqna-install/GenAIExamples/ChatQnA/docker_compose/amd/gpu/rocm
|
||||
```
|
||||
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
|
||||
eaf24161aca8 opea/nginx:latest "/docker-entrypoint.…" 37 seconds ago Up 5 seconds 0.0.0.0:18104->80/tcp, [::]:18104->80/tcp chatqna-nginx-server
|
||||
2fce48a4c0f4 opea/chatqna-ui:latest "docker-entrypoint.s…" 37 seconds ago Up 5 seconds 0.0.0.0:18101->5173/tcp, [::]:18101->5173/tcp chatqna-ui-server
|
||||
613c384979f4 opea/chatqna:latest "bash entrypoint.sh" 37 seconds ago Up 5 seconds 0.0.0.0:18102->8888/tcp, [::]:18102->8888/tcp chatqna-backend-server
|
||||
05512bd29fee opea/dataprep:latest "sh -c 'python $( [ …" 37 seconds ago Up 36 seconds (healthy) 0.0.0.0:18103->5000/tcp, [::]:18103->5000/tcp chatqna-dataprep-service
|
||||
49844d339d1d opea/retriever:latest "python opea_retriev…" 37 seconds ago Up 36 seconds 0.0.0.0:7000->7000/tcp, [::]:7000->7000/tcp chatqna-retriever
|
||||
75b698fe7de0 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18808->80/tcp, [::]:18808->80/tcp chatqna-tei-reranking-service
|
||||
342f01bfdbb2 ghcr.io/huggingface/text-generation-inference:2.3.1-rocm"python3 /workspace/…" 37 seconds ago Up 36 seconds 0.0.0.0:18008->8011/tcp, [::]:18008->8011/tcp chatqna-tgi-service
|
||||
6081eb1c119d redis/redis-stack:7.2.0-v9 "/entrypoint.sh" 37 seconds ago Up 36 seconds 0.0.0.0:6379->6379/tcp, [::]:6379->6379/tcp, 0.0.0.0:8001->8001/tcp, [::]:8001->8001/tcp chatqna-redis-vector-db
|
||||
eded17420782 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18090->80/tcp, [::]:18090->80/tcp chatqna-tei-embedding-service
|
||||
```
|
||||
|
||||
The example uses the Nano text editor. You can use any convenient text editor:
|
||||
if used TGI with FaqGen:
|
||||
|
||||
#### If you use vLLM based application
|
||||
|
||||
```bash
|
||||
nano set_env_vllm.sh
|
||||
```
|
||||
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
|
||||
eaf24161aca8 opea/nginx:latest "/docker-entrypoint.…" 37 seconds ago Up 5 seconds 0.0.0.0:18104->80/tcp, [::]:18104->80/tcp chatqna-nginx-server
|
||||
2fce48a4c0f4 opea/chatqna-ui:latest "docker-entrypoint.s…" 37 seconds ago Up 5 seconds 0.0.0.0:18101->5173/tcp, [::]:18101->5173/tcp chatqna-ui-server
|
||||
613c384979f4 opea/chatqna:latest "bash entrypoint.sh" 37 seconds ago Up 5 seconds 0.0.0.0:18102->8888/tcp, [::]:18102->8888/tcp chatqna-backend-server
|
||||
e0ef1ea67640 opea/llm-faqgen:latest "bash entrypoint.sh" 37 seconds ago Up 36 seconds 0.0.0.0:18011->9000/tcp, [::]:18011->9000/tcp chatqna-llm-faqgen
|
||||
05512bd29fee opea/dataprep:latest "sh -c 'python $( [ …" 37 seconds ago Up 36 seconds (healthy) 0.0.0.0:18103->5000/tcp, [::]:18103->5000/tcp chatqna-dataprep-service
|
||||
49844d339d1d opea/retriever:latest "python opea_retriev…" 37 seconds ago Up 36 seconds 0.0.0.0:7000->7000/tcp, [::]:7000->7000/tcp chatqna-retriever
|
||||
75b698fe7de0 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18808->80/tcp, [::]:18808->80/tcp chatqna-tei-reranking-service
|
||||
342f01bfdbb2 ghcr.io/huggingface/text-generation-inference:2.3.1-rocm"python3 /workspace/…" 37 seconds ago Up 36 seconds 0.0.0.0:18008->8011/tcp, [::]:18008->8011/tcp chatqna-tgi-service
|
||||
6081eb1c119d redis/redis-stack:7.2.0-v9 "/entrypoint.sh" 37 seconds ago Up 36 seconds 0.0.0.0:6379->6379/tcp, [::]:6379->6379/tcp, 0.0.0.0:8001->8001/tcp, [::]:8001->8001/tcp chatqna-redis-vector-db
|
||||
eded17420782 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18090->80/tcp, [::]:18090->80/tcp chatqna-tei-embedding-service
|
||||
```
|
||||
|
||||
#### If you use vLLM based application with FaqGen
|
||||
if used vLLM:
|
||||
|
||||
```bash
|
||||
nano set_env_vllm_faqgen.sh
|
||||
```
|
||||
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
|
||||
eaf24161aca8 opea/nginx:latest "/docker-entrypoint.…" 37 seconds ago Up 5 seconds 0.0.0.0:18104->80/tcp, [::]:18104->80/tcp chatqna-nginx-server
|
||||
2fce48a4c0f4 opea/chatqna-ui:latest "docker-entrypoint.s…" 37 seconds ago Up 5 seconds 0.0.0.0:18101->5173/tcp, [::]:18101->5173/tcp chatqna-ui-server
|
||||
613c384979f4 opea/chatqna:latest "bash entrypoint.sh" 37 seconds ago Up 5 seconds 0.0.0.0:18102->8888/tcp, [::]:18102->8888/tcp chatqna-backend-server
|
||||
05512bd29fee opea/dataprep:latest "sh -c 'python $( [ …" 37 seconds ago Up 36 seconds (healthy) 0.0.0.0:18103->5000/tcp, [::]:18103->5000/tcp chatqna-dataprep-service
|
||||
49844d339d1d opea/retriever:latest "python opea_retriev…" 37 seconds ago Up 36 seconds 0.0.0.0:7000->7000/tcp, [::]:7000->7000/tcp chatqna-retriever
|
||||
75b698fe7de0 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18808->80/tcp, [::]:18808->80/tcp chatqna-tei-reranking-service
|
||||
342f01bfdbb2 opea/vllm-rocm:latest "python3 /workspace/…" 37 seconds ago Up 36 seconds 0.0.0.0:18008->8011/tcp, [::]:18008->8011/tcp chatqna-vllm-service
|
||||
6081eb1c119d redis/redis-stack:7.2.0-v9 "/entrypoint.sh" 37 seconds ago Up 36 seconds 0.0.0.0:6379->6379/tcp, [::]:6379->6379/tcp, 0.0.0.0:8001->8001/tcp, [::]:8001->8001/tcp chatqna-redis-vector-db
|
||||
eded17420782 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18090->80/tcp, [::]:18090->80/tcp chatqna-tei-embedding-service
|
||||
```
|
||||
|
||||
#### If you use TGI based application
|
||||
if used vLLM with FaqGen:
|
||||
|
||||
```bash
|
||||
nano set_env.sh
|
||||
```
|
||||
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
|
||||
eaf24161aca8 opea/nginx:latest "/docker-entrypoint.…" 37 seconds ago Up 5 seconds 0.0.0.0:18104->80/tcp, [::]:18104->80/tcp chatqna-nginx-server
|
||||
2fce48a4c0f4 opea/chatqna-ui:latest "docker-entrypoint.s…" 37 seconds ago Up 5 seconds 0.0.0.0:18101->5173/tcp, [::]:18101->5173/tcp chatqna-ui-server
|
||||
613c384979f4 opea/chatqna:latest "bash entrypoint.sh" 37 seconds ago Up 5 seconds 0.0.0.0:18102->8888/tcp, [::]:18102->8888/tcp chatqna-backend-server
|
||||
e0ef1ea67640 opea/llm-faqgen:latest "bash entrypoint.sh" 37 seconds ago Up 36 seconds 0.0.0.0:18011->9000/tcp, [::]:18011->9000/tcp chatqna-llm-faqgen
|
||||
05512bd29fee opea/dataprep:latest "sh -c 'python $( [ …" 37 seconds ago Up 36 seconds (healthy) 0.0.0.0:18103->5000/tcp, [::]:18103->5000/tcp chatqna-dataprep-service
|
||||
49844d339d1d opea/retriever:latest "python opea_retriev…" 37 seconds ago Up 36 seconds 0.0.0.0:7000->7000/tcp, [::]:7000->7000/tcp chatqna-retriever
|
||||
75b698fe7de0 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18808->80/tcp, [::]:18808->80/tcp chatqna-tei-reranking-service
|
||||
342f01bfdbb2 opea/vllm-rocm:latest "python3 /workspace/…" 37 seconds ago Up 36 seconds 0.0.0.0:18008->8011/tcp, [::]:18008->8011/tcp chatqna-vllm-service
|
||||
6081eb1c119d redis/redis-stack:7.2.0-v9 "/entrypoint.sh" 37 seconds ago Up 36 seconds 0.0.0.0:6379->6379/tcp, [::]:6379->6379/tcp, 0.0.0.0:8001->8001/tcp, [::]:8001->8001/tcp chatqna-redis-vector-db
|
||||
eded17420782 ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 "text-embeddings-rou…" 37 seconds ago Up 36 seconds 0.0.0.0:18090->80/tcp, [::]:18090->80/tcp chatqna-tei-embedding-service
|
||||
```
|
||||
|
||||
#### If you use TGI based application with FaqGen
|
||||
If any issues are encountered during deployment, refer to the [Troubleshooting](../../../../README_miscellaneous.md#troubleshooting) section.
|
||||
|
||||
```bash
|
||||
nano set_env_faqgen.sh
|
||||
```
|
||||
### Validate the Pipeline
|
||||
|
||||
If you are in a proxy environment, also set the proxy-related environment variables:
|
||||
|
||||
```bash
|
||||
export http_proxy="Your_HTTP_Proxy"
|
||||
export https_proxy="Your_HTTPs_Proxy"
|
||||
```
|
||||
|
||||
Set the values of the variables:
|
||||
|
||||
- **HOST_IP, HOST_IP_EXTERNAL** - These variables are used to configure the name/address of the service in the operating system environment for the application services to interact with each other and with the outside world.
|
||||
|
||||
If your server uses only an internal address and is not accessible from the Internet, then the values for these two variables will be the same and the value will be equal to the server's internal name/address.
|
||||
|
||||
If your server uses only an external, Internet-accessible address, then the values for these two variables will be the same and the value will be equal to the server's external name/address.
|
||||
|
||||
If your server is located on an internal network, has an internal address, but is accessible from the Internet via a proxy/firewall/load balancer, then the HOST_IP variable will have a value equal to the internal name/address of the server, and the EXTERNAL_HOST_IP variable will have a value equal to the external name/address of the proxy/firewall/load balancer behind which the server is located.
|
||||
|
||||
We set these values in the file set_env\*\*\*\*.sh
|
||||
|
||||
- **Variables with names like "**\*\*\*\*\*\*\_PORT"\*\* - These variables set the IP port numbers for establishing network connections to the application services.
|
||||
The values shown in the file set_env.sh or set_env_vllm they are the values used for the development and testing of the application, as well as configured for the environment in which the development is performed. These values must be configured in accordance with the rules of network access to your environment's server, and must not overlap with the IP ports of other applications that are already in use.
|
||||
|
||||
#### Set variables with script set_env\*\*\*\*.sh
|
||||
|
||||
#### If you use vLLM based application
|
||||
|
||||
```bash
|
||||
. set_env_vllm.sh
|
||||
```
|
||||
|
||||
#### If you use vLLM based application with FaqGen
|
||||
|
||||
```bash
|
||||
. set_env_faqgen_vllm.sh
|
||||
```
|
||||
|
||||
#### If you use TGI based application
|
||||
|
||||
```bash
|
||||
. set_env.sh
|
||||
```
|
||||
|
||||
#### If you use TGI based application with FaqGen
|
||||
|
||||
```bash
|
||||
. set_env_faqgen.sh
|
||||
```
|
||||
|
||||
### Start the services:
|
||||
|
||||
#### If you use vLLM based application
|
||||
|
||||
```bash
|
||||
docker compose -f compose_vllm.yaml up -d
|
||||
```
|
||||
|
||||
#### If you use vLLM based application with FaqGen
|
||||
|
||||
```bash
|
||||
docker compose -f compose_faqgen_vllm.yaml up -d
|
||||
```
|
||||
|
||||
#### If you use TGI based application
|
||||
|
||||
```bash
|
||||
docker compose -f compose.yaml up -d
|
||||
```
|
||||
|
||||
#### If you use TGI based application with FaqGen
|
||||
|
||||
```bash
|
||||
docker compose -f compose_faqgen.yaml up -d
|
||||
```
|
||||
|
||||
All containers should be running and should not restart:
|
||||
|
||||
##### If you use vLLM based application:
|
||||
|
||||
- chatqna-redis-vector-db
|
||||
- chatqna-dataprep-service
|
||||
- chatqna-tei-embedding-service
|
||||
- chatqna-retriever
|
||||
- chatqna-tei-reranking-service
|
||||
- chatqna-vllm-service
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
- chatqna-nginx-server
|
||||
|
||||
##### If you use vLLM based application with FaqGen:
|
||||
|
||||
- chatqna-redis-vector-db
|
||||
- chatqna-dataprep-service
|
||||
- chatqna-tei-embedding-service
|
||||
- chatqna-retriever
|
||||
- chatqna-tei-reranking-service
|
||||
- chatqna-vllm-service
|
||||
- chatqna-llm-faqgen
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
- chatqna-nginx-server
|
||||
|
||||
##### If you use TGI based application:
|
||||
|
||||
- chatqna-redis-vector-db
|
||||
- chatqna-dataprep-service
|
||||
- chatqna-tei-embedding-service
|
||||
- chatqna-retriever
|
||||
- chatqna-tei-reranking-service
|
||||
- chatqna-tgi-service
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
- chaqna-nginx-server
|
||||
|
||||
##### If you use TGI based application with FaqGen:
|
||||
|
||||
- chatqna-redis-vector-db
|
||||
- chatqna-dataprep-service
|
||||
- chatqna-tei-embedding-service
|
||||
- chatqna-retriever
|
||||
- chatqna-tei-reranking-service
|
||||
- chatqna-tgi-service
|
||||
- chatqna-llm-faqgen
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
- chaqna-nginx-server
|
||||
|
||||
---
|
||||
|
||||
## Validate the Services
|
||||
|
||||
### 1. Validate TEI Embedding Service
|
||||
|
||||
```bash
|
||||
curl http://${HOST_IP}:${CHATQNA_TEI_EMBEDDING_PORT}/embed \
|
||||
-X POST \
|
||||
-d '{"inputs":"What is Deep Learning?"}' \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to text:
|
||||
|
||||
```textmate
|
||||
[[0.00037115702,-0.06356819,0.0024758505,..................,0.022725677,0.016026087,-0.02125421,-0.02984927,-0.0049473033]]
|
||||
```
|
||||
|
||||
If the service response has a meaningful response in the value,
|
||||
then we consider the TEI Embedding Service to be successfully launched
|
||||
|
||||
### 2. Validate Retriever Microservice
|
||||
|
||||
```bash
|
||||
export your_embedding=$(python3 -c "import random; embedding = [random.uniform(-1, 1) for _ in range(768)]; print(embedding)")
|
||||
curl http://${HOST_IP}:${CHATQNA_REDIS_RETRIEVER_PORT}/v1/retrieval \
|
||||
-X POST \
|
||||
-d "{\"text\":\"test\",\"embedding\":${your_embedding}}" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to JSON:
|
||||
|
||||
```json
|
||||
{ "id": "e191846168aed1f80b2ea12df80844d2", "retrieved_docs": [], "initial_query": "test", "top_n": 1, "metadata": [] }
|
||||
```
|
||||
|
||||
If the response corresponds to the form of the provided JSON, then we consider the
|
||||
Retriever Microservice verification successful.
|
||||
|
||||
### 3. Validate TEI Reranking Service
|
||||
|
||||
```bash
|
||||
curl http://${HOST_IP}:${CHATQNA_TEI_RERANKING_PORT}/rerank \
|
||||
-X POST \
|
||||
-d '{"query":"What is Deep Learning?", "texts": ["Deep Learning is not...", "Deep learning is..."]}' \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to JSON:
|
||||
|
||||
```json
|
||||
[
|
||||
{ "index": 1, "score": 0.94238955 },
|
||||
{ "index": 0, "score": 0.120219156 }
|
||||
]
|
||||
```
|
||||
|
||||
If the response corresponds to the form of the provided JSON, then we consider the TEI Reranking Service
|
||||
verification successful.
|
||||
|
||||
### 4. Validate the vLLM/TGI Service
|
||||
|
||||
#### If you use vLLM:
|
||||
|
||||
```bash
|
||||
DATA='{"model": "meta-llama/Meta-Llama-3-8B-Instruct", '\
|
||||
'"messages": [{"role": "user", "content": "What is a Deep Learning?"}], "max_tokens": 64}'
|
||||
|
||||
curl http://${HOST_IP}:${CHATQNA_VLLM_SERVICE_PORT}/v1/chat/completions \
|
||||
-X POST \
|
||||
-d "$DATA" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to JSON:
|
||||
|
||||
```json
|
||||
{
|
||||
"id": "chatcmpl-91003647d1c7469a89e399958f390f67",
|
||||
"object": "chat.completion",
|
||||
"created": 1742877228,
|
||||
"model": "meta-llama/Meta-Llama-3-8B-Instruct",
|
||||
"choices": [
|
||||
{
|
||||
"index": 0,
|
||||
"message": {
|
||||
"role": "assistant",
|
||||
"content": "Deep Learning ( DL) is a subfield of Machine Learning (ML) that focuses on the design of algorithms and architectures inspired by the structure and function of the human brain. These algorithms are designed to analyze and interpret data that is presented in the form of patterns or signals, and they often mimic the way the human brain",
|
||||
"tool_calls": []
|
||||
},
|
||||
"logprobs": null,
|
||||
"finish_reason": "length",
|
||||
"stop_reason": null
|
||||
}
|
||||
],
|
||||
"usage": { "prompt_tokens": 16, "total_tokens": 80, "completion_tokens": 64, "prompt_tokens_details": null },
|
||||
"prompt_logprobs": null
|
||||
}
|
||||
```
|
||||
|
||||
If the service response has a meaningful response in the value of the "choices.message.content" key,
|
||||
then we consider the vLLM service to be successfully launched
|
||||
|
||||
#### If you use TGI:
|
||||
|
||||
```bash
|
||||
DATA='{"inputs":"What is a Deep Learning?",'\
|
||||
'"parameters":{"max_new_tokens":64,"do_sample": true}}'
|
||||
|
||||
curl http://${HOST_IP}:${CHATQNA_TGI_SERVICE_PORT}/generate \
|
||||
-X POST \
|
||||
-d "$DATA" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to JSON:
|
||||
|
||||
```json
|
||||
{
|
||||
"generated_text": " What is its application in Computer Vision?\nWhat is a Deep Learning?\nDeep learning is a subfield of machine learning that involves the use of artificial neural networks to model high-level abstractions in data. It involves the use of deep neural networks, which are composed of multiple layers, to learn complex patterns in data. The"
|
||||
}
|
||||
```
|
||||
|
||||
If the service response has a meaningful response in the value of the "generated_text" key,
|
||||
then we consider the TGI service to be successfully launched
|
||||
|
||||
### 5. Validate the LLM Service (if your used application with FaqGen)
|
||||
|
||||
```bash
|
||||
DATA='{"messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source '\
|
||||
'text embeddings and sequence classification models. TEI enables high-performance extraction for the most '\
|
||||
'popular models, including FlagEmbedding, Ember, GTE and E5.","max_tokens": 128}'
|
||||
|
||||
curl http://${HOST_IP}:${CHATQNA_LLM_FAQGEN_PORT}/v1/faqgen \
|
||||
-X POST \
|
||||
-d "$DATA" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to JSON:
|
||||
|
||||
```json
|
||||
{
|
||||
"id": "58f0632f5f03af31471b895b0d0d397b",
|
||||
"text": " Q: What is Text Embeddings Inference (TEI)?\n A: TEI is a toolkit for deploying and serving open source text embeddings and sequence classification models.\n\n Q: What models does TEI support?\n A: TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5.\n\n Q: What is the purpose of TEI?\n A: The purpose of TEI is to enable high-performance extraction for text embeddings and sequence classification models.\n\n Q: What are the benefits of using TEI?\n A: The benefits of using TEI include high",
|
||||
"prompt": "Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."
|
||||
}
|
||||
```
|
||||
|
||||
If the service response has a meaningful response in the value of the "text" key,
|
||||
then we consider the LLM service to be successfully launched
|
||||
|
||||
### 6. Validate the MegaService
|
||||
Once the ChatQnA services are running, test the pipeline using the following command:
|
||||
|
||||
```bash
|
||||
curl http://${HOST_IP}:${CHATQNA_BACKEND_SERVICE_PORT}/v1/chatqna \
|
||||
@@ -531,91 +229,105 @@ curl http://${HOST_IP}:${CHATQNA_BACKEND_SERVICE_PORT}/v1/chatqna \
|
||||
-d '{"messages": "What is the revenue of Nike in 2023?"}'
|
||||
```
|
||||
|
||||
Checking the response from the service. The response should be similar to text:
|
||||
**Note** : Access the ChatQnA UI by web browser through this URL: `http://${HOST_IP_EXTERNAL}:${CHATQNA_NGINX_PORT}`
|
||||
|
||||
```textmate
|
||||
data: b' What'
|
||||
data: b' is'
|
||||
data: b' the'
|
||||
data: b' revenue'
|
||||
data: b' of'
|
||||
data: b' Nike'
|
||||
data: b' in'
|
||||
data: b' '
|
||||
data: b'202'
|
||||
data: b'3'
|
||||
data: b'?\n'
|
||||
data: b' '
|
||||
data: b' Answer'
|
||||
data: b':'
|
||||
data: b' According'
|
||||
data: b' to'
|
||||
data: b' the'
|
||||
data: b' search'
|
||||
data: b' results'
|
||||
data: b','
|
||||
data: b' the'
|
||||
data: b' revenue'
|
||||
data: b' of'
|
||||
data: b''
|
||||
### Cleanup the Deployment
|
||||
|
||||
data: [DONE]
|
||||
|
||||
```
|
||||
|
||||
If the output lines in the "data" keys contain words (tokens) containing meaning, then the service
|
||||
is considered launched successfully.
|
||||
|
||||
### 7. Validate the Frontend (UI)
|
||||
|
||||
To access the UI, use the URL - http://${EXTERNAL_HOST_IP}:${CHATQNA_NGINX_PORT}
|
||||
A page should open when you click through to this address:
|
||||
|
||||
|
||||
|
||||
If a page of this type has opened, then we believe that the service is running and responding,
|
||||
and we can proceed to functional UI testing.
|
||||
|
||||
Let's enter the task for the service in the "Enter prompt here" field.
|
||||
For example, "What is a Deep Learning?" and press Enter.
|
||||
After that, a page with the result of the task should open:
|
||||
|
||||
#### If used application without FaqGen
|
||||
|
||||
|
||||
|
||||
#### If used application with FaqGen
|
||||
|
||||
|
||||
|
||||
If the result shown on the page is correct, then we consider the verification of the UI service to be successful.
|
||||
|
||||
### 5. Stop application
|
||||
|
||||
#### If you use vLLM
|
||||
To stop the containers associated with the deployment, execute the following command:
|
||||
|
||||
```bash
|
||||
cd ~/chatqna-install/GenAIExamples/ChatQnA/docker_compose/amd/gpu/rocm
|
||||
docker compose -f compose_vllm.yaml down
|
||||
```
|
||||
|
||||
#### If you use vLLM with FaqGen
|
||||
|
||||
```bash
|
||||
cd ~/chatqna-install/GenAIExamples/ChatQnA/docker_compose/amd/gpu/rocm
|
||||
docker compose -f compose_faqgen_vllm.yaml down
|
||||
```
|
||||
|
||||
#### If you use TGI
|
||||
|
||||
```bash
|
||||
cd ~/chatqna-install/GenAIExamples/ChatQnA/docker_compose/amd/gpu/rocm
|
||||
# if used TGI
|
||||
docker compose -f compose.yaml down
|
||||
# if used TGI with FaqGen
|
||||
# docker compose -f compose_faqgen.yaml down
|
||||
# if used vLLM
|
||||
# docker compose -f compose_vllm.yaml down
|
||||
# if used vLLM with FaqGen
|
||||
# docker compose -f compose_faqgen_vllm.yaml down
|
||||
|
||||
|
||||
```
|
||||
|
||||
#### If you use TGI with FaqGen
|
||||
## ChatQnA Docker Compose Files
|
||||
|
||||
```bash
|
||||
cd ~/chatqna-install/GenAIExamples/ChatQnA/docker_compose/amd/gpu/rocm
|
||||
docker compose -f compose_faqgen.yaml down
|
||||
```
|
||||
In the context of deploying an ChatQnA pipeline on an Intel® Xeon® platform, we can pick and choose different large language model serving frameworks, or single English TTS/multi-language TTS component. The table below outlines the various configurations that are available as part of the application. These configurations can be used as templates and can be extended to different components available in [GenAIComps](https://github.com/opea-project/GenAIComps.git).
|
||||
|
||||
| File | Description |
|
||||
| ------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------ |
|
||||
| [compose.yaml](./compose.yaml) | The LLM serving framework is TGI. Default compose file using TGI as serving framework and redis as vector database |
|
||||
| [compose_faqgen.yaml](./compose_faqgen.yaml) | The LLM serving framework is TGI with FaqGen. All other configurations remain the same as the default |
|
||||
| [compose_vllm.yaml](./compose_vllm.yaml) | The LLM serving framework is vLLM. Compose file using vllm as serving framework and redis as vector database |
|
||||
| [compose_faqgen_vllm.yaml](./compose_faqgen_vllm.yaml) | The LLM serving framework is vLLM with FaqGen. Compose file using vllm as serving framework and redis as vector database |
|
||||
|
||||
## Validate MicroServices
|
||||
|
||||
1. TEI Embedding Service
|
||||
|
||||
```bash
|
||||
curl http://${HOST_IP}:${CHATQNA_TEI_EMBEDDING_PORT}/embed \
|
||||
-X POST \
|
||||
-d '{"inputs":"What is Deep Learning?"}' \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
2. Retriever Microservice
|
||||
|
||||
```bash
|
||||
export your_embedding=$(python3 -c "import random; embedding = [random.uniform(-1, 1) for _ in range(768)]; print(embedding)")
|
||||
curl http://${HOST_IP}:${CHATQNA_REDIS_RETRIEVER_PORT}/v1/retrieval \
|
||||
-X POST \
|
||||
-d "{\"text\":\"test\",\"embedding\":${your_embedding}}" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
3. TEI Reranking Service
|
||||
|
||||
```bash
|
||||
curl http://${HOST_IP}:${CHATQNA_TEI_RERANKING_PORT}/rerank \
|
||||
-X POST \
|
||||
-d '{"query":"What is Deep Learning?", "texts": ["Deep Learning is not...", "Deep learning is..."]}' \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
4. vLLM/TGI Service
|
||||
|
||||
If you use vLLM:
|
||||
|
||||
```bash
|
||||
DATA='{"model": "meta-llama/Meta-Llama-3-8B-Instruct", '\
|
||||
'"messages": [{"role": "user", "content": "What is a Deep Learning?"}], "max_tokens": 64}'
|
||||
|
||||
curl http://${HOST_IP}:${CHATQNA_VLLM_SERVICE_PORT}/v1/chat/completions \
|
||||
-X POST \
|
||||
-d "$DATA" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
If you use TGI:
|
||||
|
||||
```bash
|
||||
DATA='{"inputs":"What is a Deep Learning?",'\
|
||||
'"parameters":{"max_new_tokens":64,"do_sample": true}}'
|
||||
|
||||
curl http://${HOST_IP}:${CHATQNA_TGI_SERVICE_PORT}/generate \
|
||||
-X POST \
|
||||
-d "$DATA" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
5. LLM Service (if your used application with FaqGen)
|
||||
|
||||
```bash
|
||||
DATA='{"messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source '\
|
||||
'text embeddings and sequence classification models. TEI enables high-performance extraction for the most '\
|
||||
'popular models, including FlagEmbedding, Ember, GTE and E5.","max_tokens": 128}'
|
||||
|
||||
curl http://${HOST_IP}:${CHATQNA_LLM_FAQGEN_PORT}/v1/faqgen \
|
||||
-X POST \
|
||||
-d "$DATA" \
|
||||
-H 'Content-Type: application/json'
|
||||
```
|
||||
|
||||
## Conclusion
|
||||
|
||||
This guide should enable developers to deploy the default configuration or any of the other compose yaml files for different configurations. It also highlights the configurable parameters that can be set before deployment.
|
||||
|
||||
@@ -165,7 +165,7 @@ services:
|
||||
|
||||
chatqna-nginx-server:
|
||||
image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
|
||||
container_name: chaqna-nginx-server
|
||||
container_name: chatqna-nginx-server
|
||||
depends_on:
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
|
||||
@@ -187,7 +187,7 @@ services:
|
||||
|
||||
chatqna-nginx-server:
|
||||
image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
|
||||
container_name: chaqna-nginx-server
|
||||
container_name: chatqna-nginx-server
|
||||
depends_on:
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
|
||||
@@ -192,7 +192,7 @@ services:
|
||||
|
||||
chatqna-nginx-server:
|
||||
image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
|
||||
container_name: chaqna-nginx-server
|
||||
container_name: chatqna-nginx-server
|
||||
depends_on:
|
||||
- chatqna-backend-server
|
||||
- chatqna-ui-server
|
||||
|
||||
Reference in New Issue
Block a user