-
Notifications
You must be signed in to change notification settings - Fork 6
/
example_config.yaml
83 lines (83 loc) · 4.36 KB
/
example_config.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
accountNumber: "012345678901"
region: us-east-1
authConfig:
authority:
clientId:
adminGroup:
jwtGroupsProperty:
s3BucketModels: hf-models-gaiic
########################### OPTIONAL BELOW #######################################
# profile: AWS CLI profile for deployment.
# vpcId: VPC ID for the application. (e.g. vpc-0123456789abcdef)
# The following is an array of subnet objects for the application. These contain a subnetId(e.g. [subnet-fedcba9876543210] and ipv4CidrBlock
# subnets:
# - subnetId:
# ipv4CidrBlock:
# The following configuration will allow for using a custom domain for the chat user interface.
# If this option is specified, the API Gateway invocation URL will NOT work on its own as the application URL.
# Users must use the custom domain for the user interface to work if this option is populated.
# apiGatewayConfig:
# domainName:
# restApiConfig:
# sslCertIamArn: ARN of the self-signed cert to be used throughout the system
# Some customers will want to download required libs prior to deployment, provide a path to the zipped resources
# lambdaLayerAssets:
# authorizerLayerPath: /path/to/authorizer_layer.zip
# commonLayerPath: /path/to/common_layer.zip
# fastapiLayerPath: /path/to/fastapi_layer.zip
# ragLayerPath: /path/to/rag_layer.zip
# sdkLayerPath: /path/to/sdk_layer.zip
# stackSynthesizer: CliCredentialsStackSynthesizer
# deploymentPrefix: Prefix for deployment resources.
# webAppAssetsPath: Optional path to precompiled webapp assets. If not specified the web application will be built at deploy time.
# permissionsBoundaryAspect:
# permissionsBoundaryPolicyName: CustomPermissionBoundary
# rolePrefix: CustomPrefix
# policyPrefix: CustomPrefix
# instanceProfilePrefix: CustomPrefix
# vpcId: vpc-0123456789abcdef,
# aws-iso partition mountS3 package location
# mountS3DebUrl: https://mountpoint-s3-release-us-iso-east-1.s3.us-iso-east-1.c2s.ic.gov/latest/x86_64/mount-s3.deb
# aws-iso-b partition mountS3 package location
# mountS3DebUrl: https://mountpoint-s3-release-us-isob-east-1.s3.us-isob-east-1.sc2s.sgov.gov/latest/x86_64/mount-s3.deb
# List of AWS account numbers for ECR repositories.
# accountNumbersEcr:
# - 012345678901
# ragRepositories:
# - repositoryId: pgvector-rag
# type: pgvector
# rdsConfig:
# username: postgres
# - repositoryId: default
# type: opensearch
# opensearchConfig:
# dataNodes: 2
# dataNodeInstanceType: r6g.large.search
# masterNodes: 0
# masterNodeInstanceType: r6g.large.search
# volumeSize: 300
# If adding an existing PGVector database, this configurations assumes:
# 1. The database has been configured to have pgvector installed and enabled: https://aws.amazon.com/about-aws/whats-new/2023/05/amazon-rds-postgresql-pgvector-ml-model-integration/
# 2. The database is accessible by RAG-related lambda functions (add inbound PostgreSQL access on the database's security group for all Lambda RAG security groups)
# 3. A secret ID exists in SecretsManager holding the database password within a json block of '{"password":"your_password_here"}'. This is the same format that RDS natively provides a password in SecretsManager.
# If the passwordSecretId or dbHost are not provided, then a sample database will be created for you. Only the username is required.
# - repositoryId: pgvector-rag
# type: pgvector
# rdsConfig:
# username: postgres
# passwordSecretId: # password ID as stored in SecretsManager. Example: "rds!db-aa88493d-be8d-4a3f-96dc-c668165f7826"
# dbHost: # Host name of database. Example hostname from RDS: "my-db-name.291b2f03.us-east-1.rds.amazonaws.com"
# dbName: postgres
# You can optionally provide a list of models and the deployment process will ensure they exist in your model bucket and try to download them if they don't exist
# ecsModels:
# - modelName: mistralai/Mistral-7B-Instruct-v0.2
# inferenceContainer: tgi
# baseImage: ghcr.io/huggingface/text-generation-inference:2.0.1
# - modelName: intfloat/e5-large-v2
# inferenceContainer: tei
# baseImage: ghcr.io/huggingface/text-embeddings-inference:1.2.3
# - modelName: mistralai/Mixtral-8x7B-Instruct-v0.1
# inferenceContainer: tgi
# baseImage: ghcr.io/huggingface/text-generation-inference:2.0.1
# litellmConfig:
# db_key: sk-d7a77bcb-3e23-483c-beec-2700f2baeeb1 # A key is required for model management purposes - must start with sk-