forked from cloudera/CML_llm-hol
-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy path.project-metadata.yaml
60 lines (56 loc) · 1.98 KB
/
.project-metadata.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
name: LLM Hands on Lab with CML
description: "Hands on Lab which demonstrates a number of concepts including web scraping, vector databases, model deployment, model usage, Langchain, application building, and instruction following/tuning."
author: "Cloudera"
date: "2023-12-20"
specification_version: 1.0
prototype_version: 1.0
environment_variables:
#AWS_ACCESS_KEY_ID:
# default: ""
# description: "Access key ID for interacting with AWS Bedrock"
#AWS_SECRET_ACCESS_KEY:
# default: ""
# description: "Secret ket for interacting with AWS Bedrock"
#AWS_DEFAULT_REGION:
# default: "us-west-2"
# description: "AWS Region where Bedrock models are available"
#PINECONE_API_KEY:
# default: ""
# description: "API Keys for Pinecone service"
#PINECONE_ENVIRONMENT:
# default: ""
# description: "Environment within Pinecone service"
PINECONE_INDEX:
default: ""
description: "Index within Pinecone Environment"
runtimes:
- editor: JupyterLab
kernel: Python 3.10
edition: Nvidia GPU
tasks:
- type: run_session
name: Install Dependencies
script: 0_install_prerequisites/download_reqs_set_vars.py
short_summary: Install Dependencies
kernel: python3
cpu: 2
memory: 12
- type: run_session
name: Setup Chroma Server
script: 0_install_prerequisites/setup-chroma.py
short_summary: Setup Chroma Server Vector DB
long_summary: Resolve sqlite dependency with python version and control logging in Chroma server package
kernel: python3
cpu: 2
memory: 4
- type: create_job
name: Pull and Convert HTMLS to TXT
entity_label: htmls_to_txt
script: 2_populate_vector_db/html_to_text.py
arguments: None
short_summary: Create job to source htmls, download, and convert them to text.
long_summary: Create job to source htmls, download, and convert them to text. Update html_links.txt with the htmls you wish to download and convert.
cpu: 2
memory: 4
environment:
TASK_TYPE: CREATE/RUN_JOB