forked from nkityd09/cml_speech_to_text
-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy path.project-metadata.yaml
69 lines (63 loc) · 2.04 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
61
62
63
64
65
66
67
68
69
name: CML Speech To Text
description: |
This AMP uses OpenAI Whisper to transcribe Audio and an LLM model of your choice to summarize the transcribed text.
author: Cloudera Inc.
specification_version: 1.0
prototype_version: 1.0
date: "2023-09-27"
environment_variables:
HF_MODEL:
default: "meta-llama/Llama-2-7b-chat-hf"
description: >-
Provide HuggingFace Model you want to use
HF_TOKEN:
default: ""
description: >-
Provide HuggingFace Token for Gated Models
runtimes:
- editor: JupyterLab
kernel: Python 3.9
edition: S2T Edition
tasks:
- type: run_session
name: Validate GPU Availibility
script: session-resource-validation/check_gpu_resources.py
short_summary: Check for GPU availibility.
long_summary: Check GPUs are enabled on this workspace and are currently schedulable.
kernel: python3
cpu: 2
memory: 8
- type: run_session
name: Install Dependencies
script: session-install-deps/install_setup_tools.py
short_summary: Install Dependencies
kernel: python3
cpu: 2
memory: 16
- type: run_session
name: Validate GPU Capability
script: session-resource-validation/check_gpu_capability.py
short_summary: Check for GPU capability.
long_summary: Check GPU device supports the CUDA capabilities required.
kernel: python3
cpu: 2
memory: 16
gpu: 1
- type: run_session
name: Install Dependencies
script: session-install-deps/install_dependencies.py
short_summary: Install Dependencies
kernel: python3
cpu: 2
memory: 16
- type: start_application
name: Speech To Text App
subdomain: cmlllm
script: gradio-app/app.py
short_summary: Start CML LLM Chatbot application
long_summary: This application requires an available GPU to run the LLM model. Startup may be delayed if autoscaling is being performed or fail if GPU cannot be scheduled on this workspace. Please contact your administrator for GPU scheduling.
cpu: 4
memory: 32
gpu: 2
environment_variables:
TASK_TYPE: START_APPLICATION