The python tool for Open Platform for AI
Project description
PyPAI
This is a python tool for Open Platform for AI.
Installation
pip install pypai
Demo
- Create and run
submit.py
in your project folder to generate the configuration:
from pypai import PAI
# Create a PAI cluster
pai = PAI(username='xxx', passwd='xxx')
# Generate the configuration
pai.generate_config()
-
Open
pai_configuration.json
and modify it,The path of code dir MUST follows '/$PAI_DEFAULT_FS_URI/.../$PAI_USER_NAME~$PAI_JOB_NAME'. The default path is $PAI_DEFAULT_FS_URI/Users/$PAI_USER_NAME/$PAI_USER_NAME~$PAI_JOB_NAME. -
Change and run
submit.py
to submit your job:
from pypai import PAI
# Create a PAI cluster
pai = PAI(username='xxx', passwd='xxx')
# Submit job
pai.submit()
All the code in your project folder will be uploaded and your job will be submitted to the PAI.
API
class PAI:
@paremeters:
username: (str) PAI username
passwd: (str) PAI password
url: (str) PAI master node IP
worker: (int) thread number for uploading code
function generate_config:
@paremeters:
jobName='test'
image='192.168.193.253:5000/pytorch'
dataDir=''
outputDir=''
codeDir='$PAI_DEFAULT_FS_URI/Users/$PAI_USER_NAME/$PAI_USER_NAME~$PAI_JOB_NAME'
gpuType=''
function submit:
@paremeters:
config_path='./pai_configuration.json'
dir_path='./' code path
dest_dir='/Users/' code root path in the hdfs
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
No source distribution files available for this release.See tutorial on generating distribution archives.