AWS setup
2017.09.28 register AMS account ,it is free till 2018.09.27
AWS free Tier https://amazonaws-china.com/cn/free/?sc_ichannel=ha≻_icampaign=free-tier≻_icontent=2234
Get started from here :http://course.fast.ai/start.html
Platform: AWS t2.micro instance ,which is free.
AWS /aws.amazon.com
AWS Graph console: console.aws.amazon.com
command interface :
choose EC2 instanc:
https://us-west-2.console.aws.amazon.com/ec2/v2/home?region=us-west-2
Step 1: Choose an Amazon Machine Image (AMI):
- Deep Learning AMI Ubuntu Version 2.3_Sep2017 - ami-d6ee1dae
Configure T2 instance:
share files: http://files.fast.ai/
pip install kaggle-cli
brew install awscli
aws configure #input access key & Secret access key
cdfast
./setup_t2.sh #t2-micro is free
Output:
True
Waiting for instance start...
All done. Find all you need to connect in the fast-ai-commands.txt file
and to remove the stack call fast-ai-remove.sh
Connect to your instance:
ssh -i /Users/davidmeng/.ssh/aws-key-fast-ai.pem [email protected]
#connect to AWS from local
$aws-ssh
ubuntu@ip-10-0-0-12:~$ nvidia-smi #check GPU ; Since t2-micro has no GPU
#launch jupyter notebook
jupyter notebook
#Open the browser with the [IP]:8888
#IP 出现在ssh -i
e.g. http://34.215.132.142:8888
#pwd: dl_course
nbs -> new[conda]
#if you want to close the instance,go to the AWS console :
https://us-west-2.console.aws.amazon.com/ec2/v2/home?region=us-west-2#Instances:
# Action -> instance State ->stop
aws alias
- add aws-alias.sh to ~/.zshrc and reload it.
- get t2.micro IP
aws-get-t2 aws-start #display IP ssh ubuntu@$instanceIp- alias
T2.micro
34.208.138.154:8888
T2.xlarge
34.215.132.142:8888