AWS Crash Course for Python Data Science with NVIDIA GPUs: EC2, IAM, SPOT, Tunneling

In this video I provide a tutorial for launching EC2 instances for use with data science from Windows 10. I show how to launch a spot instance to save money, preparing a Putty PPK, and using a Putty tunnel to access JupyterLab.

Putty
https://www.putty.org/

WinSCP
https://winscp.net

Commands from Video

nvidia-smi
top
docker run -it --rm --gpus all -p 8888:8888 tensorflow/tensorflow:latest-gpu-py3-jupyter

0:55 AIM
3:10 Launch EC2 and Key Pairs
4:49 Instance Details and Spot Instances
6:51 Login Key Pair
8:32 Login to EC2 Instance with Putty
9:03 Generate a Putty PPK File
10:21 SSH Tunnels and Putty
16:42 Jupyter

Follow Me/Subscribe:
https://www.youtube.com/user/HeatonResearch
https://github.com/jeffheaton
https://twitter.com/jeffheaton

Support Me on Patreon: https://www.patreon.com/jeffheaton

Тэги:

#jeff_heaton #aws #amazon_web_services #docker #aws_tunneling #ssh_tunnel #spot_instances #aws_spot_instances #aws_spot_instances_for_machine_learning #aws_iam #python #python_aws #nvidia #gpu #cuda #aws_cli #aws_gpu #aws_spot_gpu #NVIDIA #GPU #RAPIDS #NGC #AWS #Putty #putty_tunnel #putty_tunnel_jupyter #WinSCP #AWS_IAM #IAM #data_science #machine_learning #deep_learning #ai #artificial_intelligence #CUDA #CUDA_drivers #tutorial #NGC_Tutorial #EC2_Jupyter_Lab #NGC_containers #ngc #pytorch

Скачать видео