ma7dev

Sudomaze - Currently working on

Today

• Lab 02a: PyTorch Lightning
• Lab 02b: Training a CNN on Synthetic Handwriting Data
• Lab 03: Transformers and Paragraphs
• srush/Tensor-Puzzles: Solve puzzles. Improve your pytorch.

---

ma7dev

ma7dev

Intro

It wasn't obvious on PyTorch's documentation of how to use PyTorch Profiler (as of today, 8/12/2021), so I have spent some time to understand how to use it and this gist contains a simple example to use.

Instructions

• Install the required packages:

python>=1.9.0
torchvision>=0.10.0
numpy
matplotlib

ma7dev

# Create 4 tunnels, each for different ports, with only https enabled
# This way the ngrok process stays bellow the Free plan limit (4 tunnels)
# command: ngrok start --all # to start all of them
# command: ngrok start note tb # to run jupyter notebook server and tensorboard server only
# refer to this page for more info: https://ngrok.com/docs#multiple-tunnels

authtoken: ...
log: ngrok.log
tunnels:
  # to run jupyter notebook server

ma7dev

name: Automated Experiments
on:
  push:
    branches: [ main ]
jobs:
  build_on_cn-gpu5:
    runs-on: [ cn-gpu5 ]
    timeout-minutes: 4320
    defaults:
      run:

ma7dev

To run access a remote jupyter notebook, you will need to do the following:

1. Setting ssh config file (~/.ssh/config):

Host FLIP # access point server
    HostName access.engr.oregonstate.edu
    User ONIDusername
Host DGX # target server
    HostName submit-b.hpc.engr.oregonstate.edu
    User ONIDusername

ma7dev

A simple example for Streamlit and Flask project

• server.py contains the API code for the Flask server (back-end)
• main.py contains the streamlit code (front-end)

Requirements

pip install plotly streamlit pandas requests flask

ma7dev