amir-saniyan

Ubuntu 22.04 for Deep Learning

In the name of God

This gist contains steps to setup Ubuntu 22.04 for deep learning.

---

Install Ubuntu 22.04

karpathy

"""
stable diffusion dreaming
creates hypnotic moving videos by smoothly walking randomly through the sample space

example way to run this script:

$ python stablediffusionwalk.py --prompt "blueberry spaghetti" --name blueberry

to stitch together the images, e.g.:
$ ffmpeg -r 10 -f image2 -s 512x512 -i blueberry/frame%06d.jpg -vcodec libx264 -crf 10 -pix_fmt yuv420p blueberry.mp4

henrik

tmux cheatsheet

As configured in my dotfiles.

start new:

tmux

start new with session name:

aluo-x

import torch
import torch.distributed as dist
import torch.nn as nn
import torch.multiprocessing as mp

from torch.nn.parallel import DistributedDataParallel as DDP
from fairscale.nn.data_parallel import ShardedDataParallel as ShardedDDP
from fairscale.optim.oss import OSS
from fairscale.nn.data_parallel import FullyShardedDataParallel as FSDP
import os

TengdaHan

Multi-node-training on slurm with PyTorch

What's this?

• A simple note for how to start multi-node-training on slurm scheduler with PyTorch.
• Useful especially when scheduler is too busy that you cannot get multiple GPUs allocated, or you need more than 4 GPUs for a single job.
• Requirement: Have to use PyTorch DistributedDataParallel(DDP) for this purpose.
• Warning: might need to re-factor your own code.
• Warning: might be secretly condemned by your colleagues because using too many GPUs.

tarlen5

"""
author: Timothy C. Arlen
date: 28 Feb 2018

Calculate Mean Average Precision (mAP) for a set of bounding boxes corresponding to specific
image Ids. Usage:

> python calculate_mean_ap.py

Will display a plot of precision vs recall curves at 10 distinct IoU thresholds as well as output

yoavg

Reinforcement Learning for Language Models

Yoav Goldberg, April 2023.

Why RL?

With the release of the ChatGPT model and followup large language models (LLMs), there was a lot of discussion of the importance of "RLHF training", that is, "reinforcement learning from human feedback". I was puzzled for a while as to why RL (Reinforcement Learning) is better than learning from demonstrations (a.k.a supervised learning) for training language models. Shouldn't learning from demonstrations (or, in language model terminology "instruction fine tuning", learning to immitate human written answers) be sufficient? I came up with a theoretical argument that was somewhat convincing. But I came to realize there is an additional argumment which not only supports the case of RL training, but also requires it, in particular for models like ChatGPT. This additional argument is spelled out in (the first half of) a talk by John Schulman from OpenAI. This post pretty much

schlamar

import os
import sys
import traceback
from functools import wraps
from multiprocessing import Process, Queue


def processify(func):
    '''Decorator to run a function as a process.
    Be sure that every argument and the return value

aaronpolhamus

n02119789 1 kit_fox
n02100735 2 English_setter
n02110185 3 Siberian_husky
n02096294 4 Australian_terrier
n02102040 5 English_springer
n02066245 6 grey_whale
n02509815 7 lesser_panda
n02124075 8 Egyptian_cat
n02417914 9 ibex
n02123394 10 Persian_cat

ctmakro

# if input image is in range 0..1, please first multiply img by 255
# assume image is ndarray of shape [height, width, channels] where channels can be 1, 3 or 4
def imshow(img):
    import cv2
    import IPython
    _,ret = cv2.imencode('.jpg', img) 
    i = IPython.display.Image(data=ret)
    IPython.display.display(i)