tcapelle

Check fiest that a upstream repo is setup here. $ git remote -v Basic commands to sync/merge git with origin.

$ git checkout master
> Switched to branch 'master'

$ git merge upstream/master
> Updating a422352..5fdff0f

tcapelle

tcapelle

Quick install :

1. Installer Radiance et ajouter les binaires au Path de windows. Faire un test simple, ouvrir une terminal (je conseil cmder) et lance la commande:

>>rvu
rvu: fatal - missing octree argument

L'output doit etre rvu: fatal - missing octree argument, ca veut dire que il trouve rvu sur ton path.

2. Créer un environment sur conda pour travailler sur Radiance

tcapelle

tcapelle

Basic Linux Setup

This gist aims to setup a LInux machine for python work from scratch.

ZSH

Other terminal app that is plain better.

• theme

TMUX

• start new with session name: tmux new -s myname

tcapelle

tcapelle

def _get_col_idxs(df, cols):
    "return cols index to perform iloc"
    return [df.columns.get_loc(c) for c in L(cols) if c in df]

def _iloc(df, rows, cols=None):
    "Iloc that supports col names"
    if isinstance(cols, (tuple, list, str)):
        cols = _get_col_idxs(df, cols)
        return df.iloc[rows, cols]
    return df.iloc[rows, slice(None)]

tcapelle

from torch import nn
from functools import partial
from einops.layers.torch import Rearrange, Reduce

class PreNormResidual(nn.Module):
    def __init__(self, dim, fn):
        super().__init__()
        self.fn = fn
        self.norm = nn.LayerNorm(dim)

tcapelle

Setup NGC Docker

you will need to setup docker with nvidia runtime, containers

• Install docker and nvidia-docker:

sudo apt-get install -y docker nvidia-container-toolkit

tcapelle