Dounm

# !/usr/bin/env python3

import torch
from torch import distributed as dist

dist.init_process_group(backend="nccl")
torch.cuda.set_device(f"cuda:{dist.get_rank()}")

if dist.get_rank() == 0:
    print("I am rank 0")

Dounm

# Inspired by https://github.com/vpenso/ganglia-sensors/blob/master/lib/python_modules/infiniband.py#/

import logging
import re
import sys
import json
import time
import subprocess

Dounm

#include <stdio.h>

#define NUM_STREAMS 4
cudaError_t memcpyUsingStreams (float           *fDest,
                                float           *fSrc,
                                int             iBytes,
                                cudaMemcpyKind  eDirection)
{
    cudaStream_t    *pCuStream = NULL;
    int             iIndex = 0 ;

Dounm

#!/usr/bin/env bash

clang_format=clang-format
autopep=autopep8


for FILE in $(git diff --cached --name-only)
do
  if [[ $FILE == *.py ]]; then
    if ! command -v $autopep &> /dev/null; then

Dounm

// t.h
#include <iostream>

using namespace std;

template <typename T1, typename T2> 
class A { 
  public:
  void foo();
};