duskvirkus

How to debug a docker container by entering the shell because I always forget

If not already a service add keep alive to the Dockerfile

ENTRYPOINT ["tail", "-f", "/dev/null"]

Then to enter shell do the following

duskvirkus

#!/bin/bash
# Installs s5cmd

if command -v s5cmd &> /dev/null
then
    echo "s5cmd is already installed"
    exit 0
fi

START_DIR=$(pwd)

duskvirkus

import concurrent.futures

from tqdm import tqdm

import random
import time

def run_on_thread_example(packed_args):
    arg_one, arg_two = packed_args

duskvirkus

import concurrent.futures
from tqdm import tqdm

def run_multiprocess(func, iterator):
    with concurrent.futures.ProcessPoolExecutor() as executor:
        results = list(tqdm(executor.map(func, iterator), total=len(iterator)))
    return results

duskvirkus

Table t;

int rows = 8;
int cols = 11;

float rowSize;
float colSize;

ArrayList<String> toDisplay = new ArrayList<String>();

duskvirkus

cmake_minimum_required(VERSION 3.0)

project(cxxtorch)

find_package(Torch REQUIRED)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${TORCH_CXX_FLAGS}")

add_executable(cxxtorch main.cxx)
target_link_libraries(cxxtorch "${TORCH_LIBRARIES}")
set_property(TARGET cxxtorch PROPERTY CXX_STANDARD 14)

duskvirkus

Local ML Setup

Minimal ubuntu server install to machine learning development setup mostly thanks to lambda.

Currently using ubuntu server 20.04.

By no means is this a definitive guide more a list for me to remember what I like to do I definitly recommend looking into each thing individually for up to date instructions on installation.

Gnome3

duskvirkus

Simple Python Profiler

in code:

call configure_profiling

use @profile_decorator on functions you want to time

don't forget to call

duskvirkus

duskvirkus

absl-py
alabaster
albumentations
altair
appdirs
argon2-cffi
arviz
astor
astropy
astunparse