progressbars.jl's Introduction

ProgressBars.jl (formerly Tqdm.jl)

A fast, extensible progress bar for Julia. This is a Julia clone of the great Python package tqdm.

Installation

Run the following in a julia prompt:

using Pkg
Pkg.add("ProgressBars")

Usage

julia> using ProgressBars

julia> for i in ProgressBar(1:100000) #wrap any iterator
          #code
       end
100.00%┣████████████████████████████████████████████████▉┫ 100000/100000 [00:12<00:00 , 8616.43 it/s]

There is a tqdm alias, so that people coming from python will feel right at home :)

julia> using ProgressBars

julia> for i in tqdm(1:100000) #wrap any iterator
          #code
       end
100.00%┣████████████████████████████████████████████████▉┫ 100000/100000 [00:12<00:00 , 8616.43 it/s]

Or with a set description (e.g. for loss values when training neural networks)

julia> iter = ProgressBar(1:100)
       for i in iter
          # ... Neural Network Training Code
          loss = exp(-i)
          set_description(iter, string(@sprintf("Loss: %.2f", loss)))
       end
Loss: 0.02 3.00%┣█▌                                                  ┫ 3/100 00:00<00:02, 64.27 it/s]

Postfixes are also possible, if that's your kind of thing:

julia> iter = ProgressBar(1:100)
       for i in iter
          # ... Neural Network Training Code
          loss = exp(-i)
          set_postfix(iter, Loss=@sprintf("%.2f", loss))
       end
100.0%┣████████████████████████████████████████████┫ 1000/1000 [00:02<00:00, 420.4 it/s, Loss: 0.37]

Now with added support for Threads.@threads for:

julia> a = []
       Threads.@threads for i in ProgressBar(1:1000)
         push!(a, i * 2)
       end
100.00%┣█████████████████████████████████████████████████████▉┫ 1000/1000 00:00<00:00, 28753.50 it/s]

Recommend Projects