Complete list

The complete list of packages currently available on TLDR is below.

• Create folder "dir"

mkdir("dir")

• Reads a line from input

a = readline()

CSV

A package built to be a fast and flexible pure-Julia library for handling delimited text files.

• Read a CSV input.

CSV.File(source)

• Write table to file.

CSV.write(file, table)

• Read a csv input iterating over its rows. Only the current row values are buffered.

for row in CSV.Rows(source)
# do something
end

Crayons

Tools to write strings in different colors and styles to terminals

• Print a string using Crayons for formatting

print(Crayon(foreground = :red, bold = true, underline = true), "Bold red text with underline")

• Create a new Crayon with red foreground color and bold text formatting

r_fg = Crayon(foreground = :red, bold = true)

• Merge two crayons with different properties in this case the new crayon will have red foreground color and green background color

merged = merge(Crayon(foreground:red), Crayon(background: green))

DataFrames

Tools for working with tabular data.

• Get the column names

names(df)

• Get the column names as Symbols

propertynames(df)

• Look at the last X rows of a data frame

last(df, X)

• Look at the firs X rows of a data frame

first(df, X)

• Add a column named A to the the DataFrame df

df.A = 1:8

• Return a new data frame df2 that is a copy of df

df2 = copy(df)

• Print all rows and/or columns of the DataFrame df

show(df, allrows=true, allcols=true)

• Create a new DataFrame by passing the column headers and contents

df = DataFrame(A = 1:3, B = [:odd, :even, :odd])

• Replace "None" values by zero in a single column col_1 of dataframe df

replace!(df.col_1, "None" => 0)

• Return a data frame with some elementary statistics and information about each column

describe(df)

• Add a new row as a tuple or vector, where the order of elements matches that of the columns of df

push!(df, (1,"M",...))

DifferentialEquations

Package for numerically solving differential equations.

• Define a discrete time evolution problem.

prob = DiscreteProblem(f, u0, tspan)

• Define an Ordinary Differential Equation problem.

prob = ODEProblem(eom, u0, tspan)

• Define a Differential Algebraic Equation problem.

prob = DAEProblem(eom, u0, du0, tspan)

• Numerically solve a differential equation problem.

sol = solve(prob)

• Define a Stochastic Differential Equation problem.

prob = SDEProblem(f, g, u0, tspan)

• Define a Random Ordinary Differential Equation problem.

prob = RODEProblem(eom, u0, tspan)

• Define a callback that is applied when the condition function is true.

cb = DiscreteCallback(condition, affect!)

• Define a callback that is applied when the continuous condition function hits zero.

cb = ContinuousCallback(condition, affect!)

Distributions

A Julia package for probability distributions and associated functions.

• Create a Normal distribution with mean 0 and standard deviation 1.

d = Normal()

• Create a Normal distribution with mean μ and standard deviation 1.

d = Normal(μ)

• Create a Normal distribution with mean μ and standard deviation σ.

d = Normal(μ, σ)

Flux

The Julia Machine Learning Library

• Create a Dense layer with a sigmoid activation function

layer = Dense(10, 5, σ)

• Chain different layers to form a deep neural network model

model = Chain(Dense(10, 5, σ), Dense(5,3), softmax)

• Create a Gradient Descent optimiser with learning rate 0.1

opt = Descent(0.1)

• Extract the parameters of a model m which can be later passed on to the gradient function

ps = Flux.params(m)

• Train the model. For each datapoint in data, compute the gradient of loss with respect to params through backpropagation and call the optimizer opt. An optional callback is given with the keyword argument cb

train!(loss, params, data, opt; cb)

JSON

• Convert dict to JSON

JSON.json(dict)

Krylov

Hand-picked Krylov methods for linear systems and least squares problems

• Solve linear system Ax = b using the Conjugate gradient method

x, stats = cg(A, b)

• Solve the least-squares problem min ½‖Ax - b‖² using the Conjugate gradient method

x, stats = cgls(A, b)

• Minimize the quadratic function f(x) = ½xᵀAx - bᵀx subject to the trust region ‖x‖ ≤ Δ using the Steihaug-Toint Conjugate gradient variant

x, stats = cg(A, b, radius=Δ)

LinearAlgebra

• Solve a linear system Ax = b using the Cholesky factorization

F = cholesky(A)
F \ b

Plots

Plotting API and toolset

• Update an existing plot

plot!(x, y)

• Save an existing plot to a file

savefig("plot.png")

• Create a basic Boxplot on a vector of numbers

boxplot(["Series 1"], y)

• Create a basic Violin plot on a vector of numbers

violin(["Series 1"], y)

• Create a basic line plot on two vectors of numbers

plot(x, y)

• Create a basic scatter plot on two vectors of numbers

scatter(x, y)

Statistics

Standard library module for basic statistics functionality.

• Compute the median of all elements in a collection itr.

median(itr)

• Compute the median of all elements of a vector v, overwriting the input vector.

median!(v)

• Compute the quantile(s) of a collection itr at a specified probability or vector or tuple of probabilities p on the interval [0,1].

q = quantile(itr, p)

• Compute the quantile(s) of a vector v at a specified probability or vector or tuple of probabilities p on the interval [0,1], overwriting v

q = quantile!(v, p)

TLDR

A package for fast help and snippets.

• Enter tldr> mode.

}

• Search for commands and packages related to the keyword.

tldr"keyword"

Zygote

Automatic Differentiation in Julia

• Computes the gradient of f at each argument, returning a tuple.

gradient(f, args...)