## Learn to Use ITensor

main / getting_started / running C++v3 | C++v2 | Julia

# Running ITensor Codes

After installing the ITensor library the next step is creating and running codes. We will look at a minimal Julia code based on ITensor and various ways to run it.

## Basic Code Outline

The basic outline of a code which uses the ITensor library is as follows

using ITensors

let
# ... your own code goes here ...
# For example:
i = Index(2,"i")
j = Index(3,"j")
T = randomITensor(i,j)
@show T
end


The reason we recommend the let...end block is that global scope in Julia can have some surprising behaviors, at least through Julia version 1.4. Putting your code into a let block avoids these issues.

## Running a Script

Now say you put the above code into a file named code.jl. Then you can run this code on the command line as follows

\$ julia code.jl


This script-like mode of running Julia is convenient for running longer jobs, such as on a cluster.

## Running Interactively

However, sometimes you want to do rapid development when first writing and testing a code. For this kind of work, the long startup and compilation times currently incurred by the 1.x versions of Julia can be a nuisance. Fortunately there is a nice solution: repeatedly load your code into a running Julia session.

To set up this kind of session, take the following steps:

1. Enter the interactive mode of Julia, by inputting the command julia on the command line. You will now be in the Julia "REPL" (read-eval-print loop) with the prompt julia> on the left of your screen.
2. To run a code such as the code.jl file discussed above, input the command

 julia> include("code.jl")


Note that you must be in the same folder as code.jl for this to work; otherwise input the entire path to the code.jl file. The code will run and you will see its output in the REPL.

3. Now say you want to modify and re-run the code. To do this, just edit the file in another terminal window or editor program, without closing your Julia session. Now run the command

 julia> include("code.jl")


again and your updated code will run, but this time skipping any of the precompilation overhead incurred on previous steps.

The above steps to running a code interactively has a big advantage that you only have to pay the startup time of compiling ITensor and other libraries you are using once. Further changes to your code only incur very small extra compilation times, facilitating rapid development.

## Other Ways of Running ITensor

The ITensors.jl library should be runnable through any type of setting that runs Julia code, such as Jupyter notebooks. If you encounter a mode of running ITensor that doesn't work, please contact us by emailing support -at- itensor.org.

## Why Are There Pauses?

You may often notice that your Julia code will run very quickly through one section, then take a long time to run a certain function. For example, while running an ITensor DMRG calculation, the first sweep is often reported as taking 10 or more seconds while the second and third sweeps take only a second. The reason for this is that Julia is a just in time (JIT) compiled language. So instead of compiling at the beginning like a C++ code, the compilation happens at the very last minute when you call a function for the first time. As of Julia version 1.4, the JIT time for Julia can be quite long, but it is expected to get much better with future releases of Julia. Also we are working on various precompilation strategies with ITensors.jl to make it so that more code only has to be compiled once, until you upgrade ITensors.jl again.

For strategies to avoid the JIT compilation overhead and reduced code pauses, see this page on compilation.

Back to Getting Started
Back to Main