This is a long preassignment that involves lots of software installation and testing. Please leave a total of at least 2 hours to complete this preassignment. That may seem like a long time, but once you've done it you'll have a powerful suite of software that you can use through your career at MIT and beyond.
If You Encounter Problems
- If you receive an error message, first try Googling it!
- If you tried that and couldn't solve it, write an email to
[email protected]and[email protected]describing the problem in as much detail as possible, preferably including screenshots.
We are assuming that you have the latest version of R (4.3.1 unless you have a macOS 10.13 laptop, in which case you need 4.2.3) installed. You may need to update your installation if you have an older version.
- Install R: Navigate to
http://cran.wustl.eduand follow the instructions for your operating system. - Download RStudio: Navigate to
https://www.rstudio.com/products/rstudio/download/and download RStudio Desktop with an Open Source License (For macOS 10.13 users, click on previous versions and then go to the installers for the 2022.07.2 version). - Test Your Installation: Open RStudio and type 1+2 into the Console window, and press "Enter." If you see the expected result, you are ready to move on!
In the RStudio console, type
pkgs <- c('tidyverse')
install.packages(pkgs)Once the installation is complete, try to load the package
library(tidyverse)If you encounter any error messages that you are unable to handle, please email us.
Donwload the Pre-Assignment folder from Canvas in 15.003_FA23/Files/Pre-Assignment, it should contain a data folder with wine.csv as well as R_Basics.R.
On the top left corner of R Studio, Click on File -> Open File, and navigate to where you have downloaded the folder to open the R_Basics.R file. You will see a new window open up on the top left corner with some scripts already written for you. Follow along the instructions in the comment and work through the basic syntax.
Please try to complete the steps below before the first day of class. We will only be using Julia and Gurobi on the second day, but we have very limited time in class and we will not be able to help you with installation problems during the teaching time. If you have difficulties with the installations below, please email [email protected] and [email protected] and include as much information as possible so that we can assist you.
Note that you will need to be connected to the MIT network to activate the Gurobi installation, but the other steps can be completed from any location.
Julia is programming language developed at MIT. To install Julia, go to https://julialang.org/downloads/ and download the appropriate version for your operating system. See here for more detailed instructions.
We will assume that everyone has installed the most recent version of Julia (v1.9.2). If you have an older version installed, we recommend that you install the newer version as well.
To confirm that Julia is installed, open a Julia window by clicking on the Julia icon in your applications menu (note: mac users should make sure Julia is copied into their applications folder). You should see a prompt at the bottom of the new window that looks like this:
julia>Type 1+1 after the prompt and press enter/return.
julia> 1+1
2If you see the response above, Julia is working!
JuMP is a Julia package that we will use to create optimization models in class. To install this package, run the following lines in the Julia window:
julia> using Pkg
julia> Pkg.add("JuMP")This might take quite a while to finish, so don’t worry if it looks like nothing is happening in the Julia window. You will know that the process is complete when you see the command prompt (julia>) appear at the bottom of your screen.
To test if the package is installed correctly, run the following commands
julia> using JuMP
julia> m = Model()You should see the output below
A JuMP Model
Feasibility problem with:
Variables: 0
Model mode: AUTOMATIC
CachingOptimizer state: NO_OPTIMIZER
Solver name: No optimizer attached.Jupyter is a free, open-source program that will allow us to write and run Julia code in a web browser (instead of typing everything into the command line). IJulia is a Julia package that allows Julia to communicate with the Jupyter software. Instead of installing Jupyter on its own, we can use the IJulia package to install it within Julia.
Run the following lines in a Julia window:
julia> using Pkg
julia> Pkg.add("IJulia")
julia> using IJuliaThese lines download and install the IJulia package. Now, we will try to open a Jupyter notebook. If Jupyter is not installed, Julia will ask if we want to install it now. Run the following line, and then press enter/return or y to install Jupyter:
julia> notebook()
install Jupyter via Conda, y/n? [y]: If this is successful, a Jupyter tab will open in the default browser on your computer. Click “New” in the top right corner to make a new notebook (if a menu appears, select Julia 1.9.2). A new tab will open with a blank Jupyter notebook.
Note: you must be on the MIT network to activate your academic license. We will leave time at the end of day 1 of orientation for you to complete these steps. If you will not be on campus during orientation, you can use a different solver instead without a license--see notes below.
Gurobi is a commercial optimization solver that we will use to solve optimization problems in class. Here are the basic steps that you will need to follow to install Gurobi,:
- Register for a Gurobi account on the gurobi website. Use your @mit.edu email address, and select the Academic option (not the commercial option).
- Download the Gurobi Optimizer software
hereand install. You might need to log in to the page first, the current stable version is Gurobi 10.0.2. - Create and download an Academic License to use the software
here. - Use the license file to activate the Gurobi software that you installed. Follow the instructions on the license page to run the grbgetkey command. Note that you must be connected to the MIT SECURE network to do this. If you are not on campus, please move on to the next section (IJulia) and come back to this step later.
A summary of the Gurobi installation/activation process is available here and detailed installation instructions are available here. If you get stuck trying to follow these instructions, please email us for assistance.
After installing Gurobi, we need to add a Julia package called "Gurobi" that allows Julia to communicate with the Gurobi software. Run the following lines in your Julia window:
julia> using Pkg
julia> Pkg.add("Gurobi")If you see an error message during this installation, it could be because you did not install/activate Gurobi properly. Please read through the "Installation" information here and see the instructions for setting the GUROBI_HOME environment variable in Julia;
# On Windows, this might be
ENV["GUROBI_HOME"] = "C:\\Program Files\\gurobi1002\\win64"
# ... or perhaps ...
ENV["GUROBI_HOME"] = "C:\\gurobi1002\\win64"
using Pkg
Pkg.add("Gurobi")
Pkg.build("Gurobi")
# On Mac, this might be
ENV["GUROBI_HOME"] = "/Library/gurobi1002/mac64"
using Pkg
Pkg.add("Gurobi")
Pkg.build("Gurobi")Note: check the version of Gurobi that you downloaded. The above instructions assume you downloaded version 10.0.2. If you have a different version, your path may differ (e.g. Gurobi 9.5.2 -> replace gurobi1002 with gurobi952). If this doesn't work, also check which folder you installed Gurobi in, and update the path accordingly if necessary.
If the Gurobi package is successfully installed in Julia, run the following lines, you might see a warning of Academic license - for non-commercial use only - expires 2023-08-11, this is normal:
julia> using JuMP, Gurobi
julia> model = Model(optimizer_with_attributes(Gurobi.Optimizer, "Presolve" => 0, "OutputFlag" => 0))You should see this output:
A JuMP Model
Feasibility problem with:
Variables: 0
Model mode: AUTOMATIC
CachingOptimizer state: EMPTY_OPTIMIZER
Solver name: GurobiIf you are unable to activate your Gurobi license (i.e. if you are not yet on campus), you can use an open-source solver as a temporary solution.
Also install the Cbc package, which will be the backend mixed-integer optimization solver for our optimization problems.
julia> Pkg.add("Cbc")
julia> using JuMP, Cbc
julia> model = Model(optimizer_with_attributes(Cbc.Optimizer, "Presolve" => 0, "OutputFlag" => 0))You should see this output:
A JuMP Model
Feasibility problem with:
Variables: 0
Model mode: AUTOMATIC
CachingOptimizer state: EMPTY_OPTIMIZER
Solver name: COIN Branch-and-Cut (Cbc)Once you have completed all the steps above, copy and paste the following code into a new Jupyter notebook (next to the "In []:" prompt)
using JuMP, Gurobi
model = Model(optimizer_with_attributes(Gurobi.Optimizer, "Presolve" => 0, "OutputFlag" => 0)) # or Cbc.Optimizer
@variable(model,x>=0)
@objective(model, Min, x)
optimize!(model)
print("The answer is ",JuMP.value(x))Now, click the "Run" button to run this code. You should see this output below:
The answer is 0.0If you see this output, everything is working correctly. If you see errors, one of the steps above may be incomplete. If you don't see any output, make sure that you have selected the notebook cell where you paste the code and try to run it again.
How can we manage complex, code-based workflows? How can we reliably share code between collaborators without syncing issues? How can we track multiple versions of scripts without going crazy? There are multiple solutions to these problems, but version control with git is by far the most common.
We will be using the command-line interface to Git. First of all, check if you already have git installed (in which case you can skip this step). Windows users should look for Git Bash, while macOS and Linux users should open a terminal and try running the command: git
If you don't have git installed, go to the Git project page and follow the link on the right side to download the installer for your operating system. Follow the instructions in the README file in the downloaded .zip or .dmg.
Windows: During the installation, select to use Git from the Windows command prompt, checkout Windows-style, commit UNIX-style line endings, and add a shortcut to the Desktop for easy access.
macOS: if you receive an “unidentified developer” warning, right-click the .pkg file, select Open, then click Open.
GitHub is a hosting service for git that makes it easy to share your code.
-
Sign up for an account -- remember to keep track of your username and password. Feel free to enter information about yourself and optionally a profile picture.
-
From the menu at the top right corner of the page, go to Settings, and select SSH and GPG keys.
-
Follow the GitHub instructions to set up SSH:
Check if you already have SSH keys
Generate a SSH key if you don’t have one
If you've made it this far, congratulations! You now possess a powerful set of tools for analyzing data, solving optimization problems, and collaborating on code. You're ready to go!