This guide is for setting up your own machine with the software needed to work through the weekly and mandatory exercises in IN3050/IN4050 Spring semester 2023.
Setup with Anaconda
Download and install either Anaconda or Miniconda
Install Anaconda by following this guide. To get started with Anaconda, see this page. Once you got a working shell with conda installed you can continue the guide.
You should now have access to the conda command. For a quick test, execute the conda command without any arguments.
conda
You are recommended to download the conda cheat sheet and keep it at hand.
Create environment
Download the file in3050_2023.yml you find here.
Go to the directory where you downloaded the file.
To create an environment with the name "in3050", and required dependencies, one command is sufficient
conda env create --file in3050_2023.yml
If you want to change the name of the environment, you can edit the name in the in3050_2023.yml file before executing the command.
Activate the environment
To activate the environment
conda activate in3050
If you want to get back to you standard python and packages:
conda deactivate
In windows you can choose environment from the start menu.
Problems with Mac M1-processor
There are some problems getting TensorFlow to work with the Mac M1-processor. If you experience problems with creating the environment on your Mac, you may remove the last two lines in the .yml file.
Added April 4: We have made a recipe for how you can install tensorflow on a mac here.