Isolated environments, extra exercises

Exercise 2: work with `Example-gpu`

Rehearse creating a Python virtual environment from a step-by-step instruction

This exercise if for UPPMAX users only.

In this exercise, we create another environment Example-gpu in the same step-by-step fashion as done in exercise 1:

flowchart TD
load_modules[1.Load modules]
create[2.Create]
activate[3.Activate]
install_libraries[4.Install Python libraries]
check[5.Check installed Python libraries]
use[6.Use]
deactivate[7.Deactivate]

load_modules --> create
create --> activate
activate --> install_libraries
install_libraries --> check
check --> use
use --> deactivate
deactivate --> activate

This virtual environment called Example-gpu is used for examples where the use of GPUs is demonstrated, by using the numba and PyTorch Python packages.

Because the structure is the same as Exercise 1, see exercise 1 for details.

Exercise 2.1: load the modules needed

Here we need an older Python module, as that is what available on the Snowy computer cluster:

module load python/3.9.5

Exercise 2.2: create the virtual environment

python -m venv --system-site-packages /proj/r-py-jl/<user>/python/Example-gpu

where [username] is your UPPMAX username, for example python -m venv --system-site-packages /proj/r-py-jl/sven/python/Example-gpu.

Exercise 2.3: activate the virtual environment

source /proj/r-py-jl/<user>/python/Example-gpu/bin/activate

where [username] is your UPPMAX username, for example source /proj/r-py-jl/sven/python/Example-gpu/bin/activate.

Exercise 2.4: install Python packages

Installing these packages takes around 6 minutes.

pip install --upgrade numpy scipy numba torch

Exercise 2.5: check if the Python packages are installed

pip list

Exercise 2.6: use the virtual environment

Not now :-)

Exercise 2.7: deactivate the virtual environment

deactivate

Exercise 3: export and import a virtual environment

Rehearse creating a Python virtual environment
Export a virtual environment from a step-by-step instruction
Import a virtual environment from a step-by-step instruction

In this exercise, we export and import a virtual environment. Additionally, you get to rehearse to create a virtual environment: as we have done this earlier, the answers will now be hidden :-)

Exercise 3.1: load the modules needed

Load the modules for Python 3.11.x.

module load python/3.11.8

module load GCC/12.3.0 Python/3.11.3

Exercise 3.2: create the virtual environment

Create a virtual environment with the name analysis.

python -m venv --system-site-packages /proj/r-py-jl/[username]/python/analysis

python -m venv --system-site-packages /proj/nobackup/hpc2n2024-025/[username]/python/analysis

Exercise 3.3: activate the virtual environment

Activate the virtual environment.

source /proj/r-py-jl/[username]/python/analysis/bin/activate

source /proj/nobackup/hpc2n2024-025/[username]/python/analysis/bin/activate

Exercise 3.4: install Python packages

Installing these packages takes around 13 minutes.

Create a file called requirements.txt, with the following content:

numpy==1.22.3
matplotlib==3.5.2
pandas==1.4.2

Use any text editor, for example, nano.

Install packages by using the requirements.txt file:

pip install -r requirements.txt

Yup, just do that:

pip install -r requirements.txt

Yup, just do that:

pip install -r requirements.txt

Exercise 3.5: check if the Python packages are installed

Check that the packages were installed.

pip list

pip list

Exercise 3.6: use the virtual environment

Export the Python packages our virtual environment uses:

pip freeze > requirements.txt

View the file requirements.txt.

There are many ways to view a file, for example:

cat requirements.txt