Merge branch 'main' into dependabot/pip/ruff-0.7.4

.gitignore

@@ -159,3 +159,7 @@ repo_data.csv
 student_repos/
 reviews.csv
 bert_sentiment_model.pt
+prediction*.json
+
+# vscode
+.vscode/

README.md

@@ -37,7 +37,7 @@ git clone https://github.com/SkafteNicki/dtu_mlops
 If you do not have git installed (yet) we will touch upon it in the course. The folder will contain all the exercise
 material for this course and lectures. Additionally, you should join our
 [Slack channel](https://join.slack.com/t/dtumlops/shared_invite/zt-29szus81f-JKANkZqhalPjKqKRPPY6xA)
-which we use for communication. The link may be expired, write to [me](mailto:nsde@dtu.com).
+which we use for communication. The link may be expired, write to [me](mailto:nsde@dtu.dk).
 
 ## 📂 Course organization
 

pages/faq.md

@@ -71,7 +71,7 @@ of December.
 We can give a grade on the Danish
 [7-point grading](https://ufm.dk/en/education/the-danish-education-system/grading-system) scale for foreign students who
 need it, where their home university does not accept pass/no-pass. You need to contact the course responsible
-[Nicki](mailto:nsde@dtu.com) within the first week of the course to request this. Secondly, we may need to further
+[Nicki](mailto:nsde@dtu.dk) within the first week of the course to request this. Secondly, we may need to further
 validate your work, so please be prepared for doing a short oral exam on one of the last days of the course.
 
 ## I am a EuroTEQ student, any special rules for me ❔

s10_extra/high_performance_clusters.md

@@ -21,7 +21,6 @@ Tier, the larger applications it is possible to run.
 
 <figure markdown>
 ![Image](../figures/hpc_tiers.png){ width="800" }
-<figcaption> <a href="https://gl.deic.dk/en/Supercomputing/EuroCC/HPC-Landscapes"> Image credit </a> </figcaption>
 </figure>
 
 ## Cluster architectures

s1_development_environment/editor.md

@@ -27,7 +27,7 @@ Below, you see an overview of the VS Code interface
 
 <figure markdown>
 ![Image](../figures/vscode.PNG){ width="700" }
-<figcaption> <a href=" https://www.git-tower.com/learn/cheat-sheets/vscode/"> Image credit </a> </figcaption>
+<figcaption> <a href="https://www.git-tower.com/learn/cheat-sheets/vscode/"> Image credit </a> </figcaption>
 </figure>
 
 The main components of VS Code are:

s3_reproducibility/docker.md

@@ -61,7 +61,7 @@ The whole point of using Docker is that sharing applications becomes much easier
 
 <figure markdown>
 ![Image](../figures/docker_share.png){ width="1000" }
-<figcaption> <a href="https://www.ravirajag.dev/blog/mlops-docker"> Image credit </a> </figcaption>
+<figcaption> <a href="https://deep-learning-blogs.vercel.app/blog/mlops-docker"> Image credit </a> </figcaption>
 </figure>
 
 ## ❔ Exercises

s4_debugging_and_logging/boilerplate.md

@@ -320,8 +320,8 @@ framework to do some of the heavy lifting you need to have gone through some of
 10. (Optional) As default PyTorch uses `float32` for representing floating point numbers. However, research have shown
     that neural network training is very robust towards a decrease in precision. The great benefit going from `float32`
     to `float16` is that we get approximately half the
-    [memory consumption](https://www.khronos.org/opengl/wiki/Small_Float_Formats). Try out half-precision training in
-    PyTorch lightning. You can enable this by setting the
+    [memory consumption](https://en.wikipedia.org/wiki/Half-precision_floating-point_format). Try out half-precision
+    training in PyTorch lightning. You can enable this by setting the
     [precision](https://lightning.ai/docs/pytorch/latest/common/trainer.html#precision) flag in the `Trainer`.
 
     ??? success "Solution"

s5_continuous_integration/github_actions.md

@@ -39,7 +39,7 @@ Let's take a look at how a GitHub workflow file is organized:
 <figure markdown>
 ![Image](../figures/actions.png){ width="1000" }
 <figcaption>
-<a href=" https://madewithml.com/courses/mlops/cicd/#github-actions"> Image credit </a>
+<a href="https://madewithml.com/courses/mlops/cicd/#github-actions"> Image credit </a>
 </figcaption>
 </figure>