Frequently asked questions
Get ready, find your pace, and keep learning
Start here for practical questions about the workshop. For the lesson itself, open the handbook.
Before the workshop
Who is this workshop for?
People who know basic linear algebra and want to use tensors in machine learning. You do not need previous tensor theory: new terms are introduced as you go. You will read and edit Python code in the notebooks.
What should I do beforehand?
Follow the prerequisites on the homepage: review the required linear algebra reading, complete the pre-work notebooks, and run notebook 00. Report setup or download failures before the session so they can be addressed early.
How long does the workshop take?
The live agenda lasts 210 minutes, including quizzes and breaks. Preparation happens beforehand; optional exercises and the convolution and deconvolution deep dive are take-home material. See the schedule for the session timings.
Which language is used?
The workshop is taught in English, with questions welcome in English or Spanish. The site includes both languages. Notebooks are primarily in English, with a short Spanish summary under each heading. The Spanish handbook is a machine translation; the English handbook remains the source of truth. The Kahoot how-to exists in both languages; the live questions on kahoot.it stay in English.
Running the notebooks
Do I need to install anything on my computer?
You can work in Google Colab through the links on the notebooks page. Each notebook has a Setup section that loads its own data and installs any additional packages it needs. Run Setup before the exercises. For a local Jupyter environment, use the local setup instructions.
Do the notebooks share data or variables?
No. Each notebook is self-contained, so you can open a section without running the earlier notebooks. Run that notebook’s Setup and the preparation cells for your chosen activity. Running notebook 00 checks your setup and downloads; it does not populate another notebook’s runtime.
Can I work offline?
Some notebooks use datasets bundled with the Python libraries; others download data. For local offline work, install the required packages in advance and check the dependency and network table for the notebook you want to use. Test it without a connection before relying on it during a session.
What if a cell fails or a download does not work?
Check the first error and confirm that Setup and the required preparation cells have run in the current runtime. If you restarted the runtime, run them again. For a download failure, check your connection and the notebook’s network requirements. During the workshop, share the notebook number, the cell, and the error message with the facilitator in Discord.
If code runs but a result seems wrong, compare it with the notebook’s expected result and use the worked mistakes to practice checking shapes, axes, and assumptions.
Learning during and after the session
Do I need to finish every exercise during the workshop?
No. Follow the notebook’s Core prep cells and Core activity first. Use Predict → Run → Explain → Check: predict the result, run the code, explain what happened, and check your understanding. Save optional exercises for later if you need more time.
Where are the solutions?
Exercises have folded solution cells in the notebooks, and the handbook contains worked solutions. Try the exercise before opening the answer. The notebooks are saved without outputs, so running the cells produces the results you compare with the explanation.
Can I use the material for self-study?
Yes. Start with the prerequisites, then read the handbook alongside the corresponding notebooks. Pause to solve exercises and use the entry and exit checks to assess your reasoning. The references provide further reading, and the take-home deep dive extends the workshop.
How should I use the companion resources?
The companion offers summaries, audio, videos, and self-checks for review. Its NotebookLM material is machine-generated and may contain errors. Check explanations against the handbook and your notebook results.
Where should instructors start?
Open Teach this workshop for the facilitator guide, group tasks, assessments, worked mistakes, and feedback form. It explains how to fit those activities into the existing agenda.