Intro to Deep Learning review
Kaggle · kaggle
A short, practical first look at neural networks in Keras, run entirely in Kaggle notebooks.
What does it take?
| Difficulty | 3 out of 5 |
|---|---|
| Hours | Not checked yet |
| Format | Code notebooks, Interactive exercises |
| Maths needed | School-level maths |
| Coding needed | Yes, you will write code |
| Hardware needed | Not checked yet |
| Certificate | Free certificate |
| Free to audit | Yes |
| Pace | Self-paced |
| Last updated by the provider | Not checked yet |
Who should take it, and who should skip it?
Take it if you are
- People who finished Intro to Machine Learning
- Anyone who wants to try deep learning before committing
Skip it if you are
- Anyone who wants PyTorch
- People who want the theory behind backpropagation
What will you be able to do after it?
- Build and train a small neural network
- Recognise overfitting and underfitting
- Know what dropout and batch normalisation are for
What's in each module?
| Module | What it covers |
|---|---|
| A single neuron | What a unit in a network actually computes. |
| Deep networks and training | Stacking layers and fitting them with gradient descent. |
| Overfitting and regularisation | Keeping a model from memorising the training data. |
What we liked, and what we didn't
What we liked
- Quick to finish
- Free GPU notebooks
- Clear exercises
What we didn't
- Uses Keras, while most research uses PyTorch
- Shallow by design
What should you take after it?
- Practical Deep Learning for Coders
Practical Deep Learning for Coders has 9 lessons of about 90 minutes, costs nothing and has no certificate. You deploy your own model by the end of lesson 2.
- Neural Networks: Zero to Hero
Neural Networks: Zero to Hero is 8 free YouTube videos, about 14.5 hours, where Andrej Karpathy builds a GPT from scratch in code. No certificate.
How we rated it: 7.6 out of 10
| Criterion | Score out of 10 |
|---|---|
| Clarity | 8 |
| Hands-on practice | 9 |
| Up to date | 6 |
| Depth | 5 |
| Truly free | 10 |
So should you take it?
A good taster once you've done Kaggle's Intro to ML. It won't make you a deep learning engineer, but you'll train a real network in an afternoon and know whether you want more.
Where to go next
- Kaggle Learn: our platform review
What else it offers, and what costs money.
- Kaggle Intro to Machine Learning
Kaggle's 7-lesson first machine learning course takes about 3 hours, costs nothing and ends with a competition. Kaggle gives a free completion certificate.
- Machine Learning Crash Course
Google's Machine Learning Crash Course has 12 modules and takes about 13.5 hours by Google's own timings. It is free, with quiz badges and no certificate.
- AI for Beginners
A 24-lesson open curriculum on GitHub covering classic AI, ML and a little deep learning.
Last checked .
Not sure this is the one?
Answer two questions and we'll point you to the course that fits where you are today.