Choose a learner path and build your knowledge.
Read the notebook, run the standalone version in Colab, and keep track of what you have completed on this device.
Learning from data
Build the mathematical and practical habits used across machine learning.
0 / 12 complete
- Stage 1
Set up your workbench
Start with arrays and prepare data for a model.
- Stage 2
Fit your first models
Use regression to model an outcome and classify an observation.
- Stage 3
Learn how models learn
Follow optimization and diagnose model error.
- Stage 4
Find structure without labels
Reduce dimensions and group related observations.
Neural networks by hand
Understand how neural networks represent data and learn from error.
0 / 10 complete
- Stage 1
Build a neural layer
Learn the parts of a feed-forward network.
- Stage 2
Train with gradients
Trace an error backwards through the network.
- Stage 3
Make training stable
Use optimization and normalization techniques.
- Stage 4
Build deeper networks
Use residual connections and recognize handwritten digits.
Language models end to end
Prepare text, represent words, and build transformer-based language systems.
0 / 13 complete
- Stage 1
Prepare text
Normalize text and split it into model inputs.
- Stage 2
Represent words
Learn word vectors and local language models.
- Stage 3
Model sequences
Carry information through an ordered sequence.
- Stage 4
Build with attention
Move from position information to transformer applications.
Explore individual topics
Find a notebook and explore how its prerequisites connect.
Select a topic to explore
Choose a notebook from the list to see its overview, prerequisites, and completion status.