# Task

Implement `get_positional_encoding` — a function that precomputes the full positional encoding matrix for a given sequence length and embedding dimension.

The function should:

1. Accept `seq_len` and `d_model` as arguments;
2. Return a tensor of shape `(seq_len, d_model)` where each row is the encoding for one position;
3. Apply `sin` to even indices and `cos` to odd indices using the sinusoidal formula from the previous chapter.

Run it with `seq_len=10` and `d_model=16`, print the resulting tensor, and inspect how the values change across positions and dimensions. Then create a random embedding tensor of shape `(10, 16)` and add the positional encoding to it to confirm the shapes are compatible.



Master the transformer architecture by building its core components from scratch. Explore self-attention, multi-head attention, positional encoding, encoder-decoder structure, and layer normalization, all within a consistent domain. Finish by assembling a complete transformer and tackling a capstone challenge.

Build a deep understanding of the transformer architecture by constructing each component from scratch, culminating in a capstone challenge.

Challenge: Implementing Positional Encoding

Task