Understanding einsum for Deep learning: implement a transformer with multi-head self-attention from scratch

Learn about the einsum notation and einops by coding a custom multi-head self-attention unit and a transformer block