Add transformer example with RoPE and MoE-like mechanisms#3078

Merged

davisking merged 29 commits into

davisking:masterfrom

Jul 25, 2025

Commits on Apr 28, 2025

Implementation of linear_ layer for neural networks. This layer provides an optimized linear transformation for multi-dimensional inputs.
Cydral
committed

Commits on May 2, 2025

Minor change
Cydral
committed

Commits on May 3, 2025

Update dlib/dnn/layers.h

davisking
and
Copilot
authored

Commits on May 6, 2025

Commits on May 22, 2025

Commits on May 23, 2025

Update dlib/test/dnn.cpp
davisking
authored

Commits on May 30, 2025

Commits on May 31, 2025

Fix batch building
Cydral
committed

Commits on Jun 3, 2025

Slight improvement in model definition.
Cydral
committed

Commits on Jun 7, 2025

Commits on Jun 8, 2025

Commits on Jun 16, 2025

Commits on Jun 17, 2025

Decoding part of the bpe_tokenizer updated.
Cydral
committed

Commits on Jun 27, 2025

Network definition update
Cydral
committed