moskomule/memory_efficient_attention.pytorch
A human-readable PyTorch implementation of "Self-attention Does Not Need O(n^2) Memory" (Rabe&Staats'21).
PythonApache-2.0
A human-readable PyTorch implementation of "Self-attention Does Not Need O(n^2) Memory" (Rabe&Staats'21).
PythonApache-2.0