Issues

• Title: setup.py fails on CPU-only machines due to unhandled PermissionError

  idiap/fast-transformers#134 · Filter by author belassiter opened on Oct 25, 2025
  1

• TypeError: canonicalize_version() got an unexpected keyword argument 'strip_trailing_zero'

  idiap/fast-transformers#132 · Filter by author luispintoc opened on Aug 28, 2024

• Speed of linear attention slower than the attention implemented in pytorch

  idiap/fast-transformers#130 · Filter by author yzeng58 opened on Jun 24, 2024

• [WinError 2] The system cannot find the file specified: build_ext

  idiap/fast-transformers#129 · Filter by author cliffordkleinsr opened on Feb 29, 2024
  2

• ERROR: Could not build wheels for pytorch-fast-transformers, which is required to install pyproject.toml-based projects

  idiap/fast-transformers#128 · Filter by author ouusan opened on Dec 27, 2023
  13

• ImportError

  idiap/fast-transformers#127 · Filter by author PaulaTeeuwen opened on Nov 25, 2023
  1

• Provenance of algorithms

  idiap/fast-transformers#126 · Filter by author taibai123abc opened on Sep 7, 2023

• Error about causal_product_cpu.cpython-38-darwin.so on Mac

  idiap/fast-transformers#124 · Filter by author XiaoqZhang opened on Jul 17, 2023
  2

• Detailed implementation of clustered_sparse_dot_product

  idiap/fast-transformers#120 · Filter by author HanielF opened on Jan 24, 2023

• Understanding how to define key, query and value for the cross attention calculation

  idiap/fast-transformers#119 · Filter by author neuronphysics opened on Dec 18, 2022

• Example for NLP

  idiap/fast-transformers#118 · Filter by author Bachstelze opened on Dec 7, 2022

• Cuda version

  idiap/fast-transformers#117 · Filter by author jiaji-huang opened on Dec 7, 2022
  2

• Can't officially save Linear Attention model

  idiap/fast-transformers#114 · Filter by author maulberto3 opened on Jun 24, 2022
  2

• Any decoder example?

  idiap/fast-transformers#113 · Filter by author ahmedraza1996 opened on Jun 19, 2022
  1

• Runtime error on causal_product_cpu on GCC/G++ 11

  idiap/fast-transformers#110 · Filter by author lsisoft opened on Feb 8, 2022
  3

• how causal mask constructed in training batch model with linear causal attention?

  idiap/fast-transformers#109 · Filter by author Howuhh opened on Nov 26, 2021

• Parallel complexity of Linear Attention is O(N)?

  idiap/fast-transformers#108 · Filter by author haozheji opened on Nov 16, 2021
  1

• causal-linear do not use attn_mask ?

  idiap/fast-transformers#105 · Filter by author davidliujiafeng opened on Oct 14, 2021
  1

• layernorm eps is not copied properly when cloning HF_Bert

  idiap/fast-transformers#103 · Filter by author huu4ontocord opened on Sep 26, 2021
  2

• local_dot_product_cuda fails when queries and keys have different lengths

  idiap/fast-transformers#98 · Filter by author tridao opened on Aug 8, 2021

• Make fast-transformers JIT Compilable

  idiap/fast-transformers#88 · Filter by author AndriyMulyar opened on Apr 28, 2021
  1

• Implementing tensorflow version of CausalDotProduct

  idiap/fast-transformers#73 · Filter by author JamesDeAntonis opened on Mar 16, 2021
  1

• Linear Transformers are Fast Weight Memory Systems

  new-attention
  idiap/fast-transformers#70 · Filter by author angeloskath opened on Mar 10, 2021

• How to install without compiling linear attention?

  idiap/fast-transformers#69 · Filter by author hadaev8 opened on Mar 10, 2021

• Training with fp16 by DeepSpeed

  idiap/fast-transformers#62 · Filter by author e4exp opened on Feb 11, 2021