Attention for PyTorch with Linear Memory Footprint

Unofficially implements https://arxiv.org/abs/2112.05682 to get Linear Memory Cost on Attention (+ some sidekick speedup on the GPU when compared to reference implementation in JAX)

Usage:

git clone https://github.com/CHARM-Tx/linear_mem_attention_pytorch
cd linear_mem_attention_pytorch
python setup.py install

Usage:

High Level

from linear_mem_attention_torch.fast_attn import Attention

batch, length, features = 2, 2**8, 64
x, ctx = torch.randn(2, batch, length, features)
mask = torch.randn(batch, length) < 1.

attn = Attention(dim=features, heads = 8, dim_head = 64, bias=False)

# self-attn
v_self = attn(x, x, mask, query_chunk_size=1024, key_chunk_size=4096)

# cross-attn
v_cross = attn(x, ctx, mask, query_chunk_size=1024, key_chunk_size=4096)

Low level

from linear_mem_attention_torch import attention

batch, length, heads, features = 2, 2**8, 8, 64
mask = torch.randn(batch, length) < 1.
q, k, v = torch.randn(3, batch, length, heads, features)

v_ = attention(q, k, v, mask, query_chunk_size=1024, key_chunk_size=4096)

Benchmarks

See examples/example_benchamrk.ipynb for more information.

Citations:

@misc{rabe2021selfattention,
      title={Self-attention Does Not Need $O(n^2)$ Memory}, 
      author={Markus N. Rabe and Charles Staats},
      year={2021},
      eprint={2112.05682},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}

Attention for PyTorch with Linear Memory Footprint

Related tags

Overview

Attention for PyTorch with Linear Memory Footprint

Usage:

Usage:

High Level

Low level

Benchmarks

Citations:

Owner

Hierarchical Time Series Forecasting with a familiar API

This is a Python Module For Encryption, Hashing And Other stuff

Employee-Managment - Company employee registration software in the face recognition system

Customer Segmentation using RFM

Reinforcement Learning for the Blackjack

Learning to Communicate with Deep Multi-Agent Reinforcement Learning in PyTorch

Code for C2-Matching (CVPR2021). Paper: Robust Reference-based Super-Resolution via C2-Matching.

A framework for multi-step probabilistic time-series/demand forecasting models

Demo project for real time anomaly detection using kafka and python

A platform to display the carbon neutralization information for researchers, decision-makers, and other participants in the community.

The code repository for EMNLP 2021 paper "Vision Guided Generative Pre-trained Language Models for Multimodal Abstractive Summarization".

Github Traffic Insights as Prometheus metrics.

Deeprl - Standard DQN and dueling network for simple games

A simple pytorch pipeline for semantic segmentation.

DeconvNet : Learning Deconvolution Network for Semantic Segmentation

Voice of Pajlada with model and weights.

Conservative Q Learning for Offline Reinforcement Reinforcement Learning in JAX

Python inverse kinematics for your robot model based on Pinocchio.

Multiple-criteria decision-making (MCDM) with Electre, Promethee, Weighted Sum and Pareto

Make Watson Assistant send messages to your Discord Server