A python library for highly configurable transformers - easing model architecture search and experimentation.

Related tags

Deep Learningdeep-learningtransformerspytorchartificial-intelligence
Overview

configaformers (re-factor in progress)

A python library for highly configurable transformers - easing model architecture search and experimentation. It is premised on building small and independent modules that enables users to configure custom transformer architectures.

Special thanks to lucidrains (https://github.com/lucidrains) and Kharr.

Usage

Quick demo that will configure a 768-wide, 12-layer transformer, with a language modeling head.

Import, and create token embedding block:

import torch
from model_builder import ConfigaFormer

emb = []
model_dim = 768

emb.append({'type': 'embedding',
            'output_dim': model_dim,
            'num_classes': 50257})

Create self-attention module:

attn = []

# Make residual and norm
attn.append({'type': 'make_stream', 'output_name': 'residual'})
attn.append({'type': 'norm', 'norm_type': 'layer_norm'})

# Make QKVs
attn.append({'type': 'linear', 'output_name': 'queries'})
attn.append({'type': 'linear', 'output_name': 'keys'})
attn.append({'type': 'linear', 'output_name': 'values'})

attn.append({'type': 'make_heads', 'input_name': 'queries', 'output_name': 'queries', 'num_heads': 12})
attn.append({'type': 'make_heads', 'input_name': 'keys', 'output_name': 'keys', 'num_heads': 12})

attn.append({'type': 'rope', 'input_name': 'queries', 'output_name': 'queries', 'rotate_dim': 16})
attn.append({'type': 'rope', 'input_name': 'keys', 'output_name': 'keys', 'rotate_dim': 16})

# Perform attention
attn.append({'type': 'mha_dots',
             'input_name_queries': 'queries',
             'input_name_keys': 'keys'})
attn.append({'type': 'attention_offset'})
attn.append({'type': 'mha_sum',
             'input_name_values': 'values'})

# Mix
attn.append({'type': 'linear'})

# Add residual
attn.append({'type': 'merge_streams',
             'input_name_1': 'residual',
             'merge_type': 'add'})

Create FFN module:

ffn = []

# Make residual and norm
ffn.append({'type': 'make_stream', 'output_name': 'residual'})
ffn.append({'type': 'norm', 'norm_type': 'layer_norm'})

# Proj Up
ffn.append({'type': 'linear', 'output_dim': 768*4})

# Activation
ffn.append({'type': 'activation'})

# Proj Down
ffn.append({'type': 'linear', 'output_dim': 768})

# Add residual
ffn.append({'type': 'merge_streams',
             'input_name_1': 'residual',
             'merge_type': 'add'})

Create language modeling head:

to_logits = []
to_logits.append({'type': 'linear', 'output_dim': 50257})

Create blocks, initialize input shapes, and init the model:

transformer_block = attn + ffn
classifier = ffn + to_logits

blocks = [{"config": emb,
           "repeat": 1},
          {"config": transformer_block,
           "repeat": 12},
          {"config": classifier,
           "repeat": 1},
          ]
          
my_config = {'blocks' = blocks}
input_streams = {'emb_ids': ['B', 'L_in'],
                 'attn_offset': ['B', 12, 'L_in', 'L_in'],}

model = ConfigaFormer(model_config=my_config,
                     input_streams=input_streams).cuda()

This will print out the transformer config:

Block #1, 1x
embedding -> Input(s): emb_ids (BSZ, L_in) - Output(s): x (BSZ, L_in, 768)


Block #2, 12x
make_stream -> Input(s): x (BSZ, L_in, 768) - Output(s): residual (BSZ, L_in, 768)
norm -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): queries (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): keys (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): values (BSZ, L_in, 768)
make_heads -> Input(s): queries (BSZ, L_in, 768) - Output(s): queries (BSZ, 12, L_in, 64)
make_heads -> Input(s): keys (BSZ, L_in, 768) - Output(s): keys (BSZ, 12, L_in, 64)
rope -> Input(s): queries (BSZ, 12, L_in, 64), rope_16 (2048, 16) - Output(s): queries (BSZ, 12, L_in, 64)
rope -> Input(s): keys (BSZ, 12, L_in, 64), rope_16 (2048, 16) - Output(s): keys (BSZ, 12, L_in, 64)
mha_dots -> Input(s): queries (BSZ, 12, L_in, 64), keys (BSZ, 12, L_in, 64) - Output(s): attn_dots (BSZ, 12, L_in, L_in)
attention_offset -> Input(s): attn_dots (BSZ, 12, L_in, L_in), attn_offset (BSZ, 12, L_in, L_in) - Output(s): attn_dots (BSZ, 12, L_in, L_in)
mha_sum -> Input(s): values (BSZ, L_in, 768), attn_dots (BSZ, 12, L_in, L_in) - Output(s): x (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)
merge_streams -> Input(s): residual (BSZ, L_in, 768), x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)
make_stream -> Input(s): x (BSZ, L_in, 768) - Output(s): residual (BSZ, L_in, 768)
norm -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 3072)
activation -> Input(s): x (BSZ, L_in, 3072) - Output(s): x (BSZ, L_in, 3072)
linear -> Input(s): x (BSZ, L_in, 3072) - Output(s): x (BSZ, L_in, 768)
merge_streams -> Input(s): residual (BSZ, L_in, 768), x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)


Block #3, 1x
make_stream -> Input(s): x (BSZ, L_in, 768) - Output(s): residual (BSZ, L_in, 768)
norm -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 3072)
activation -> Input(s): x (BSZ, L_in, 3072) - Output(s): x (BSZ, L_in, 3072)
linear -> Input(s): x (BSZ, L_in, 3072) - Output(s): x (BSZ, L_in, 768)
merge_streams -> Input(s): residual (BSZ, L_in, 768), x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 768)
linear -> Input(s): x (BSZ, L_in, 768) - Output(s): x (BSZ, L_in, 50257)

Before running, we need to get the attention offset (in this case, AliBi with a causal mask):

from attention_offset_module import get_alibi

attn_offset = get_alibi(num_heads=12)

Now we can use the model:

input_data = {'emb_ids': batch_ids.view(bsz, 1024).cuda(),
              'attn_offset': attn_offset.cuda()}

logits = model(input_data)['x'].view(bsz, 1024, 50257)

TODO

  1. Token shifting, down/up sampling
  2. Create higher abstractions for FFN and self-attention
  3. everything else
Owner
Anthony Fuller
Anthony Fuller
GitHub Repository
Piotr - IoT firmware emulation instrumentation for training and research

Piotr: Pythonic IoT exploitation and Research Introduction to Piotr Piotr is an emulation helper for Qemu that provides a convenient way to create, sh

Damien Cauquil 51 Nov 09, 2022
A toolset for creating Qualtrics-based IAT experiments

Qualtrics IAT Tool A web app for generating the Implicit Association Test (IAT) running on Qualtrics Online Web App The app is hosted by Streamlit, a

0 Feb 12, 2022
LSTC: Boosting Atomic Action Detection with Long-Short-Term Context

LSTC: Boosting Atomic Action Detection with Long-Short-Term Context This Repository contains the code on AVA of our ACM MM 2021 paper: LSTC: Boosting

Tencent YouTu Research 9 Oct 11, 2022
An implementation of Video Frame Interpolation via Adaptive Separable Convolution using PyTorch

This work has now been superseded by: https://github.com/sniklaus/revisiting-sepconv sepconv-slomo This is a reference implementation of Video Frame I

Simon Niklaus 984 Dec 16, 2022
Smart edu-autobooking - Johnson @ DMI-UNICT study room self-booking system

smart_edu-autobooking Sistema di autoprenotazione per l'aula studio [email protected]

Davide Carnemolla 17 Jun 20, 2022
An essential implementation of BYOL in PyTorch + PyTorch Lightning

Essential BYOL A simple and complete implementation of Bootstrap your own latent: A new approach to self-supervised Learning in PyTorch + PyTorch Ligh

Enrico Fini 48 Sep 27, 2022
A simple baseline for the 2022 IEEE GRSS Data Fusion Contest (DFC2022)

DFC2022 Baseline A simple baseline for the 2022 IEEE GRSS Data Fusion Contest (DFC2022) This repository uses TorchGeo, PyTorch Lightning, and Segmenta

isaac 24 Nov 28, 2022
Code for the head detector (HeadHunter) proposed in our CVPR 2021 paper Tracking Pedestrian Heads in Dense Crowd.

Head Detector Code for the head detector (HeadHunter) proposed in our CVPR 2021 paper Tracking Pedestrian Heads in Dense Crowd. The head_detection mod

Ramana Sundararaman 76 Dec 06, 2022
MassiveSumm: a very large-scale, very multilingual, news summarisation dataset

MassiveSumm: a very large-scale, very multilingual, news summarisation dataset This repository contains links to data and code to fetch and reproduce

Daniel Varab 19 Dec 16, 2022
Head and Neck Tumour Segmentation and Prediction of Patient Survival Project

Head-and-Neck-Tumour-Segmentation-and-Prediction-of-Patient-Survival Welcome to the Head and Neck Tumour Segmentation and Prediction of Patient Surviv

5 Oct 20, 2022
Clockwork Convnets for Video Semantic Segmentation

Clockwork Convnets for Video Semantic Segmentation This is the reference implementation of arxiv:1608.03609: Clockwork Convnets for Video Semantic Seg

Evan Shelhamer 141 Nov 21, 2022
Code for Piggyback: Adapting a Single Network to Multiple Tasks by Learning to Mask Weights

Piggyback: https://arxiv.org/abs/1801.06519 Pretrained masks and backbones are available here: https://uofi.box.com/s/c5kixsvtrghu9yj51yb1oe853ltdfz4q

Arun Mallya 165 Nov 22, 2022
Official Implementation of Domain-Aware Universal Style Transfer

Domain Aware Universal Style Transfer Official Pytorch Implementation of 'Domain Aware Universal Style Transfer' (ICCV 2021) Domain Aware Universal St

KibeomHong 80 Dec 30, 2022
Huawei Hackathon 2021 - Sweden (Stockholm)

huawei-hackathon-2021 Contributors DrakeAxelrod Challenge Requirements: python=3.8.10 Standard libraries (no importing) Important factors: Data depend

Drake Axelrod 32 Nov 08, 2022
Code for our ICASSP 2021 paper: SA-Net: Shuffle Attention for Deep Convolutional Neural Networks

SA-Net: Shuffle Attention for Deep Convolutional Neural Networks (paper) By Qing-Long Zhang and Yu-Bin Yang [State Key Laboratory for Novel Software T

Qing-Long Zhang 199 Jan 08, 2023
The Fundamental Clustering Problems Suite (FCPS) summaries 54 state-of-the-art clustering algorithms, common cluster challenges and estimations of the number of clusters as well as the testing for cluster tendency.

FCPS Fundamental Clustering Problems Suite The package provides over sixty state-of-the-art clustering algorithms for unsupervised machine learning pu

9 Nov 27, 2022
Image De-raining Using a Conditional Generative Adversarial Network

Image De-raining Using a Conditional Generative Adversarial Network [Paper Link] [Project Page] He Zhang, Vishwanath Sindagi, Vishal M. Patel In this

He Zhang 216 Dec 18, 2022
This is the winning solution of the Endocv-2021 grand challange.

Endocv2021-winner [Paper] This is the winning solution of the Endocv-2021 grand challange. Dependencies pytorch # tested with 1.7 and 1.8 torchvision

Vajira Thambawita 14 Dec 03, 2022
A new video text spotting framework with Transformer

TransVTSpotter: End-to-end Video Text Spotter with Transformer Introduction A Multilingual, Open World Video Text Dataset and End-to-end Video Text Sp

weijiawu 67 Jan 03, 2023
PyTorch implementation of CVPR'18 - Perturbative Neural Networks

This is an attempt to reproduce results in Perturbative Neural Networks paper. See original repo for details.

Michael Klachko 57 May 14, 2021