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

Related tags

Deep Learning deep-learning transformers pytorch artificial-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
You might also like...
"NAS-Bench-301 and the Case for Surrogate Benchmarks for Neural Architecture Search".

NAS-Bench-301 This repository containts code for the paper: "NAS-Bench-301 and the Case for Surrogate Benchmarks for Neural Architecture Search". The

AutoML-Freiburg-Hannover 57 Nov 30, 2022
PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and Multi-Step Knowledge Distillation
PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and Multi-Step Knowledge Distillation

PocketNet This is the official repository of the paper: PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and M

Fadi Boutros 40 Dec 22, 2022
Differentiable architecture search for convolutional and recurrent networks
Differentiable architecture search for convolutional and recurrent networks

Differentiable Architecture Search Code accompanying the paper DARTS: Differentiable Architecture Search Hanxiao Liu, Karen Simonyan, Yiming Yang. arX

Hanxiao Liu 3.7k Jan 9, 2023
DeepHyper: Scalable Asynchronous Neural Architecture and Hyperparameter Search for Deep Neural Networks
DeepHyper: Scalable Asynchronous Neural Architecture and Hyperparameter Search for Deep Neural Networks

What is DeepHyper? DeepHyper is a software package that uses learning, optimization, and parallel computing to automate the design and development of

DeepHyper Team 214 Jan 8, 2023
[ICLR 2021] "Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective" by Wuyang Chen, Xinyu Gong, Zhangyang Wang

Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective [PDF] Wuyang Chen, Xinyu Gong, Zhangyang Wang In ICLR 2

VITA 156 Nov 28, 2022
[ICLR 2021] HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark
[ICLR 2021] HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark

HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark Accepted as a spotlight paper at ICLR 2021. Table of content File structure Prerequi

null 72 Jan 3, 2023
Deep Multimodal Neural Architecture Search
Deep Multimodal Neural Architecture Search

MMNas: Deep Multimodal Neural Architecture Search This repository corresponds to the PyTorch implementation of the MMnas for visual question answering

Vision and Language Group@ MIL 23 Dec 21, 2022
Official implementation of Rethinking Graph Neural Architecture Search from Message-passing (CVPR2021)
Official implementation of Rethinking Graph Neural Architecture Search from Message-passing (CVPR2021)

Rethinking Graph Neural Architecture Search from Message-passing Intro The GNAS can automatically learn better architecture with the optimal depth of

Shaofei Cai 48 Sep 30, 2022
Block-wisely Supervised Neural Architecture Search with Knowledge Distillation (CVPR 2020)
Block-wisely Supervised Neural Architecture Search with Knowledge Distillation (CVPR 2020)

DNA This repository provides the code of our paper: Blockwisely Supervised Neural Architecture Search with Knowledge Distillation. Illustration of DNA

Changlin Li 215 Dec 19, 2022
Owner
Anthony Fuller
Anthony Fuller
GitHub
Lightweight, Python library for fast and reproducible experimentation :microscope:

Steppy What is Steppy? Steppy is a lightweight, open-source, Python 3 library for fast and reproducible experimentation. Steppy lets data scientist fo

minerva.ml 134 Jul 10, 2022
BossNAS: Exploring Hybrid CNN-transformers with Block-wisely Self-supervised Neural Architecture Search

BossNAS This repository contains PyTorch evaluation code, retraining code and pretrained models of our paper: BossNAS: Exploring Hybrid CNN-transforme

Changlin Li 127 Dec 26, 2022
Densely Connected Search Space for More Flexible Neural Architecture Search (CVPR2020)

DenseNAS The code of the CVPR2020 paper Densely Connected Search Space for More Flexible Neural Architecture Search. Neural architecture search (NAS)

Jamin Fong 291 Nov 18, 2022
piSTAR Lab is a modular platform built to make AI experimentation accessible and fun. (pistar.ai)

piSTAR Lab WARNING: This is an early release. Overview piSTAR Lab is a modular deep reinforcement learning platform built to make AI experimentation a

piSTAR Lab 0 Aug 1, 2022
Deep Image Search is an AI-based image search engine that includes deep transfor learning features Extraction and tree-based vectorized search.

Deep Image Search - AI-Based Image Search Engine Deep Image Search is an AI-based image search engine that includes deep transfer learning features Ex

null 139 Jan 1, 2023
Naszilla is a Python library for neural architecture search (NAS)

A repository to compare many popular NAS algorithms seamlessly across three popular benchmarks (NASBench 101, 201, and 301). You can implement your ow

null 270 Jan 3, 2023
Arch-Net: Model Distillation for Architecture Agnostic Model Deployment

Arch-Net: Model Distillation for Architecture Agnostic Model Deployment The official implementation of Arch-Net: Model Distillation for Architecture A

MEGVII Research 22 Jan 5, 2023
TorchPQ is a python library for Approximate Nearest Neighbor Search (ANNS) and Maximum Inner Product Search (MIPS) on GPU using Product Quantization (PQ) algorithm.

Efficient implementations of Product Quantization and its variants using Pytorch and CUDA

null 146 Dec 28, 2022
Fake-user-agent-traffic-geneator - Python CLI Tool to generate fake traffic against URLs with configurable user-agents

Fake traffic generator for Gartner Demo Generate fake traffic to URLs with custo

New Relic Experimental 3 Oct 31, 2022
FADNet++: Real-Time and Accurate Disparity Estimation with Configurable Networks

FADNet++: Real-Time and Accurate Disparity Estimation with Configurable Networks

HKBU High Performance Machine Learning Lab 6 Nov 18, 2022