Pytorch implementation of paper "Efficient Nearest Neighbor Language Models" (EMNLP 2021)

Overview

Efficient Nearest Neighbor Language Models

This is implementation of the paper:

Efficient Nearest Neighbor Language Models
Junxian He, Graham Neubig, Taylor Berg-Kirkpatrick
EMNLP 2021

This repo implements several techniques to speed up the evaluation of non-parametric, nearest neighbor language models. Specifically, we improve the efficiency along three axes: adaptive retrieval, datastore prunning, and dimension reduction.

Install Dependencies

This repository is largly based on the knnlm repo which is a fork of Fairseq (commit da544b). Please use the exact commit page to determine software requirements for using this code.

git clone [email protected]:jxhe/efficient-knnlm.git

cd efficient-knnlm
pip install --editable .
pip install faiss

Hardware

Experiments for this paper were conducted on machines that contain 32 CPUs, 100GB of RAM, and one NVIDIA 3090 24GB GPU. Saving the Wikitext-103 datastore requires 200GB of disk space. Note that the number of CPUs has a great impact on the speed.

Running Efficient kNNLM

Preparation

Data

We share Fairseq's instructions on how to prepare the data here.

mkdir -p datasets/wikitext-103
cp examples/language_model/wikitext-103/prepare-wikitext-103.sh datasets/wikitext-103

cd datasets/wikitext-103
bash prepare-wikitext-103.sh
cd ../..

TEXT=datasets/wikitext-103
python preprocess.py \
    --only-source \
    --trainpref $TEXT/wiki.train.tokens \
    --validpref $TEXT/wiki.valid.tokens \
    --testpref $TEXT/wiki.test.tokens \
    --destdir data-bin/wikitext-103 \
    --workers 20

Download the language model checkpoint pretrained on WikiText-103

# the model checkpoint link is from the knnlm repo
wget https://nlp.stanford.edu/projects/knnlm/wt103_checkpoint_best.pt -P knnlm_ckpt

Save the datastore

mkdir -p dstore

python eval_lm.py data-bin/wikitext-103 \
    --path knnlm_ckpt/checkpoint_best.pt \
    --sample-break-mode none --max-tokens 3072 \
    --softmax-batch 1024 --gen-subset train \
    --context-window 1536 --tokens-per-sample 1536 \
    --dstore-mmap dstore/dstore --knn-keytype 'last_ffn_input' \
    --dstore-size 103225485 --model-overrides "{'knn_keytype': 'last_ffn_input'}" \
    --save-knnlm-dstore --fp16 --dstore-fp16

Dimension Reduction

# the script applies PCA of dimension 512 by default 
# the PCA hyperparameter can be tuned in this script
# set pca=0 to revert back to the vanilla version
bash ef_knnlm/build_faiss.sh

The faiss index is saved into dstore. Try it out:

bash ef_knnlm/utils_cmd/eval_knnlm.sh \
    -d wikitext-103 \
    -s valid \
    -p dstore/dstore_size103225485_embed1024_fp16 \
    -i dstore/knn.103225485.pca512.m64.index \
    -n 103225485 \

You should already observe a speedup.

Adaptive Retrieval

prepare heldout data to train the retrieval adaptor

# this randomly selects 90% of validation data as the training data to 
# train the retrieval adaptor
bash ef_knnlm/adaptive_retrieval/prepare_heldout.sh wikitext-103

prepare features

bash ef_knnlm/adaptive_retrieval/prepare_feature_pipeline.sh

train

bash ef_knnlm/adaptive_retrieval/train_ar.sh

It saves the retrieval adaptor checkpoints into checkpoint/wikitext-103-valid

evaluation

# the cutoff ratio in adaptive retrieval
# by default we cut off half of the retrieval
cutoff=50

# please change this to the .pt file path observed from the last step
ar_ckpt=xxx

# this hyperparameter needs to be changed if 
# the datastore sizes change (e.g. datastore pruning)
size=103225485

dstore_prefix=dstore/dstore_size${size}_embed1024_fp16
index_file=dstore/knn.${size}.pca512.m64.index

bash ef_knnlm/utils_cmd/eval_knnlm.sh \
    -d wikitext-103 \
    -s test \
    -p ${dstore_prefix} \
    -i ${index_file} \
    -c knnlm_ckpt/wt103_checkpoint_best.pt \
    -n ${size} \
    -f datasets/wikitext-103 \
    -a ctxt,freq,lm_ent,lm_max,fert \
    -u ${cutoff} \
    -h ${ar_ckpt} \
    # -w "True"

Datastore Pruning

precompute all the retrieval results for every record in the datastore:

# It is possible to parallel this operation by change 
# "--start-point" and "--num" arguments so that the training
# data would be splitted into multiple smaller ones. In this case
# the retrieval results would be saved into multiple files
bash ef_knnlm/dstore_compression/save_retrieval_results.sh

The retrieval results are saved into dstore/greedy_merge, other datastore pruning algorithms may be played around using these pre-computed results.

greedy merging

# perform greedy merging to yield a new smaller datastore, 
# and build faiss index from the new datastore
bash ef_knnlm/dstore_compression/merge_compression.sh

The pruned datastore and index are saved into dstore/greedy_merging, replace the previousdstore_prefix/index_file with the new ones to use the pruned the datastore. The option -w "True"needs to be passed to eval_knnlm.sh to read the generated datastore weights file from greedy merging.

Reference

@inproceedings{he2021eff,
title={Efficient Nearest Neighbor Language Models},
author={Junxian He and Graham Neubig and Taylor Berg-Kirkpatrick},
booktitle={Proceedings of EMNLP},
year={2021}
}
Owner
Junxian He
NLP/ML PhD student at CMU
Junxian He
Implementations for the ICLR-2021 paper: SEED: Self-supervised Distillation For Visual Representation.

Implementations for the ICLR-2021 paper: SEED: Self-supervised Distillation For Visual Representation.

Jacob 27 Oct 23, 2022
Datasets for new state-of-the-art challenge in disentanglement learning

High resolution disentanglement datasets This repository contains the Falcor3D and Isaac3D datasets, which present a state-of-the-art challenge for co

NVIDIA Research Projects 37 May 26, 2022
Domain Generalization for Mammography Detection via Multi-style and Multi-view Contrastive Learning

MSVCL_MICCAI2021 Installation Please follow the instruction in pytorch-CycleGAN-and-pix2pix to install. Example Usage An example of vendor-styles tran

Jaron Lee 11 Oct 19, 2022
PiRank: Learning to Rank via Differentiable Sorting

PiRank: Learning to Rank via Differentiable Sorting This repository provides a reference implementation for learning PiRank-based models as described

54 Dec 17, 2022
A real-time motion capture system that estimates poses and global translations using only 6 inertial measurement units

TransPose Code for our SIGGRAPH 2021 paper "TransPose: Real-time 3D Human Translation and Pose Estimation with Six Inertial Sensors". This repository

Xinyu Yi 261 Dec 31, 2022
Small repo describing how to use Hugging Face's Wav2Vec2 with PyCTCDecode

🤗 Transformers Wav2Vec2 + PyCTCDecode Introduction This repo shows how 🤗 Transformers can be used in combination with kensho-technologies's PyCTCDec

Patrick von Platen 102 Oct 22, 2022
Pytorch implementation of our paper under review -- 1xN Pattern for Pruning Convolutional Neural Networks

1xN Pattern for Pruning Convolutional Neural Networks (paper) . This is Pytorch re-implementation of "1xN Pattern for Pruning Convolutional Neural Net

Mingbao Lin (林明宝) 29 Nov 29, 2022
All the code and files related to the MI-Lab of UE19CS305 course in sem 5

Machine-Intelligence-Lab-CS305 The compilation of all the code an drelated files from MI-Lab UE19CS305 (of batch 2019-2023) offered by PES University

Arvind Krishna 3 Nov 10, 2022
Contrastive Learning of Structured World Models

Contrastive Learning of Structured World Models This repository contains the official PyTorch implementation of: Contrastive Learning of Structured Wo

Thomas Kipf 371 Jan 06, 2023
Implementation of "Distribution Alignment: A Unified Framework for Long-tail Visual Recognition"(CVPR 2021)

Implementation of "Distribution Alignment: A Unified Framework for Long-tail Visual Recognition"(CVPR 2021)

105 Nov 07, 2022
Implementation of NÜWA, state of the art attention network for text to video synthesis, in Pytorch

NÜWA - Pytorch (wip) Implementation of NÜWA, state of the art attention network for text to video synthesis, in Pytorch. This repository will be popul

Phil Wang 463 Dec 28, 2022
The easiest tool for extracting radiomics features and training ML models on them.

Simple pipeline for experimenting with radiomics features Installation git clone https://github.com/piotrekwoznicki/ClassyRadiomics.git cd classrad pi

Piotr Woźnicki 17 Aug 04, 2022
PyTorch implementation of the ExORL: Exploratory Data for Offline Reinforcement Learning

ExORL: Exploratory Data for Offline Reinforcement Learning This is an original PyTorch implementation of the ExORL framework from Don't Change the Alg

Denis Yarats 52 Jan 01, 2023
Confident Semantic Ranking Loss for Part Parsing

Confident Semantic Ranking Loss for Part Parsing

Jiachen Xu 5 Oct 22, 2022
Repository for MeshTalk supplemental material and code once the (already approved) 16 GHS captures our lab will make publicly available are released.

meshtalk This repository contains code to run MeshTalk for face animation from audio. If you use MeshTalk, please cite @inproceedings{richard2021mesht

Meta Research 221 Jan 06, 2023
G-NIA model from "Single Node Injection Attack against Graph Neural Networks" (CIKM 2021)

Single Node Injection Attack against Graph Neural Networks This repository is our Pytorch implementation of our paper: Single Node Injection Attack ag

Shuchang Tao 18 Nov 21, 2022
DrNAS: Dirichlet Neural Architecture Search

This paper proposes a novel differentiable architecture search method by formulating it into a distribution learning problem. We treat the continuously relaxed architecture mixing weight as random va

Xiangning Chen 37 Jan 03, 2023
DeepMind Alchemy task environment: a meta-reinforcement learning benchmark

The DeepMind Alchemy environment is a meta-reinforcement learning benchmark that presents tasks sampled from a task distribution with deep underlying structure.

DeepMind 188 Dec 25, 2022
An All-MLP solution for Vision, from Google AI

MLP Mixer - Pytorch An All-MLP solution for Vision, from Google AI, in Pytorch. No convolutions nor attention needed! Yannic Kilcher video Install $ p

Phil Wang 784 Jan 06, 2023
GeDML is an easy-to-use generalized deep metric learning library

GeDML is an easy-to-use generalized deep metric learning library

Borui Zhang 32 Dec 05, 2022