Explore extreme compression for pre-trained language models

Last update: Nov 14, 2022

Related tags

Deep Learning Xcompression

Overview

Explore extreme compression for pre-trained language models

Code for paper "Exploring extreme parameter compression for pre-trained language models ICLR2022"

Before Training

install some libraries

 pip install tensorly==0.5.0

Torch is needed, torch 1.0-1.4 is preferred

Install horovod for distributed learning

Configuration Install horovod on GPU

pip install horovod[pytorch]

loading pre-trained models

wget https://huggingface.co/bert-base-uncased/resolve/main/pytorch_model.bin -P  models/bert-base-uncased
wget https://huggingface.co/bert-base-uncased/resolve/main/vocab.txt -P  models/bert-base-uncased
cp models/bert-base-uncased/pytorch_model.bin models/bert-td-72-384/pytorch_model.bin 
cp models/bert-base-uncased/vocab.txt models/bert-td-72-384/vocab.txt

generate training data for given corpora (e.g., saved in the path "corpora" )

python pregenerate_training_data.py --train_corpus ${CORPUS_RAW} \ 
                  --bert_model ${BERT_BASE_DIR}$ \
                  --reduce_memory --do_lower_case \
                  --epochs_to_generate 3 \
                  --output_dir ${CORPUS_JSON_DIR}$

task data augmentation

python data_augmentation.py --pretrained_bert_model ${BERT_BASE_DIR}$ \
                            --glove_embs ${GLOVE_EMB}$ \
                            --glue_dir ${GLUE_DIR}$ \  
                            --task_name ${TASK_NAME}$

Decomposing BERT

decomposition and general distillation

Run with horovod

mpirun -np 8 -bind-to none -map-by slot -x NCCL_DEBUG=INFO -x LD_LIBRARY_PATH -x PATH -mca pml ob1 -mca btl ^openib python3 general_distill.py --teacher_model models/bert-base-uncased --student_model models/bert-gd-72-384 --pregenerated_data data/pregenerated_data --num_train_epochs 2.0 --train_batch_size 32 --output_dir output/bert-gd-72-384 -use_swap --do_lower_case

To restrict sharing among SAN or FFN, add "ops" and set "ops" to be "san" or "ffn" in bert-gd-72-384/config.json

ops = "san"

Evaluation

Task distillation with data augmentation in fine-tuning phase

Rename a pretrained model as "", for instance, change step_0_pytorch_model.bin to pytorch_model.bin, and change load_compressed_model from false to true in output/config.json

Task distillation for distributed training

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 task_distill.py --teacher_model models/bert-base-uncasedi/STS-B --student_model models/bert-gd-72-384 --task_name STS-B --aug_train --data_dir data/glue_data/SST-2 --max_seq_length 128 --train_batch_size 32 --aug_train --learning_rate 2e-5 --num_train_epochs 3.0 --output_dir ./output/36-256-STS-B

Task distillation for single gpu

python3  task_distill.py  --teacher_model models/bert-base-uncased   --student_model  models/bert-td-72-384  --output output_demo  --data_dir  data/glue_data/SST-2   --task_name  SST-2  --do_lower_case --aug_train

For augmentation, you should add --aug_train

Get test result for model

python run_glue.py --model_name_or_path  models/bert-td-72-384/SST-2 --task_name SST-2 --do_eval --do_predict --data_dir data/glue_data/STS-B --max_seq_length 128 --save_steps 500 --save_total_limit 2 --output_dir ./output/SST-2

Explore extreme compression for pre-trained language models

Related tags

Overview

Explore extreme compression for pre-trained language models

Before Training

install some libraries

loading pre-trained models

generate training data for given corpora (e.g., saved in the path "corpora" )

task data augmentation

Decomposing BERT

decomposition and general distillation

Evaluation

Task distillation with data augmentation in fine-tuning phase

Owner

twinkle

Official PyTorch implementation of RIO

Code for "Modeling Indirect Illumination for Inverse Rendering", CVPR 2022

Official Pytorch implementation of ICLR 2018 paper Deep Learning for Physical Processes: Integrating Prior Scientific Knowledge.

SegNet-Basic with Keras

Unofficial implementation of Fast-SCNN: Fast Semantic Segmentation Network

Implementation for our ICCV 2021 paper: Dual-Camera Super-Resolution with Aligned Attention Modules

A library for implementing Decentralized Graph Neural Network algorithms.

Deduplicating Training Data Makes Language Models Better

A python library for self-supervised learning on images.

Official repository for the paper "Going Beyond Linear Transformers with Recurrent Fast Weight Programmers"

mmdetection version of TinyBenchmark.

Keras like implementation of Deep Learning architectures from scratch using numpy.

Pytorch implementation of "MOSNet: Deep Learning based Objective Assessment for Voice Conversion"

VD-BERT: A Unified Vision and Dialog Transformer with BERT

hySLAM is a hybrid SLAM/SfM system designed for mapping

A Python-based development platform for automated trading systems - from backtesting to optimisation to livetrading.

Controlling Hill Climb Racing with Hand Tacking

An Implicit Function Theorem (IFT) optimizer for bi-level optimizations

[ICCV 2021] FaPN: Feature-aligned Pyramid Network for Dense Image Prediction

Self-training for Few-shot Transfer Across Extreme Task Differences