ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Related tags

Deep Learningalbert
Overview

ALBERT

***************New March 28, 2020 ***************

Add a colab tutorial to run fine-tuning for GLUE datasets.

***************New January 7, 2020 ***************

v2 TF-Hub models should be working now with TF 1.15, as we removed the native Einsum op from the graph. See updated TF-Hub links below.

***************New December 30, 2019 ***************

Chinese models are released. We would like to thank CLUE team for providing the training data.

Version 2 of ALBERT models is released.

In this version, we apply 'no dropout', 'additional training data' and 'long training time' strategies to all models. We train ALBERT-base for 10M steps and other models for 3M steps.

The result comparison to the v1 models is as followings:

Average SQuAD1.1 SQuAD2.0 MNLI SST-2 RACE
V2
ALBERT-base 82.3 90.2/83.2 82.1/79.3 84.6 92.9 66.8
ALBERT-large 85.7 91.8/85.2 84.9/81.8 86.5 94.9 75.2
ALBERT-xlarge 87.9 92.9/86.4 87.9/84.1 87.9 95.4 80.7
ALBERT-xxlarge 90.9 94.6/89.1 89.8/86.9 90.6 96.8 86.8
V1
ALBERT-base 80.1 89.3/82.3 80.0/77.1 81.6 90.3 64.0
ALBERT-large 82.4 90.6/83.9 82.3/79.4 83.5 91.7 68.5
ALBERT-xlarge 85.5 92.5/86.1 86.1/83.1 86.4 92.4 74.8
ALBERT-xxlarge 91.0 94.8/89.3 90.2/87.4 90.8 96.9 86.5

The comparison shows that for ALBERT-base, ALBERT-large, and ALBERT-xlarge, v2 is much better than v1, indicating the importance of applying the above three strategies. On average, ALBERT-xxlarge is slightly worse than the v1, because of the following two reasons: 1) Training additional 1.5 M steps (the only difference between these two models is training for 1.5M steps and 3M steps) did not lead to significant performance improvement. 2) For v1, we did a little bit hyperparameter search among the parameters sets given by BERT, Roberta, and XLnet. For v2, we simply adopt the parameters from v1 except for RACE, where we use a learning rate of 1e-5 and 0 ALBERT DR (dropout rate for ALBERT in finetuning). The original (v1) RACE hyperparameter will cause model divergence for v2 models. Given that the downstream tasks are sensitive to the fine-tuning hyperparameters, we should be careful about so called slight improvements.

ALBERT is "A Lite" version of BERT, a popular unsupervised language representation learning algorithm. ALBERT uses parameter-reduction techniques that allow for large-scale configurations, overcome previous memory limitations, and achieve better behavior with respect to model degradation.

For a technical description of the algorithm, see our paper:

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut

Release Notes

  • Initial release: 10/9/2019

Results

Performance of ALBERT on GLUE benchmark results using a single-model setup on dev:

Models MNLI QNLI QQP RTE SST MRPC CoLA STS
BERT-large 86.6 92.3 91.3 70.4 93.2 88.0 60.6 90.0
XLNet-large 89.8 93.9 91.8 83.8 95.6 89.2 63.6 91.8
RoBERTa-large 90.2 94.7 92.2 86.6 96.4 90.9 68.0 92.4
ALBERT (1M) 90.4 95.2 92.0 88.1 96.8 90.2 68.7 92.7
ALBERT (1.5M) 90.8 95.3 92.2 89.2 96.9 90.9 71.4 93.0

Performance of ALBERT-xxl on SQuaD and RACE benchmarks using a single-model setup:

Models SQuAD1.1 dev SQuAD2.0 dev SQuAD2.0 test RACE test (Middle/High)
BERT-large 90.9/84.1 81.8/79.0 89.1/86.3 72.0 (76.6/70.1)
XLNet 94.5/89.0 88.8/86.1 89.1/86.3 81.8 (85.5/80.2)
RoBERTa 94.6/88.9 89.4/86.5 89.8/86.8 83.2 (86.5/81.3)
UPM - - 89.9/87.2 -
XLNet + SG-Net Verifier++ - - 90.1/87.2 -
ALBERT (1M) 94.8/89.2 89.9/87.2 - 86.0 (88.2/85.1)
ALBERT (1.5M) 94.8/89.3 90.2/87.4 90.9/88.1 86.5 (89.0/85.5)

Pre-trained Models

TF-Hub modules are available:

Example usage of the TF-Hub module in code:

tags = set()
if is_training:
  tags.add("train")
albert_module = hub.Module("https://tfhub.dev/google/albert_base/1", tags=tags,
                           trainable=True)
albert_inputs = dict(
    input_ids=input_ids,
    input_mask=input_mask,
    segment_ids=segment_ids)
albert_outputs = albert_module(
    inputs=albert_inputs,
    signature="tokens",
    as_dict=True)

# If you want to use the token-level output, use
# albert_outputs["sequence_output"] instead.
output_layer = albert_outputs["pooled_output"]

Most of the fine-tuning scripts in this repository support TF-hub modules via the --albert_hub_module_handle flag.

Pre-training Instructions

To pretrain ALBERT, use run_pretraining.py:

pip install -r albert/requirements.txt
python -m albert.run_pretraining \
    --input_file=... \
    --output_dir=... \
    --init_checkpoint=... \
    --albert_config_file=... \
    --do_train \
    --do_eval \
    --train_batch_size=4096 \
    --eval_batch_size=64 \
    --max_seq_length=512 \
    --max_predictions_per_seq=20 \
    --optimizer='lamb' \
    --learning_rate=.00176 \
    --num_train_steps=125000 \
    --num_warmup_steps=3125 \
    --save_checkpoints_steps=5000

Fine-tuning on GLUE

To fine-tune and evaluate a pretrained ALBERT on GLUE, please see the convenience script run_glue.sh.

Lower-level use cases may want to use the run_classifier.py script directly. The run_classifier.py script is used both for fine-tuning and evaluation of ALBERT on individual GLUE benchmark tasks, such as MNLI:

pip install -r albert/requirements.txt
python -m albert.run_classifier \
  --data_dir=... \
  --output_dir=... \
  --init_checkpoint=... \
  --albert_config_file=... \
  --spm_model_file=... \
  --do_train \
  --do_eval \
  --do_predict \
  --do_lower_case \
  --max_seq_length=128 \
  --optimizer=adamw \
  --task_name=MNLI \
  --warmup_step=1000 \
  --learning_rate=3e-5 \
  --train_step=10000 \
  --save_checkpoints_steps=100 \
  --train_batch_size=128

Good default flag values for each GLUE task can be found in run_glue.sh.

You can fine-tune the model starting from TF-Hub modules instead of raw checkpoints by setting e.g. --albert_hub_module_handle=https://tfhub.dev/google/albert_base/1 instead of --init_checkpoint.

You can find the spm_model_file in the tar files or under the assets folder of the tf-hub module. The name of the model file is "30k-clean.model".

After evaluation, the script should report some output like this:

***** Eval results *****
  global_step = ...
  loss = ...
  masked_lm_accuracy = ...
  masked_lm_loss = ...
  sentence_order_accuracy = ...
  sentence_order_loss = ...

Fine-tuning on SQuAD

To fine-tune and evaluate a pretrained model on SQuAD v1, use the run_squad_v1.py script:

pip install -r albert/requirements.txt
python -m albert.run_squad_v1 \
  --albert_config_file=... \
  --output_dir=... \
  --train_file=... \
  --predict_file=... \
  --train_feature_file=... \
  --predict_feature_file=... \
  --predict_feature_left_file=... \
  --init_checkpoint=... \
  --spm_model_file=... \
  --do_lower_case \
  --max_seq_length=384 \
  --doc_stride=128 \
  --max_query_length=64 \
  --do_train=true \
  --do_predict=true \
  --train_batch_size=48 \
  --predict_batch_size=8 \
  --learning_rate=5e-5 \
  --num_train_epochs=2.0 \
  --warmup_proportion=.1 \
  --save_checkpoints_steps=5000 \
  --n_best_size=20 \
  --max_answer_length=30

You can fine-tune the model starting from TF-Hub modules instead of raw checkpoints by setting e.g. --albert_hub_module_handle=https://tfhub.dev/google/albert_base/1 instead of --init_checkpoint.

For SQuAD v2, use the run_squad_v2.py script:

pip install -r albert/requirements.txt
python -m albert.run_squad_v2 \
  --albert_config_file=... \
  --output_dir=... \
  --train_file=... \
  --predict_file=... \
  --train_feature_file=... \
  --predict_feature_file=... \
  --predict_feature_left_file=... \
  --init_checkpoint=... \
  --spm_model_file=... \
  --do_lower_case \
  --max_seq_length=384 \
  --doc_stride=128 \
  --max_query_length=64 \
  --do_train \
  --do_predict \
  --train_batch_size=48 \
  --predict_batch_size=8 \
  --learning_rate=5e-5 \
  --num_train_epochs=2.0 \
  --warmup_proportion=.1 \
  --save_checkpoints_steps=5000 \
  --n_best_size=20 \
  --max_answer_length=30

You can fine-tune the model starting from TF-Hub modules instead of raw checkpoints by setting e.g. --albert_hub_module_handle=https://tfhub.dev/google/albert_base/1 instead of --init_checkpoint.

Fine-tuning on RACE

For RACE, use the run_race.py script:

pip install -r albert/requirements.txt
python -m albert.run_race \
  --albert_config_file=... \
  --output_dir=... \
  --train_file=... \
  --eval_file=... \
  --data_dir=...\
  --init_checkpoint=... \
  --spm_model_file=... \
  --max_seq_length=512 \
  --max_qa_length=128 \
  --do_train \
  --do_eval \
  --train_batch_size=32 \
  --eval_batch_size=8 \
  --learning_rate=1e-5 \
  --train_step=12000 \
  --warmup_step=1000 \
  --save_checkpoints_steps=100

You can fine-tune the model starting from TF-Hub modules instead of raw checkpoints by setting e.g. --albert_hub_module_handle=https://tfhub.dev/google/albert_base/1 instead of --init_checkpoint.

SentencePiece

Command for generating the sentence piece vocabulary:

spm_train \
--input all.txt --model_prefix=30k-clean --vocab_size=30000 --logtostderr
--pad_id=0 --unk_id=1 --eos_id=-1 --bos_id=-1
--control_symbols=[CLS],[SEP],[MASK]
--user_defined_symbols="(,),\",-,.,–,£,€"
--shuffle_input_sentence=true --input_sentence_size=10000000
--character_coverage=0.99995 --model_type=unigram
Owner
Google Research
Google Research
Cryptocurrency Prediction with Artificial Intelligence (Deep Learning via LSTM Neural Networks)

Cryptocurrency Prediction with Artificial Intelligence (Deep Learning via LSTM Neural Networks)- Emirhan BULUT

Emirhan BULUT 102 Nov 18, 2022
A simple editor for captions in .SRT file extension

WaySRT A simple editor for captions in .SRT file extension The program doesn't use any external dependecies, just run: python way_srt.py {file_name.sr

Gustavo Lopes 3 Nov 16, 2022
Benchmark for the generalization of 3D machine learning models across different remeshing/samplings of a surface.

Discretization Robust Correspondence Benchmark One challenge of machine learning on 3D surfaces is that there are many different representations/sampl

Nicholas Sharp 10 Sep 30, 2022
Baseline of DCASE 2020 task 4

Couple Learning for SED This repository provides the data and source code for sound event detection (SED) task. The improvement of the Couple Learning

21 Oct 18, 2022
Realtime YOLO Monster Detection With Non Maximum Supression

Realtime-YOLO-Monster-Detection-With-Non-Maximum-Supression Table of Contents In

5 Oct 07, 2022
Do Neural Networks for Segmentation Understand Insideness?

This is part of the code to reproduce the results of the paper Do Neural Networks for Segmentation Understand Insideness? [pdf] by K. Villalobos (*),

biolins 0 Mar 20, 2021
:fire: 2D and 3D Face alignment library build using pytorch

Face Recognition Detect facial landmarks from Python using the world's most accurate face alignment network, capable of detecting points in both 2D an

Adrian Bulat 6k Dec 31, 2022
Code repository for our paper "Learning to Generate Scene Graph from Natural Language Supervision" in ICCV 2021

Scene Graph Generation from Natural Language Supervision This repository includes the Pytorch code for our paper "Learning to Generate Scene Graph fro

Yiwu Zhong 64 Dec 24, 2022
The repository for the paper "When Do You Need Billions of Words of Pretraining Data?"

pretraining-learning-curves This is the repository for the paper When Do You Need Billions of Words of Pretraining Data? Edge Probing We use jiant1 fo

ML² AT CILVR 19 Nov 25, 2022
Complete U-net Implementation with keras

U Net Lowered with Keras Complete U-net Implementation with keras Original Paper Link : https://arxiv.org/abs/1505.04597 Special Implementations : The

Sagnik Roy 14 Oct 10, 2022
DeepAL: Deep Active Learning in Python

DeepAL: Deep Active Learning in Python Python implementations of the following active learning algorithms: Random Sampling Least Confidence [1] Margin

Kuan-Hao Huang 583 Jan 03, 2023
Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes (CVPR 2021 Oral)

Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Surfaces Official code release for NGLOD. For technical details, please refer t

659 Dec 27, 2022
QR2Pass-project - A proof of concept for an alternative (passwordless) authentication system to a web server

QR2Pass This is a proof of concept for an alternative (passwordless) authenticat

4 Dec 09, 2022
Cancer-and-Tumor-Detection-Using-Inception-model - In this repo i am gonna show you how i did cancer/tumor detection in lungs using deep neural networks, specifically here the Inception model by google.

Cancer-and-Tumor-Detection-Using-Inception-model In this repo i am gonna show you how i did cancer/tumor detection in lungs using deep neural networks

Deepak Nandwani 1 Jan 01, 2022
Optimizers-visualized - Visualization of different optimizers on local minimas and saddle points.

Optimizers Visualized Visualization of how different optimizers handle mathematical functions for optimization. Contents Installation Usage Functions

Gautam J 1 Jan 01, 2022
Code for the paper "Curriculum Dropout", ICCV 2017

Curriculum Dropout Dropout is a very effective way of regularizing neural networks. Stochastically "dropping out" units with a certain probability dis

Pietro Morerio 21 Jan 02, 2022
(EI 2022) Controllable Confidence-Based Image Denoising

Image Denoising with Control over Deep Network Hallucination Paper and arXiv preprint -- Our frequency-domain insights derive from SFM and the concept

Images and Visual Representation Laboratory (IVRL) at EPFL 5 Dec 18, 2022
A denoising diffusion probabilistic model (DDPM) tailored for conditional generation of protein distograms

Denoising Diffusion Probabilistic Model for Proteins Implementation of Denoising Diffusion Probabilistic Model in Pytorch. It is a new approach to gen

Phil Wang 108 Nov 23, 2022
End-To-End Optimization of LiDAR Beam Configuration

End-To-End Optimization of LiDAR Beam Configuration arXiv | IEEE Xplore This repository is the official implementation of the paper: End-To-End Optimi

Niclas 30 Nov 28, 2022
Creative Applications of Deep Learning w/ Tensorflow

Creative Applications of Deep Learning w/ Tensorflow This repository contains lecture transcripts and homework assignments as Jupyter Notebooks for th

Parag K Mital 1.5k Dec 30, 2022