Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.

Last update: Dec 29, 2022

Related tags

Deep Learning wechsel

Overview

WECHSEL

Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.

arXiv: https://arxiv.org/abs/2112.06598

Models from the paper are available on the HuggingFace Hub:

Installation

We distribute a Python Package via PyPI:

pip install wechsel

Alternatively, clone the repository, install requirements.txt and run the code in wechsel/.

Example usage

Transferring English roberta-base to Swahili:

import torch
from transformers import AutoModel, AutoTokenizer
from datasets import load_dataset
from wechsel import WECHSEL, load_embeddings

source_tokenizer = AutoTokenizer.from_pretrained("roberta-base")
model = AutoModel.from_pretrained("roberta-base")

target_tokenizer = source_tokenizer.train_new_from_iterator(
    load_dataset("oscar", "unshuffled_deduplicated_sw", split="train")["text"],
    vocab_size=len(source_tokenizer)
)

wechsel = WECHSEL(
    load_embeddings("en"),
    load_embeddings("sw"),
    bilingual_dictionary="swahili"
)

target_embeddings, info = wechsel.apply(
    source_tokenizer,
    target_tokenizer,
    model.get_input_embeddings().weight.detach().numpy(),
)

model.get_input_embeddings().weight.data = torch.from_numpy(target_embeddings)

# use `model` and `target_tokenizer` to continue training in Swahili!

Bilingual dictionaries

We distribute 3276 bilingual dictionaries from English to other languages for use with WECHSEL in dicts/.

Citation

Please cite WECHSEL as

@misc{minixhofer2021wechsel,
      title={WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models}, 
      author={Benjamin Minixhofer and Fabian Paischer and Navid Rekabsaz},
      year={2021},
      eprint={2112.06598},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Code for WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.

Related tags

Overview

WECHSEL

Installation

Example usage

Bilingual dictionaries

Citation

Owner

Institute of Computational Perception

pytorch implementation of fast-neural-style

An example to implement a new backbone with OpenMMLab framework.

🌎 The Modern Declarative Data Flow Framework for the AI Empowered Generation.

Official repo for BMVC2021 paper ASFormer: Transformer for Action Segmentation

PyStan, a Python interface to Stan, a platform for statistical modeling. Documentation: https://pystan.readthedocs.io

EsViT: Efficient self-supervised Vision Transformers

A Gura parser implementation for Python

Keeping it safe - AI Based COVID-19 Tracker using Deep Learning and facial recognition

Pytorch Implementation for (STANet+ and STANet)

Modelisation on galaxy evolution using PEGASE-HR

Code for training and evaluation of the model from "Language Generation with Recurrent Generative Adversarial Networks without Pre-training"

OrienMask: Real-time Instance Segmentation with Discriminative Orientation Maps

a minimal terminal with python 😎😉

Local trajectory planner based on a multilayer graph framework for autonomous race vehicles.

Stroke-predictions-ml-model - Machine learning model to predict individuals chances of having a stroke

Pop-Out Motion: 3D-Aware Image Deformation via Learning the Shape Laplacian (CVPR 2022)

Unofficial implementation of MUSIQ (Multi-Scale Image Quality Transformer)

A MatConvNet-based implementation of the Fully-Convolutional Networks for image segmentation

Stitch it in Time: GAN-Based Facial Editing of Real Videos

Official Implementation of Neural Splines