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

Torch Implementation of "Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network"

zeus is a Python implementation of the Ensemble Slice Sampling method.

Nodule Generation Algorithm Baseline and template code for node21 generation track

An implementation of "MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing" (ICML 2019).

Public repository containing materials used for Feed Forward (FF) Neural Networks article.

ManipNet: Neural Manipulation Synthesis with a Hand-Object Spatial Representation - SIGGRAPH 2021

Cweqgen - The CW Equation Generator

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

👨‍💻 run nanosaur in simulation with Gazebo/Ingnition

Utilities and information for the signals.numer.ai tournament

Official PyTorch implementation of the preprint paper "Stylized Neural Painting", accepted to CVPR 2021.

Near-Optimal Sparse Allreduce for Distributed Deep Learning (published in PPoPP'22)

NAS-HPO-Bench-II is the first benchmark dataset for joint optimization of CNN and training HPs.

Deep Residual Learning for Image Recognition

TGRNet: A Table Graph Reconstruction Network for Table Structure Recognition

Jittor implementation of Recursive-NeRF: An Efficient and Dynamically Growing NeRF

TLDR: Twin Learning for Dimensionality Reduction

Efficiently computes derivatives of numpy code.

code for our ECCV 2020 paper "A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation"

This is the code for the paper "Motion-Focused Contrastive Learning of Video Representations" (ICCV'21).