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

Learning with Subset Stacking

Human Detection - Pedestrian Detection using OpenCV Python

This repo contains the pytorch implementation for Dynamic Concept Learner (accepted by ICLR 2021).

Official implementation for the paper "Attentive Prototypes for Source-free Unsupervised Domain Adaptive 3D Object Detection"

Neural Message Passing for Computer Vision

Official PyTorch implementation of the paper Image-Based CLIP-Guided Essence Transfer.

Implementation of Kalman Filter in Python

A dataset for online Arabic calligraphy

Reinforcement Learning for Automated Trading

Task-related Saliency Network For Few-shot learning

Immortal tracker

HNECV: Heterogeneous Network Embedding via Cloud model and Variational inference

Implementation of the federated dual coordinate descent (FedDCD) method.

PyTorch implementation of MLP-Mixer

Out-of-distribution detection using the pNML regret. NeurIPS2021

A PyTorch Implementation of FaceBoxes

Vignette is a face tracking software for characters using osu!framework.

Ağ tarayıcı.Gönderdiği paketler ile ağa bağlı olan cihazların IP adreslerini gösterir.

Character Controllers using Motion VAEs

Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow