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

A PaddlePaddle implementation of Time Interval Aware Self-Attentive Sequential Recommendation.

KinectFusion implemented in Python with PyTorch

Generate saved_model, tfjs, tf-trt, EdgeTPU, CoreML, quantized tflite and .pb from .tflite.

Artificial Intelligence playing minesweeper 🤖

Neurons Dataset API - The official dataloader and visualization tools for Neurons Datasets.

Code for Contrastive-Geometry Networks for Generalized 3D Pose Transfer

Match SafeGraph POIs with Data collected through a cultural resource survey in Washington DC.

Implementation of experiments in the paper Clockwork Variational Autoencoders (project website) using JAX and Flax

Predicts an answer in yes or no.

这是一个facenet-pytorch的库，可以用于训练自己的人脸识别模型。

[WWW 2021] Source code for "Graph Contrastive Learning with Adaptive Augmentation"

用强化学习DQN算法，训练AI模型来玩合成大西瓜游戏，提供Keras版本和PARL（paddle）版本

YOLOX-Paddle - A reproduction of YOLOX by PaddlePaddle

Code repository for Self-supervised Structure-sensitive Learning, CVPR'17

SGoLAM - Simultaneous Goal Localization and Mapping

Official implementation of the paper Image Generators with Conditionally-Independent Pixel Synthesis https://arxiv.org/abs/2011.13775

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

Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression

Alphabetical Letter Recognition

General Vision Benchmark, a project from OpenGVLab