This module supports embeddings, wrappers and utilities for Neural NLP tasks. In-depth description of resources and their implementation details are recorded right here:
VictorNLP, a PyTorch-based NLP framework, provides intuitive interfaces and carefully designed code modularity for implementing and pipelining NLP modules. For more details, refer to the VictorNLP Specification page.
To use these embeddings, you must download embedding files, VictorNLP-formatted. Navigate to .../victornlp_utils
, and execute sudo ./download_glove.sh
to download pretrained GloVes.
Note:
download_glove.sh
installsgdown
for downloading large files.
- bert-base-uncased
- kobert
To use these embeddings, you do not need to download addition files manually. However, you must correctly install dependencies. Look on the warnings!
Note: We do not support local file loading currently, but will be soon updated.
This Korean BERT embedding require some work to run on.
- Visit AI Hub, recieve an API key, and the download the model file. Refer to
EmbeddingBERTMorph_kor/readme.txt
for detailed information. - Move
bert_config.json
,pytorch_model.bin
, andvocab.korean_morp.list
toembedding/data/EmbeddingBERTMorph_kor
. - Rename
bert_config.json
toconfig.json
.