finnnnnnnnnnnn/toki-pona-word2vec

This is a 100-dimensional word2vec model trained on this Toki Pona dataset.

The file model.txt contains the model's vocab and embeddings. You can load the model with the following code:

from huggingface_hub import hf_hub_download
from gensim.models import KeyedVectors

model_path = hf_hub_download(
    repo_id="finnnnnnnnnnnn/toki-pona-word2vec",
    filename="model.txt",
)

model = KeyedVectors.load_word2vec_format(model_path, binary=False)
print(model.most_similar("mi"))