Download a Pretrained Model2Vec Model

download_embeddings_model2vec() downloads the files of a Model2Vec model, so they can be read with read_embeddings_model2vec().

Usage

download_embeddings_model2vec(url, path = tempfile("model2vec-"))

Arguments

url

A string giving the URL of the folder that holds the model files. For a model on the Hugging Face Hub this has the form https://huggingface.co/<owner>/<model>/resolve/<revision>, where <revision> is a branch name such as main or a commit hash. Use a commit hash to always get the same version of the model.

path

A string giving the path to the folder to download the files into. The folder is created if it doesn't exist, and files in it with the same names are overwritten. Defaults to a new temporary folder, which is removed when the R session ends.

Value

The path to the folder holding the files, invisibly.

Details

Two files are downloaded: model.safetensors, which holds the vectors, and vocab.txt, which holds the tokens. The files of a small model are about 30 MB, so the download can take a while. The downloaded files are not needed once a recipe using the embeddings has been prepped.

The embeddings are tied to the tokenizer the model was trained with. See vignette("static-embeddings", package = "textrecipes") for how to use a downloaded model in a recipe.

See also

read_embeddings_model2vec() to read the downloaded files.

Examples

if (FALSE) {
url <- paste0(
  "https://huggingface.co/minishlab/potion-base-8M/resolve/",
  "bf8b056651a2c21b8d2565580b8569da283cab23"
)

model_dir <- download_embeddings_model2vec(url)
embeddings <- read_embeddings_model2vec(model_dir)
}