google-bert-bert-base-uncased-ov

google-bert/bert-base-uncased

This is the google-bert/bert-base-uncased model converted to OpenVINO, for accelerated inference.

An example of how to do inference on this model:

from optimum.intel import OVModelForMaskedLM
from transformers import AutoTokenizer, pipeline

# model_id should be set to either a local directory or a model available on the HuggingFace hub.
model_id = "helenai/google-bert-bert-base-uncased-ov"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = OVModelForMaskedLM.from_pretrained(model_id)
pipe = pipeline("fill-mask", model=model, tokenizer=tokenizer)
result = pipe(f"I am a {tokenizer.mask_token} model")
print(result)