Views
No views yet
bert-base-uncased -> 2-layer bidirectional GRU -> linear sentiment headtorch.sigmoid():transformer-model-inner.pt: trained on Inner-Circle English varieties.transformer-model-outer.pt: trained on Outer-Circle / Indian English.modeling_bert_gru.py: model architecture and loading helpers.config.json: architecture metadata.1from huggingface_hub import hf_hub_download
2from transformers import BertTokenizer
3from modeling_bert_gru import load_model, predict_sentiment
4
5repo_id = "YOUR_USERNAME/besstie-bert-gru-sentiment"
6
7checkpoint_path = hf_hub_download(repo_id, filename="transformer-model-inner.pt")
8
9tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
10model = load_model(checkpoint_path)
11
12result = predict_sentiment(
13 model,
14 tokenizer,
15 "Traditional friendly pub with excellent beer and warm service."
16)
17
18print(result)