Downloads · 30 days
109
1% of all-time downloads
dkurt/bert-large-uncased-whole-word-masking-squad-int8-0001
bert-large-uncased-whole-word-masking-squad-int8-0001 is a question answering model from dkurt. Use it when the input is a question plus a passage. It is set up for transformers.
This is a BERT-large model pre-trained on lower-cased English text using Whole-Word-Masking and fine-tuned on the SQuAD v1.1 training set. The model performs question answering for English language; the input is a con…
Downloads · 30 days
109
1% of all-time downloads
All-time downloads
11.4K
Public
Repo size
366 MB
Likes
0
Public
Click a slice to open those files.
.bin366 MB · 99%
From the Hugging Face model README
This is a BERT-large model pre-trained on lower-cased English text using Whole-Word-Masking and fine-tuned on the SQuAD v1.1 training set. The model performs question answering for English language; the input is a concatenated premise and question for the premise, and the output is the location of the answer to the question inside the premise. For details about the original floating-point model, check out BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.
The model has been further quantized to INT8 precision using quantization-aware fine-tuning with NNCF.
Model source: Open Model Zoo