Revisiting Transformer-based Models for Long Document Classification
Paper • 2204.06683 • Published
# Load model directly
from transformers import AutoTokenizer, AutoModelForMaskedLM
tokenizer = AutoTokenizer.from_pretrained("xdai/mimic_roberta_base")
model = AutoModelForMaskedLM.from_pretrained("xdai/mimic_roberta_base")Quick Links
YAML Metadata Error:"language[0]" must only contain lowercase characters
YAML Metadata Error:"language[0]" with value "English" is not valid. It must be an ISO 639-1, 639-2 or 639-3 code (two/three letters), or a special value like "code", "multilingual". If you want to use BCP-47 identifiers, you can specify them in language_bcp47.
Continue pre-training RoBERTa-base using discharge summaries from MIMIC-III datasets.
Details can be found in the following paper
Xiang Dai and Ilias Chalkidis and Sune Darkner and Desmond Elliott. 2022. Revisiting Transformer-based Models for Long Document Classification. (https://arxiv.org/abs/2204.06683)
- Important hyper-parameters
| Max sequence | 128 |
| Batch size | 128 |
| Learning rate | 5e-5 |
| Training epochs | 15 |
| Training time | 40 GPU-hours |
- Downloads last month
- 9
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="xdai/mimic_roberta_base")