bionlp/bluebert_pubmed_mimic_uncased_L-24_H-1024_A-16

The BlueBERT model was pre-trained on PubMed abstracts and clinical notes (MIMIC-III) The model was used to pre-train the models using pre-processed texts . The corpus contains around4000M words extracted from the PubMed ASCII code version . The training procedure was done using the NLTK Treebank tokenizer to tokenize the text . The data and codes are publicly available to download . The information produced on this website is not intended for direct diagnostic use or medical decision-making without review and oversight by a clinical professional . The tool shows the results of research conducted in the Computational Biology Branch, NCBI. NIH does not independently verify the validity or utility,