This repository consists of a sample of PUBMED RCT predictor which was trained on 20,000 lines/data. The model starts from using a simple tokenizer(sentence) embedding model to predict the line type. As the notebook progresses, It uses the character alongside the token embeddings to predict the outputs and thus acts as a hybrid model. This is followed by using a "tri-input" model which takes character embeddings alongside token (sentence) embeddings and the line number of each lines to predict the type of line and classify them respectively.
-
Notifications
You must be signed in to change notification settings - Fork 0
This repository consists of a sample of PUBMED RCT predictor which was trained on 20,000 lines/data. The model starts from using a simple tokenizer(sentence) embedding model to predict the line type. As the notebook progresses, It uses the character alongside the token embeddings to predict the outputs and thus acts as a hybrid model. This is fo…
ikiiftekhar2/PubMed-RCT-Line-predictor-classifier
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
This repository consists of a sample of PUBMED RCT predictor which was trained on 20,000 lines/data. The model starts from using a simple tokenizer(sentence) embedding model to predict the line type. As the notebook progresses, It uses the character alongside the token embeddings to predict the outputs and thus acts as a hybrid model. This is fo…
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published