OTAR3088
non-profit
AI & ML interests
None defined yet.
Recent Activity
View all activity
This collection consolidates NER models trained using a 2-class label schema, predicting CellLine and Cell_Tissue (combined CellType and Tissue)
-
OTAR3088/CeLLaTe-ner-2class-bioformer8l-baseline
Token Classification • 42.3M • Updated • 13 -
OTAR3088/CeLLaTe-ner-2class-bioformer16l-baseline
Token Classification • 41.4M • Updated • 27 -
OTAR3088/CeLLaTe-ner-2class-pubmedbert-baseline
Token Classification • 0.1B • Updated • 11 -
OTAR3088/CeLLaTe-ner-2class-pubmedbert-tapt-tokenizer-original-v1
Token Classification • 0.1B • Updated • 15
Datasets used in the curation of CeLLaTe data. These downstream curated-data have gone on to being used for NER transformer model production.
Current datasets containing annotations of Adverse Events (AE, AdEv), used for train/test cycles
This collection consolidates NER models trained using a 3-class label schema, predicting CellLine , CellType and Tissue entities
This collection contains raw dataset zips, consisting of LabelStudio-formatted .json files, named by their corresponding Europe PMC identifiers.
Data produced relating to model train/test cycles which have been archived - due to data versioning or priorities changing
Root version of CeLLaTe. Best represents curation guides set out for CeLLaTe prior to any testing / edits of data during model production.
This collection consolidates NER models trained using a 3-class label schema, predicting CellLine , CellType and Tissue entities
This collection consolidates NER models trained using a 2-class label schema, predicting CellLine and Cell_Tissue (combined CellType and Tissue)
-
OTAR3088/CeLLaTe-ner-2class-bioformer8l-baseline
Token Classification • 42.3M • Updated • 13 -
OTAR3088/CeLLaTe-ner-2class-bioformer16l-baseline
Token Classification • 41.4M • Updated • 27 -
OTAR3088/CeLLaTe-ner-2class-pubmedbert-baseline
Token Classification • 0.1B • Updated • 11 -
OTAR3088/CeLLaTe-ner-2class-pubmedbert-tapt-tokenizer-original-v1
Token Classification • 0.1B • Updated • 15
This collection contains raw dataset zips, consisting of LabelStudio-formatted .json files, named by their corresponding Europe PMC identifiers.
Datasets used in the curation of CeLLaTe data. These downstream curated-data have gone on to being used for NER transformer model production.
Current datasets containing annotations of Adverse Events (AE, AdEv), used for train/test cycles
Data produced relating to model train/test cycles which have been archived - due to data versioning or priorities changing
Root version of CeLLaTe. Best represents curation guides set out for CeLLaTe prior to any testing / edits of data during model production.