IEETA/MultiClinNER-SV

MultiClinNER SV Models

Clinical NER models for SV, trained with Multi-Head CRF architecture.

Best Model

• Model: KB-C64-H3-E60-Arandom-%0.1-P0.5-42
• Best F1: 0.7051
• Branch: main

Usage

# Load the best model (main branch)
from transformers import AutoTokenizer, AutoModelForTokenClassification

model = AutoModelForTokenClassification.from_pretrained("IEETA/MultiClinNER-SV")
tokenizer = AutoTokenizer.from_pretrained("IEETA/MultiClinNER-SV")

# Load a specific model variant
model = AutoModelForTokenClassification.from_pretrained("IEETA/MultiClinNER-SV", revision="BRANCH_NAME")

All Models (11 variants)

Branch	Model	Best?
main	KB-C64-H3-E60-Arandom-%0.1-P0.5-42	Yes
KB-C64-H3-E60-Arandom-pct0.1-P0.2-123	KB-C64-H3-E60-Arandom-%0.1-P0.2-123
KB-C64-H3-E60-Arandom-pct0.1-P0.2-42	KB-C64-H3-E60-Arandom-%0.1-P0.2-42
KB-C64-H3-E60-Arandom-pct0.1-P0.2-999	KB-C64-H3-E60-Arandom-%0.1-P0.2-999
KB-C64-H3-E60-Arandom-pct0.1-P0.5-123	KB-C64-H3-E60-Arandom-%0.1-P0.5-123
KB-C64-H3-E60-Arandom-pct0.25-P0.2-42	KB-C64-H3-E60-Arandom-%0.25-P0.2-42
KB-C64-H3-E60-Arandom-pct0.25-P0.2-999	KB-C64-H3-E60-Arandom-%0.25-P0.2-999
KB-C64-H3-E60-Aukn-pct0.1-P0.5-123	KB-C64-H3-E60-Aukn-%0.1-P0.5-123
KB-C64-H3-E60-Aukn-pct0.1-P0.5-42	KB-C64-H3-E60-Aukn-%0.1-P0.5-42
KB-C64-H3-E60-Aukn-pct0.25-P0.2-123	KB-C64-H3-E60-Aukn-%0.25-P0.2-123
KB-C64-H3-E60-Aukn-pct0.25-P0.2-42	KB-C64-H3-E60-Aukn-%0.25-P0.2-42