en_med12_trf / README.md
karl2990's picture
updating model card
4c0267d
|
raw
history blame
1.73 kB
metadata
tags:
  - spacy
  - token-classification
  - text-classification
language:
  - en
model-index:
  - name: en_med12_trf
    results:
      - task:
          name: NER
          type: token-classification
        metrics:
          - name: NER Precision
            type: precision
            value: 0.9987402508
          - name: NER Recall
            type: recall
            value: 0.9991845056
          - name: NER F Score
            type: f_score
            value: 0.9989623288
Feature Description
Name en_med12_trf
Version 0.0.0
spaCy >=3.4.1,<3.5.0
Default Pipeline transformer, ner, textcat
Components transformer, ner, textcat
Vectors 0 keys, 0 unique vectors (0 dimensions)
Sources n/a
License Apache License, Version 2.0
Author Karl Renius

Label Scheme

View label scheme (15 labels for 2 components)
Component Labels
ner Denominator_Unit, Denominator_Value, Dose_Form, Medication_Name, NDC, Numerator_Unit, Numerator_Value, Product_Package_Type, Product_Package_Type_Value, Quantity_Factor_Unit, Quantity_Factor_Unit_Value, Quantity_Factor_Value
textcat MEDICATION, OTHER, DEVICE

Accuracy

Type Score
ENTS_F 99.90
ENTS_P 99.87
ENTS_R 99.92
CATS_SCORE 100.00
CATS_MICRO_P 100.00
CATS_MICRO_R 100.00
CATS_MICRO_F 100.00
CATS_MACRO_P 100.00
CATS_MACRO_R 100.00
CATS_MACRO_F 100.00
CATS_MACRO_AUC 100.00
CATS_MACRO_AUC_PER_TYPE 0.00
TRANSFORMER_LOSS 19686.79
NER_LOSS 513489.20
TEXTCAT_LOSS 2.08