Hugging Face's logo

RGarrido03
/
modernbert-conll2003-ner-large

Token Classification
Transformers
Safetensors
English
modernbert
ner
Eval Results (legacy)
Model card Files
xet
 Community
RGarrido03 commited on
Commit
62fdc4a
·
verified ·
1 Parent(s): 90fe529

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -13
README.md CHANGED
@@ -62,7 +62,7 @@ This model is intended for identifying entities (Persons, Organizations, Locatio
62
  ```python
63
  from transformers import pipeline
64
 
65
- ner_pipeline = pipeline("ner", model="RGarrido03/modernbert-conll2003-ner", aggregation_strategy="simple")
66
  text = "The CERN headquarters are located in Geneva, Switzerland."
67
  results = ner_pipeline(text)
68
 
@@ -96,26 +96,26 @@ The following hyperparameters were used during training:
96
 
97
  | Metric | Value |
98
  | :------------ | :----- |
99
- | **Accuracy** | 0.9711 |
100
- | **F1 Score** | 0.8851 |
101
- | **Precision** | 0.8721 |
102
- | **Recall** | 0.8985 |
103
- | **Loss** | 0.1873 |
104
 
105
  ## Evaluation on Validation Split
106
 
107
  | Metric | Value |
108
  | :------------ | :----- |
109
- | **Accuracy** | 0.9871 |
110
- | **F1 Score** | 0.9416 |
111
- | **Precision** | 0.9357 |
112
- | **Recall** | 0.9475 |
113
- | **Loss** | 0.0625 |
114
 
115
  ## Environmental Impact
116
 
117
- - **Runtime:** ~11.5 minutes (694 seconds)
118
- - **Hardware:** MacBook Pro, M5 Pro 24GB (Training speed: ~62 samples/sec)
119
 
120
  ## Citation
121
 
 
62
  ```python
63
  from transformers import pipeline
64
 
65
+ ner_pipeline = pipeline("ner", model="RGarrido03/modernbert-conll2003-ner-large", aggregation_strategy="simple")
66
  text = "The CERN headquarters are located in Geneva, Switzerland."
67
  results = ner_pipeline(text)
68
 
 
96
 
97
  | Metric | Value |
98
  | :------------ | :----- |
99
+ | **Accuracy** | 0.9743 |
100
+ | **F1 Score** | 0.8969 |
101
+ | **Precision** | 0.8834 |
102
+ | **Recall** | 0.9108 |
103
+ | **Loss** | 0.1762 |
104
 
105
  ## Evaluation on Validation Split
106
 
107
  | Metric | Value |
108
  | :------------ | :----- |
109
+ | **Accuracy** | 0.9880 |
110
+ | **F1 Score** | 0.9456 |
111
+ | **Precision** | 0.9388 |
112
+ | **Recall** | 0.9525 |
113
+ | **Loss** | 0.0484 |
114
 
115
  ## Environmental Impact
116
 
117
+ - **Runtime:** ~44 minutes
118
+ - **Hardware:** MacBook Pro, M5 Pro 24GB
119
 
120
  ## Citation
121