Hugging Face's logo

msperka
/
HeRo-finetuned-ner

Token Classification
Transformers
TensorBoard
Safetensors
roberta
generated_from_trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
HeRo-finetuned-ner
File size: 2,194 Bytes 
0eac0d1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
 
---
base_model: HeNLP/HeRo
tags:
- generated_from_trainer
datasets:
- nemo_corpus
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: HeRo-finetuned-ner
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: nemo_corpus
      type: nemo_corpus
      config: flat_token
      split: validation
      args: flat_token
    metrics:
    - name: Precision
      type: precision
      value: 0.8625592417061612
    - name: Recall
      type: recall
      value: 0.8484848484848485
    - name: F1
      type: f1
      value: 0.855464159811986
    - name: Accuracy
      type: accuracy
      value: 0.9769208008679356
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# HeRo-finetuned-ner

This model is a fine-tuned version of [HeNLP/HeRo](https://huggingface.co/HeNLP/HeRo) on the nemo_corpus dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1244
- Precision: 0.8626
- Recall: 0.8485
- F1: 0.8555
- Accuracy: 0.9769

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.2734        | 1.0   | 618  | 0.1445          | 0.8125    | 0.7576 | 0.7841 | 0.9667   |
| 0.0939        | 2.0   | 1236 | 0.1258          | 0.8449    | 0.8380 | 0.8414 | 0.9748   |
| 0.0545        | 3.0   | 1854 | 0.1244          | 0.8626    | 0.8485 | 0.8555 | 0.9769   |


### Framework versions

- Transformers 4.35.2
- Pytorch 2.0.1+cpu
- Datasets 2.15.0
- Tokenizers 0.15.0