---
license: apache-2.0
base_model: allenai/longformer-base-4096
tags:
- generated_from_trainer
datasets:
- essays_su_g
metrics:
- accuracy
model-index:
- name: longformer-spans
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: essays_su_g
      type: essays_su_g
      config: spans
      split: test
      args: spans
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.9393385646207316
---

<!-- 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. -->

# longformer-spans

This model is a fine-tuned version of [allenai/longformer-base-4096](https://huggingface.co/allenai/longformer-base-4096) on the essays_su_g dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1675
- B: {'precision': 0.8321678321678322, 'recall': 0.898961284230406, 'f1-score': 0.864275987290059, 'support': 1059.0}
- I: {'precision': 0.9499635384529085, 'recall': 0.9635846372688478, 'f1-score': 0.956725608722671, 'support': 17575.0}
- O: {'precision': 0.9318639516670396, 'recall': 0.8980053908355795, 'f1-score': 0.9146214242573986, 'support': 9275.0}
- Accuracy: 0.9393
- Macro avg: {'precision': 0.9046651074292601, 'recall': 0.9201837707782777, 'f1-score': 0.9118743400900429, 'support': 27909.0}
- Weighted avg: {'precision': 0.939478772950926, 'recall': 0.9393385646207316, 'f1-score': 0.9392251443558882, 'support': 27909.0}

## 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: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | B                                                                                                                  | I                                                                                                                   | O                                                                                                                  | Accuracy | Macro avg                                                                                                           | Weighted avg                                                                                                        |
|:-------------:|:-----:|:----:|:---------------:|:------------------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------------:|:--------:|:-------------------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------------------:|
| No log        | 1.0   | 41   | 0.2773          | {'precision': 0.7656804733727811, 'recall': 0.6109537299339, 'f1-score': 0.6796218487394958, 'support': 1059.0}    | {'precision': 0.9200088755755256, 'recall': 0.943669985775249, 'f1-score': 0.931689230942082, 'support': 17575.0}   | {'precision': 0.8860241230496846, 'recall': 0.8632884097035041, 'f1-score': 0.8745085190039319, 'support': 9275.0} | 0.9043   | {'precision': 0.8572378239993305, 'recall': 0.8059707084708844, 'f1-score': 0.82860653289517, 'support': 27909.0}   | {'precision': 0.9028587678106511, 'recall': 0.9043319359346448, 'f1-score': 0.9031217272343576, 'support': 27909.0} |
| No log        | 2.0   | 82   | 0.1955          | {'precision': 0.7943201376936316, 'recall': 0.8715769593956563, 'f1-score': 0.8311571364250337, 'support': 1059.0} | {'precision': 0.9362793776895068, 'recall': 0.9656330014224751, 'f1-score': 0.9507296714377748, 'support': 17575.0} | {'precision': 0.9372462591346712, 'recall': 0.8711590296495957, 'f1-score': 0.9029950827000447, 'support': 9275.0} | 0.9307   | {'precision': 0.8892819248392699, 'recall': 0.9027896634892424, 'f1-score': 0.8949606301876177, 'support': 27909.0} | {'precision': 0.9312140937398226, 'recall': 0.9306675266043212, 'f1-score': 0.9303288822614897, 'support': 27909.0} |
| No log        | 3.0   | 123  | 0.1872          | {'precision': 0.7751385589865399, 'recall': 0.9244570349386213, 'f1-score': 0.8432385874246339, 'support': 1059.0} | {'precision': 0.9386327328816174, 'recall': 0.96950213371266, 'f1-score': 0.9538177339901479, 'support': 17575.0}   | {'precision': 0.9483103732485576, 'recall': 0.868355795148248, 'f1-score': 0.9065736154885187, 'support': 9275.0}  | 0.9342   | {'precision': 0.8873605550389051, 'recall': 0.9207716545998431, 'f1-score': 0.9012099789677669, 'support': 27909.0} | {'precision': 0.9356451584163368, 'recall': 0.9341789386936113, 'f1-score': 0.933921194690442, 'support': 27909.0}  |
| No log        | 4.0   | 164  | 0.1684          | {'precision': 0.8173322005097706, 'recall': 0.9084041548630784, 'f1-score': 0.8604651162790699, 'support': 1059.0} | {'precision': 0.9426896055761464, 'recall': 0.9696159317211949, 'f1-score': 0.9559631998204869, 'support': 17575.0} | {'precision': 0.9440785673021375, 'recall': 0.8809703504043127, 'f1-score': 0.9114333519241495, 'support': 9275.0} | 0.9378   | {'precision': 0.9013667911293516, 'recall': 0.9196634789961954, 'f1-score': 0.9092872226745689, 'support': 27909.0} | {'precision': 0.938394544056324, 'recall': 0.9378336737253216, 'f1-score': 0.9375409414196524, 'support': 27909.0}  |
| No log        | 5.0   | 205  | 0.1675          | {'precision': 0.8321678321678322, 'recall': 0.898961284230406, 'f1-score': 0.864275987290059, 'support': 1059.0}   | {'precision': 0.9499635384529085, 'recall': 0.9635846372688478, 'f1-score': 0.956725608722671, 'support': 17575.0}  | {'precision': 0.9318639516670396, 'recall': 0.8980053908355795, 'f1-score': 0.9146214242573986, 'support': 9275.0} | 0.9393   | {'precision': 0.9046651074292601, 'recall': 0.9201837707782777, 'f1-score': 0.9118743400900429, 'support': 27909.0} | {'precision': 0.939478772950926, 'recall': 0.9393385646207316, 'f1-score': 0.9392251443558882, 'support': 27909.0}  |


### Framework versions

- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2