meta_data/README_s42_e4.md

metadata

base_model: allenai/longformer-base-4096
tags:
  - generated_from_trainer
datasets:
  - essays_su_g
metrics:
  - accuracy
model-index:
  - name: longformer-sep_tok
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: essays_su_g
          type: essays_su_g
          config: sep_tok
          split: train[80%:100%]
          args: sep_tok
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.888301632721764

longformer-sep_tok

This model is a fine-tuned version of allenai/longformer-base-4096 on the essays_su_g dataset. It achieves the following results on the evaluation set:

• Loss: 0.2543
• Claim: {'precision': 0.6146161934805467, 'recall': 0.5609404990403071, 'f1-score': 0.5865529352734571, 'support': 4168.0}
• Majorclaim: {'precision': 0.8342541436464088, 'recall': 0.8420074349442379, 'f1-score': 0.8381128584643849, 'support': 2152.0}
• O: {'precision': 0.9999115122555526, 'recall': 0.998939179632249, 'f1-score': 0.9994251094503163, 'support': 11312.0}
• Premise: {'precision': 0.8800289668490505, 'recall': 0.9059057400811729, 'f1-score': 0.8927798865352434, 'support': 12073.0}
• Accuracy: 0.8883
• Macro avg: {'precision': 0.8322027040578897, 'recall': 0.8269482134244918, 'f1-score': 0.8292176974308504, 'support': 29705.0}
• Weighted avg: {'precision': 0.8851245229744956, 'recall': 0.888301632721764, 'f1-score': 0.8864635554242416, 'support': 29705.0}

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: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Claim	Majorclaim	O	Premise	Accuracy	Macro avg	Weighted avg
No log	1.0	41	0.4315	{'precision': 0.3877917414721723, 'recall': 0.2072936660268714, 'f1-score': 0.2701688555347092, 'support': 4168.0}	{'precision': 0.6876106194690266, 'recall': 0.36105947955390333, 'f1-score': 0.473491773308958, 'support': 2152.0}	{'precision': 0.9996342021033379, 'recall': 0.9663189533239038, 'f1-score': 0.982694295860116, 'support': 11312.0}	{'precision': 0.7669997404619777, 'recall': 0.9791269775532179, 'f1-score': 0.8601782790613062, 'support': 12073.0}	0.8212	{'precision': 0.7105090758766286, 'recall': 0.6284497691144741, 'f1-score': 0.6466333009412724, 'support': 29705.0}	{'precision': 0.7966303313362657, 'recall': 0.8211748863827638, 'f1-score': 0.7960339445852997, 'support': 29705.0}
No log	2.0	82	0.2780	{'precision': 0.6238676644348169, 'recall': 0.3800383877159309, 'f1-score': 0.4723423289100939, 'support': 4168.0}	{'precision': 0.7306525037936267, 'recall': 0.8949814126394052, 'f1-score': 0.8045112781954887, 'support': 2152.0}	{'precision': 0.9998226164079823, 'recall': 0.996552333804809, 'f1-score': 0.9981847965643954, 'support': 11312.0}	{'precision': 0.8501697472651829, 'recall': 0.9334051188602667, 'f1-score': 0.8898452305748579, 'support': 12073.0}	0.8770	{'precision': 0.8011281329754022, 'recall': 0.8012443132551029, 'f1-score': 0.791220908561209, 'support': 29705.0}	{'precision': 0.8667475983527302, 'recall': 0.8770240700218819, 'f1-score': 0.866338966000359, 'support': 29705.0}
No log	3.0	123	0.2682	{'precision': 0.6295336787564767, 'recall': 0.46641074856046066, 'f1-score': 0.535832414553473, 'support': 4168.0}	{'precision': 0.7360406091370558, 'recall': 0.9433085501858736, 'f1-score': 0.8268839103869654, 'support': 2152.0}	{'precision': 0.9999115200849407, 'recall': 0.9990275813295615, 'f1-score': 0.9994693552666489, 'support': 11312.0}	{'precision': 0.8739348570518436, 'recall': 0.9089704298848671, 'f1-score': 0.8911084043848965, 'support': 12073.0}	0.8837	{'precision': 0.8098551662575791, 'recall': 0.8294293274901907, 'f1-score': 0.8133235211479959, 'support': 29705.0}	{'precision': 0.8776256659925162, 'recall': 0.8836559501767379, 'f1-score': 0.8778708228219765, 'support': 29705.0}
No log	4.0	164	0.2543	{'precision': 0.6146161934805467, 'recall': 0.5609404990403071, 'f1-score': 0.5865529352734571, 'support': 4168.0}	{'precision': 0.8342541436464088, 'recall': 0.8420074349442379, 'f1-score': 0.8381128584643849, 'support': 2152.0}	{'precision': 0.9999115122555526, 'recall': 0.998939179632249, 'f1-score': 0.9994251094503163, 'support': 11312.0}	{'precision': 0.8800289668490505, 'recall': 0.9059057400811729, 'f1-score': 0.8927798865352434, 'support': 12073.0}	0.8883	{'precision': 0.8322027040578897, 'recall': 0.8269482134244918, 'f1-score': 0.8292176974308504, 'support': 29705.0}	{'precision': 0.8851245229744956, 'recall': 0.888301632721764, 'f1-score': 0.8864635554242416, 'support': 29705.0}

Framework versions

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