meta_data/README_s42_e5.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: 0.8854738259552264

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: 0.2621
• Claim: {'precision': 0.5981237322515213, 'recall': 0.565978886756238, 'f1-score': 0.5816074950690335, 'support': 4168.0}
• Majorclaim: {'precision': 0.8415746519443111, 'recall': 0.8145910780669146, 'f1-score': 0.8278630460448643, 'support': 2152.0}
• O: {'precision': 0.9999115904871364, 'recall': 0.9998231966053748, 'f1-score': 0.9998673915926268, 'support': 11312.0}
• Premise: {'precision': 0.8798415137058301, 'recall': 0.9012672906485546, 'f1-score': 0.8904255319148937, 'support': 12073.0}
• Accuracy: 0.8855
• Macro avg: {'precision': 0.8298628720971998, 'recall': 0.8204151130192705, 'f1-score': 0.8249408661553546, 'support': 29705.0}
• Weighted avg: {'precision': 0.8832645976626653, 'recall': 0.8854738259552264, 'f1-score': 0.8842386364262106, '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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Claim	Majorclaim	O	Premise	Accuracy	Macro avg	Weighted avg
No log	1.0	41	0.3617	{'precision': 0.4788679245283019, 'recall': 0.30446257197696736, 'f1-score': 0.37224992666471113, 'support': 4168.0}	{'precision': 0.6963803349540789, 'recall': 0.5989776951672863, 'f1-score': 0.6440169872595554, 'support': 2152.0}	{'precision': 0.9991923180472045, 'recall': 0.9842644978783592, 'f1-score': 0.9916722333556, 'support': 11312.0}	{'precision': 0.8126733518241945, 'recall': 0.9464921726165825, 'f1-score': 0.8744929976276117, 'support': 12073.0}	0.8456	{'precision': 0.7467784823384449, 'recall': 0.7085492344097988, 'f1-score': 0.7206080362268695, 'support': 29705.0}	{'precision': 0.8284396858636128, 'recall': 0.8456152162935533, 'f1-score': 0.8319479048980907, 'support': 29705.0}
No log	2.0	82	0.2796	{'precision': 0.5955649419218585, 'recall': 0.4059500959692898, 'f1-score': 0.4828078185190469, 'support': 4168.0}	{'precision': 0.760759493670886, 'recall': 0.8378252788104089, 'f1-score': 0.7974347633790357, 'support': 2152.0}	{'precision': 0.9999115357395613, 'recall': 0.9992043847241867, 'f1-score': 0.999557835160948, 'support': 11312.0}	{'precision': 0.8505686125852919, 'recall': 0.9292636461525718, 'f1-score': 0.8881763844357361, 'support': 12073.0}	0.8758	{'precision': 0.8017011459793993, 'recall': 0.7930608514141144, 'f1-score': 0.7919942003736917, 'support': 29705.0}	{'precision': 0.8651534509455714, 'recall': 0.8758458172024912, 'f1-score': 0.8671393475513334, 'support': 29705.0}
No log	3.0	123	0.2584	{'precision': 0.6091815161582603, 'recall': 0.48392514395393477, 'f1-score': 0.5393769220484022, 'support': 4168.0}	{'precision': 0.7808161548169962, 'recall': 0.862453531598513, 'f1-score': 0.8196069772576728, 'support': 2152.0}	{'precision': 0.9999115748518879, 'recall': 0.9996463932107497, 'f1-score': 0.9997789664471067, 'support': 11312.0}	{'precision': 0.8697670758577274, 'recall': 0.9155139567630249, 'f1-score': 0.892054396513458, 'support': 12073.0}	0.8832	{'precision': 0.8149190804212179, 'recall': 0.8153847563815556, 'f1-score': 0.8127043155666599, 'support': 29705.0}	{'precision': 0.8763198978646256, 'recall': 0.8831509846827134, 'f1-score': 0.8783433638684701, 'support': 29705.0}
No log	4.0	164	0.2543	{'precision': 0.5829736211031175, 'recall': 0.583253358925144, 'f1-score': 0.58311345646438, 'support': 4168.0}	{'precision': 0.8634197988353626, 'recall': 0.7578996282527881, 'f1-score': 0.8072259341747091, 'support': 2152.0}	{'precision': 0.9999115904871364, 'recall': 0.9998231966053748, 'f1-score': 0.9998673915926268, 'support': 11312.0}	{'precision': 0.8795297932711795, 'recall': 0.89861674811563, 'f1-score': 0.8889708292363159, 'support': 12073.0}	0.8827	{'precision': 0.831458700924199, 'recall': 0.8098982329747342, 'f1-score': 0.8197944028670079, 'support': 29705.0}	{'precision': 0.8825947337352275, 'recall': 0.8827133479212254, 'f1-score': 0.8823636375005335, 'support': 29705.0}
No log	5.0	205	0.2621	{'precision': 0.5981237322515213, 'recall': 0.565978886756238, 'f1-score': 0.5816074950690335, 'support': 4168.0}	{'precision': 0.8415746519443111, 'recall': 0.8145910780669146, 'f1-score': 0.8278630460448643, 'support': 2152.0}	{'precision': 0.9999115904871364, 'recall': 0.9998231966053748, 'f1-score': 0.9998673915926268, 'support': 11312.0}	{'precision': 0.8798415137058301, 'recall': 0.9012672906485546, 'f1-score': 0.8904255319148937, 'support': 12073.0}	0.8855	{'precision': 0.8298628720971998, 'recall': 0.8204151130192705, 'f1-score': 0.8249408661553546, 'support': 29705.0}	{'precision': 0.8832645976626653, 'recall': 0.8854738259552264, 'f1-score': 0.8842386364262106, 'support': 29705.0}

Framework versions

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