metadata

license: apache-2.0
base_model: allenai/longformer-base-4096
tags:
  - generated_from_trainer
datasets:
  - fancy_dataset
metrics:
  - accuracy
model-index:
  - name: longformer-one-step
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: fancy_dataset
          type: fancy_dataset
          config: full_labels
          split: test
          args: full_labels
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8161524956107349

longformer-one-step

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

Loss: 0.5164
Claim: {'precision': 0.5841029946823397, 'recall': 0.47910927456382, 'f1-score': 0.5264219952074664, 'support': 4356.0}
Majorclaim: {'precision': 0.663898774219059, 'recall': 0.7694775435380385, 'f1-score': 0.7127998301846742, 'support': 2182.0}
O: {'precision': 0.9219015280135824, 'recall': 0.8781671159029649, 'f1-score': 0.8995030369961348, 'support': 9275.0}
Premise: {'precision': 0.8377274128893001, 'recall': 0.8983961640211641, 'f1-score': 0.8670017552257859, 'support': 12096.0}
Accuracy: 0.8162
Macro avg: {'precision': 0.7519076774510703, 'recall': 0.7562875245064969, 'f1-score': 0.7514316544035153, 'support': 27909.0}
Weighted avg: {'precision': 0.8125252509519227, 'recall': 0.8161524956107349, 'f1-score': 0.8125897502575133, '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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Claim	Majorclaim	O	Premise	Accuracy	Macro avg	Weighted avg
No log	1.0	41	0.7242	{'precision': 0.4451114922813036, 'recall': 0.23829201101928374, 'f1-score': 0.3104066985645933, 'support': 4356.0}	{'precision': 0.6888297872340425, 'recall': 0.11869844179651695, 'f1-score': 0.20250195465207194, 'support': 2182.0}	{'precision': 0.7629536017331648, 'recall': 0.9112668463611859, 'f1-score': 0.8305409521937798, 'support': 9275.0}	{'precision': 0.7774552148976847, 'recall': 0.9077380952380952, 'f1-score': 0.83756054769442, 'support': 12096.0}	0.7427	{'precision': 0.6685875240365489, 'recall': 0.5439988486037705, 'f1-score': 0.5452525382762162, 'support': 27909.0}	{'precision': 0.7138351496506337, 'recall': 0.7427353183560859, 'f1-score': 0.7032996725252499, 'support': 27909.0}
No log	2.0	82	0.5451	{'precision': 0.5706823375775384, 'recall': 0.40128558310376494, 'f1-score': 0.47122253673001757, 'support': 4356.0}	{'precision': 0.6872317596566524, 'recall': 0.5870760769935839, 'f1-score': 0.6332179930795848, 'support': 2182.0}	{'precision': 0.8817295464179737, 'recall': 0.8970350404312668, 'f1-score': 0.8893164448720005, 'support': 9275.0}	{'precision': 0.819134799940942, 'recall': 0.9173280423280423, 'f1-score': 0.8654551127057172, 'support': 12096.0}	0.8042	{'precision': 0.7396946108982767, 'recall': 0.7006811857141645, 'f1-score': 0.71480302184683, 'support': 27909.0}	{'precision': 0.7908462519320261, 'recall': 0.8042208606542692, 'f1-score': 0.7936967322502336, 'support': 27909.0}
No log	3.0	123	0.5164	{'precision': 0.5841029946823397, 'recall': 0.47910927456382, 'f1-score': 0.5264219952074664, 'support': 4356.0}	{'precision': 0.663898774219059, 'recall': 0.7694775435380385, 'f1-score': 0.7127998301846742, 'support': 2182.0}	{'precision': 0.9219015280135824, 'recall': 0.8781671159029649, 'f1-score': 0.8995030369961348, 'support': 9275.0}	{'precision': 0.8377274128893001, 'recall': 0.8983961640211641, 'f1-score': 0.8670017552257859, 'support': 12096.0}	0.8162	{'precision': 0.7519076774510703, 'recall': 0.7562875245064969, 'f1-score': 0.7514316544035153, 'support': 27909.0}	{'precision': 0.8125252509519227, 'recall': 0.8161524956107349, 'f1-score': 0.8125897502575133, 'support': 27909.0}

Framework versions

Transformers 4.37.1
Pytorch 2.1.2+cu121
Datasets 2.16.1
Tokenizers 0.15.1