metadata

license: apache-2.0
base_model: allenai/longformer-base-4096
tags:
  - generated_from_trainer
datasets:
  - fancy_dataset
metrics:
  - accuracy
model-index:
  - name: longformer-spans
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: fancy_dataset
          type: fancy_dataset
          config: spans
          split: test
          args: spans
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9297359274785911

longformer-spans

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: 0.1948
B: {'precision': 0.773955773955774, 'recall': 0.8923512747875354, 'f1-score': 0.8289473684210527, 'support': 1059.0}
I: {'precision': 0.9401371161027814, 'recall': 0.9597155049786629, 'f1-score': 0.949825430791756, 'support': 17575.0}
O: {'precision': 0.9301474791357037, 'recall': 0.8771967654986523, 'f1-score': 0.9028964598823661, 'support': 9275.0}
Accuracy: 0.9297
Macro avg: {'precision': 0.8814134563980863, 'recall': 0.9097545150882835, 'f1-score': 0.893889753031725, 'support': 27909.0}
Weighted avg: {'precision': 0.9305115500057043, 'recall': 0.9297359274785911, 'f1-score': 0.929642834739043, '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	B	I	O	Accuracy	Macro avg	Weighted avg
No log	1.0	41	0.2901	{'precision': 0.8336673346693386, 'recall': 0.392823418319169, 'f1-score': 0.5340179717586649, 'support': 1059.0}	{'precision': 0.9134376209164778, 'recall': 0.9402560455192034, 'f1-score': 0.9266528346324231, 'support': 17575.0}	{'precision': 0.8671531280180277, 'recall': 0.871266846361186, 'f1-score': 0.8692051199311606, 'support': 9275.0}	0.8966	{'precision': 0.8714193612012814, 'recall': 0.7347821033998527, 'f1-score': 0.7766253087740829, 'support': 27909.0}	{'precision': 0.8950290285352085, 'recall': 0.8965566663083593, 'f1-score': 0.892662800104582, 'support': 27909.0}
No log	2.0	82	0.2109	{'precision': 0.7553191489361702, 'recall': 0.8715769593956563, 'f1-score': 0.8092941692240245, 'support': 1059.0}	{'precision': 0.9298303409652446, 'recall': 0.9635846372688478, 'f1-score': 0.9464066167430425, 'support': 17575.0}	{'precision': 0.9366296908189757, 'recall': 0.8557412398921833, 'f1-score': 0.8943602456476422, 'support': 9275.0}	0.9243	{'precision': 0.8739263935734636, 'recall': 0.8969676121855624, 'f1-score': 0.883353677204903, 'support': 27909.0}	{'precision': 0.9254681860164671, 'recall': 0.9242538249310258, 'f1-score': 0.9239073450445768, 'support': 27909.0}
No log	3.0	123	0.1948	{'precision': 0.773955773955774, 'recall': 0.8923512747875354, 'f1-score': 0.8289473684210527, 'support': 1059.0}	{'precision': 0.9401371161027814, 'recall': 0.9597155049786629, 'f1-score': 0.949825430791756, 'support': 17575.0}	{'precision': 0.9301474791357037, 'recall': 0.8771967654986523, 'f1-score': 0.9028964598823661, 'support': 9275.0}	0.9297	{'precision': 0.8814134563980863, 'recall': 0.9097545150882835, 'f1-score': 0.893889753031725, 'support': 27909.0}	{'precision': 0.9305115500057043, 'recall': 0.9297359274785911, 'f1-score': 0.929642834739043, 'support': 27909.0}

Framework versions

Transformers 4.37.2
Pytorch 2.2.0+cu121
Datasets 2.17.0
Tokenizers 0.15.2