longformer-schema-linking-column

This model is a fine-tuned version of allenai/longformer-base-4096 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2077
• Accuracy: 0.9347
• F1: 0.8362
• Precision: 0.8688
• Recall: 0.8059

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: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.3589	0.1166	400	0.5544	0.7933	0.0	1.0	0.0
0.24	0.2332	800	0.2523	0.9007	0.7491	0.7842	0.7171
0.1828	0.3498	1200	0.2099	0.9130	0.7612	0.8793	0.6711
0.1545	0.4663	1600	0.3050	0.9123	0.7598	0.8755	0.6711
0.143	0.5829	2000	0.2551	0.9225	0.7986	0.8626	0.7434
0.1381	0.6995	2400	0.2263	0.9279	0.8166	0.8613	0.7763
0.1188	0.8161	2800	0.2768	0.9211	0.7860	0.8950	0.7007
0.1116	0.9327	3200	0.2077	0.9347	0.8362	0.8688	0.8059

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0