bart-finetuned-squad

This model is a fine-tuned version of p208p2002/bart-squad-qg-hl on the squad dataset. It achieves the following results on the evaluation set:

• Loss: 1.8813
• Rouge1: 50.1505
• Rouge2: 26.8606
• Rougel: 46.0203
• Rougelsum: 46.0242

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: 5.6e-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: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
1.5702	1.0	125	1.4266	49.7474	26.6965	46.3227	46.342
0.84	2.0	250	1.4845	49.8379	26.3973	45.126	45.1791
0.535	3.0	375	1.6037	50.1413	27.4581	46.7795	46.8001
0.3621	4.0	500	1.6899	49.6087	25.9818	45.0914	45.1004
0.2448	5.0	625	1.7540	49.7468	26.5312	45.5623	45.5296
0.1756	6.0	750	1.8287	49.4987	26.2315	45.3515	45.4214
0.13	7.0	875	1.8809	49.6426	26.4688	45.5167	45.5427
0.1016	8.0	1000	1.8813	50.1505	26.8606	46.0203	46.0242

Framework versions

• Transformers 4.24.0
• Pytorch 1.12.1+cu113
• Datasets 2.7.0
• Tokenizers 0.13.2