summarizer

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

• Loss: 2.4858
• Rouge1: 50.2321
• Rouge2: 24.0325
• Rougel: 45.4209
• Rougelsum: 45.4669

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: 4
• eval_batch_size: 4
• 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
2.324	1.0	460	1.8943	49.6772	24.3924	45.0788	45.1276
1.5791	2.0	920	1.8899	49.6626	24.2944	45.2762	45.4145
1.0975	3.0	1380	1.9199	49.725	24.1088	44.9682	45.0787
0.7581	4.0	1840	2.1324	48.8128	22.5742	44.3464	44.3182
0.5103	5.0	2300	2.3014	49.5098	23.3978	44.8389	44.9802
0.3448	6.0	2760	2.3828	49.8076	23.3584	44.7778	44.8633
0.2394	7.0	3220	2.4564	49.3934	23.7093	44.486	44.6473
0.1763	8.0	3680	2.4858	50.2321	24.0325	45.4209	45.4669

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3