bart_samsum_finetuned_for_asr

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

• Loss: 1.4623
• Rouge1: 54.4349
• Rouge2: 29.4619
• Rougel: 44.7701
• Rougelsum: 50.2825
• Gen Len: 30.2751

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.3855	0.9997	1841	1.5274	52.3133	27.7547	43.0233	48.3891	30.3004
1.0882	2.0	3683	1.4969	53.2457	28.4129	44.1196	49.1579	30.2637
0.8376	2.9997	5524	1.5882	52.6929	27.5974	43.3339	47.9779	30.8034
0.6756	4.0	7366	1.6617	52.5167	27.1278	43.037	48.1382	30.5299
0.5417	4.9986	9205	1.8083	52.0696	26.8054	42.6108	47.5455	30.2894

Framework versions

• Transformers 4.42.4
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1