BARTModel_ExerciseLog

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: 4.3574
• Rouge1: 0.8531
• Rouge2: 0.581
• Rougel: 0.8531
• Rougelsum: 0.8531
• Gen Len: 7.2857

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 18

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	2	7.5886	0.4437	0.1117	0.4458	0.4459	16.5714
No log	2.0	4	6.4463	0.4127	0.1117	0.4136	0.4155	16.0
No log	3.0	6	5.9474	0.5136	0.0952	0.5102	0.5204	8.2857
No log	4.0	8	5.6884	0.5537	0.2381	0.5459	0.5459	8.0
No log	5.0	10	5.5303	0.55	0.2381	0.5452	0.5452	6.5714
No log	6.0	12	5.4149	0.55	0.2381	0.5452	0.5452	6.7143
No log	7.0	14	5.2489	0.5286	0.2	0.5197	0.5197	7.0
No log	8.0	16	5.0956	0.5388	0.2	0.5286	0.5286	7.0
No log	9.0	18	4.9528	0.5388	0.2	0.5286	0.5286	7.0
No log	10.0	20	4.8291	0.5388	0.2	0.5286	0.5286	7.1429
No log	11.0	22	4.7158	0.5748	0.2714	0.567	0.5667	7.4286
No log	12.0	24	4.6173	0.7388	0.3429	0.7388	0.7388	7.2857
No log	13.0	26	4.5333	0.8531	0.581	0.8531	0.8531	7.2857
No log	14.0	28	4.4660	0.8531	0.581	0.8531	0.8531	7.2857
No log	15.0	30	4.4177	0.8531	0.581	0.8531	0.8531	7.2857
No log	16.0	32	4.3857	0.8531	0.581	0.8531	0.8531	7.2857
No log	17.0	34	4.3660	0.8531	0.581	0.8531	0.8531	7.2857
No log	18.0	36	4.3574	0.8531	0.581	0.8531	0.8531	7.2857

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1