Synopsis_summarization_t5

This model is a fine-tuned version of t5-small on an Indonesia Novel dataset. It achieves the following results on the evaluation set:

• Loss: 2.5656
• Rouge1: 0.131
• Rouge2: 0.0393
• Rougel: 0.1193
• Rougelsum: 0.12
• Gen Len: 19.0

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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	10	2.9436	0.0943	0.0169	0.0838	0.0838	19.0
2.9817	2.0	20	2.8118	0.1009	0.0192	0.0894	0.0897	19.0
2.9817	3.0	30	2.7272	0.1062	0.0214	0.0934	0.0938	19.0
2.8136	4.0	40	2.6788	0.1086	0.0216	0.0962	0.0967	19.0
2.8136	5.0	50	2.6475	0.1107	0.0243	0.0961	0.0971	19.0
2.697	6.0	60	2.6258	0.1258	0.0312	0.1123	0.1132	19.0
2.697	7.0	70	2.6174	0.1245	0.0325	0.1108	0.1114	19.0
2.6577	8.0	80	2.6132	0.1314	0.037	0.1193	0.1198	19.0
2.6577	9.0	90	2.6049	0.128	0.0371	0.1173	0.1185	19.0
2.6347	10.0	100	2.5951	0.1294	0.0383	0.1195	0.1205	19.0
2.6347	11.0	110	2.5872	0.1294	0.0383	0.1195	0.1205	19.0
2.6063	12.0	120	2.5808	0.1296	0.0383	0.1195	0.1204	19.0
2.6063	13.0	130	2.5776	0.1296	0.0383	0.1195	0.1204	19.0
2.5979	14.0	140	2.5755	0.1302	0.0371	0.1186	0.1195	19.0
2.5979	15.0	150	2.5733	0.1302	0.0371	0.1186	0.1195	19.0
2.5841	16.0	160	2.5703	0.131	0.0393	0.1193	0.12	19.0
2.5841	17.0	170	2.5680	0.131	0.0393	0.1193	0.12	19.0
2.5672	18.0	180	2.5669	0.131	0.0393	0.1193	0.12	19.0
2.5672	19.0	190	2.5658	0.131	0.0393	0.1193	0.12	19.0
2.5716	20.0	200	2.5656	0.131	0.0393	0.1193	0.12	19.0

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0