indosum-seq_bn-rf16-0

This model is a fine-tuned version of LazarusNLP/IndoNanoT5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4868
• Rouge1: 73.4412
• Rouge2: 66.5198
• Rougel: 70.4596
• Rougelsum: 72.5706
• Gen Len: 103.0

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: 0.001
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.7747	1.0	892	0.5280	70.368	63.1479	67.3463	69.4649	100.8813
0.6074	2.0	1784	0.5154	71.3426	64.1849	68.288	70.4796	104.6013
0.5576	3.0	2676	0.4920	71.6471	64.5469	68.6561	70.7418	101.8973
0.5199	4.0	3568	0.4931	72.695	65.8499	69.8428	71.8427	103.6333
0.4895	5.0	4460	0.4868	73.1452	66.2201	70.1397	72.2952	102.3827

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1