indosum-base-1

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: 0.7478
• Rouge1: 72.3325
• Rouge2: 65.1065
• Rougel: 69.2256
• Rougelsum: 71.5325
• Gen Len: 99.4045

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
1.1904	1.0	892	0.8053	65.8257	57.6167	62.6222	65.0027	95.8598
0.6851	2.0	1784	0.6779	67.8889	60.0878	64.5868	66.9914	96.2911
0.4856	3.0	2676	0.6460	70.9241	63.6363	67.8555	70.153	96.9212
0.3358	4.0	3568	0.6565	69.9002	62.4	66.5928	69.0347	101.8745
0.1973	5.0	4460	0.7478	72.0587	64.7973	68.9279	71.3028	99.3765

Framework versions

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