mbart-en-id-smaller-indo-amr-parsing-translated-nafkhan

This model was trained from scratch on the data dataset. It achieves the following results on the evaluation set:

• Loss: 1.1691
• Smatch: 0.8299
• Gen Len: 27.5333

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: 1e-06
• train_batch_size: 5
• eval_batch_size: 5
• seed: 42
• gradient_accumulation_steps: 5
• total_train_batch_size: 25
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: polynomial
• lr_scheduler_warmup_steps: 200
• num_epochs: 640.0
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Smatch	Gen Len
0.9755	39.7351	3600	1.1801	0.7484	27.3333
0.9556	79.4702	7200	1.1545	0.7573	27.0667
0.8935	119.2053	10800	1.1454	0.7924	26.6667
0.8823	158.9404	14400	1.1351	0.7891	27.3333
0.8514	198.6755	18000	1.1308	0.828	26.6
0.8385	238.4106	21600	1.1385	0.8235	27.1
0.836	278.1457	25200	1.1407	0.8399	27.5
0.8503	317.8808	28800	1.1352	0.822	26.6
0.8349	357.6159	32400	1.1478	0.8326	27.2333
0.8296	397.3510	36000	1.1506	0.8201	27.3
0.871	437.0861	39600	1.1592	0.7992	28.3
0.834	476.8212	43200	1.1534	0.825	27.5667
0.8332	516.5563	46800	1.1615	0.8225	27.4
0.8513	556.2914	50400	1.1663	0.8417	27.3667
0.8249	596.0265	54000	1.1667	0.8358	27.4
0.8439	635.7616	57600	1.1691	0.8299	27.5333

Framework versions

• Transformers 4.44.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1