md_mt5_1911_v18_retrain

This model is a fine-tuned version of Buseak/md_mt5_1911_v16_deneme on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1754
• Bleu: 0.7623
• Gen Len: 18.7866

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.6365	1.0	1250	0.3435	0.6788	18.7694
0.5732	2.0	2500	0.3064	0.7037	18.7644
0.5375	3.0	3750	0.2819	0.7114	18.7706
0.4912	4.0	5000	0.2549	0.7237	18.77
0.4648	5.0	6250	0.2394	0.7354	18.772
0.4321	6.0	7500	0.2245	0.7335	18.7762
0.4159	7.0	8750	0.2131	0.7446	18.778
0.4044	8.0	10000	0.2030	0.7478	18.7776
0.3889	9.0	11250	0.1963	0.7496	18.7852
0.3798	10.0	12500	0.1896	0.7524	18.7834
0.3733	11.0	13750	0.1836	0.757	18.7854
0.3623	12.0	15000	0.1803	0.7596	18.7852
0.3583	13.0	16250	0.1775	0.7618	18.7878
0.3643	14.0	17500	0.1758	0.7616	18.7828
0.3609	15.0	18750	0.1754	0.7623	18.7866

Framework versions

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