mt5-small-task3-dataset2

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5208
• Accuracy: 0.06
• Mse: 4.0312
• Log-distance: 0.6188
• S Score: 0.5264

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: 5.6e-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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Mse	Log-distance	S Score
10.5119	1.0	250	2.2161	0.016	4.4296	0.7765	0.4528
3.0365	2.0	500	1.7090	0.026	4.5503	0.7910	0.4512
2.2209	3.0	750	1.6007	0.052	4.7537	0.6721	0.4932
1.9292	4.0	1000	1.5895	0.042	4.1466	0.6578	0.5020
1.7982	5.0	1250	1.5695	0.052	4.7583	0.6732	0.4928
1.7379	6.0	1500	1.5367	0.046	4.2149	0.6615	0.5000
1.7081	7.0	1750	1.5376	0.054	4.1174	0.6606	0.5028
1.6768	8.0	2000	1.5462	0.054	4.1031	0.6584	0.5032
1.6515	9.0	2250	1.5256	0.052	4.1256	0.6525	0.5076
1.6235	10.0	2500	1.5512	0.052	4.1063	0.6542	0.5040
1.6289	11.0	2750	1.5346	0.06	4.1069	0.6390	0.5140
1.6077	12.0	3000	1.5385	0.058	4.0832	0.6298	0.5200
1.6014	13.0	3250	1.5204	0.058	3.9666	0.6236	0.5240
1.5998	14.0	3500	1.5201	0.06	4.0911	0.6142	0.5304
1.5994	15.0	3750	1.5208	0.06	4.0312	0.6188	0.5264

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Tokenizers 0.15.0