outputs

This model is a fine-tuned version of microsoft/deberta-v3-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0041
• Pearson: 0.9845

Model description

O modelo verifica se a mensagem é spam o não. Caso o valor seja maior ou igual a 0.6 ele é spam, caso seja menor ele não é spam.

Aqui temos algumas mensagens do dataframe de teste:

• Send a logo 2 ur lover - 2 names joined by a heart. Txt LOVE NAME1 NAME2 MOBNO eg LOVE ADAM EVE 07123456789 to 87077 Yahoo! POBox36504W45WQ TxtNO 4 no ads 150p | Spam
• Not directly behind... Abt 4 rows behind ü... | Non-Spam

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 8e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Pearson
No log	1.0	24	0.0338	0.9274
No log	2.0	48	0.0070	0.9667
No log	3.0	72	0.0110	0.9504
No log	4.0	96	0.0078	0.9634

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3