distilbert-base-uncased-finetuned-pad-mult-clf-v2

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5603
• Accuracy: 0.8182
• F1: 0.7530

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: 256
• eval_batch_size: 256
• 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	F1
No log	1.0	3	1.1843	0.5397	0.3783
No log	2.0	6	1.0491	0.5397	0.3783
No log	3.0	9	0.9692	0.5397	0.3783
1.1406	4.0	12	0.8981	0.5745	0.4501
1.1406	5.0	15	0.8277	0.7292	0.6618
1.1406	6.0	18	0.7654	0.7834	0.7178
0.862	7.0	21	0.7125	0.8046	0.7388
0.862	8.0	24	0.6680	0.8143	0.7480
0.862	9.0	27	0.6395	0.8162	0.7499
0.6971	10.0	30	0.6132	0.8162	0.7499
0.6971	11.0	33	0.5999	0.8182	0.7523
0.6971	12.0	36	0.5766	0.8221	0.7556
0.6971	13.0	39	0.5637	0.8221	0.7561
0.6012	14.0	42	0.5607	0.8182	0.7529
0.6012	15.0	45	0.5603	0.8182	0.7530

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.1+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0