MSLacerda/attribute_mining_mslacerda

attribute_minig_mslacerda

This model is a fine-tuned version of distilbert/distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5688
• Precision: 0.7424
• Recall: 0.7766
• F1: 0.7591
• Accuracy: 0.8673

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: 4.546321141328063e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 207
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	85	4.0840	0.0271	0.0229	0.0248	0.1523
No log	2.0	170	2.1874	0.3096	0.2348	0.2671	0.5099
No log	3.0	255	1.2897	0.4893	0.4684	0.4786	0.6940
No log	4.0	340	0.8647	0.6485	0.6413	0.6449	0.8027
No log	5.0	425	0.6897	0.6977	0.7051	0.7014	0.8414
1.93	6.0	510	0.6046	0.7177	0.7354	0.7264	0.8550
1.93	7.0	595	0.5833	0.7339	0.7602	0.7468	0.8628
1.93	8.0	680	0.5722	0.7474	0.7627	0.7550	0.8643
1.93	9.0	765	0.5704	0.7451	0.7695	0.7571	0.8694
1.93	10.0	850	0.5802	0.7485	0.7763	0.7622	0.8715
1.93	11.0	935	0.5723	0.7539	0.7838	0.7685	0.8754
0.1632	12.0	1020	0.5736	0.7497	0.7813	0.7652	0.8739

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1