bert-prodcat

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

• Loss: 0.7061
• Accuracy: 0.8372
• F1 Macro: 0.7861

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: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro
0.3823	0.2912	500	0.8749	0.7949	0.7250
0.5849	0.5824	1000	0.8225	0.8090	0.7437
0.6131	0.8736	1500	0.7716	0.8175	0.7532
0.4162	1.1648	2000	0.7270	0.8375	0.7784
0.4578	1.4560	2500	0.7239	0.8342	0.7761
0.4166	1.7472	3000	0.7061	0.8372	0.7861

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1