bert-finetuned-sentiment

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

• Loss: 0.6699
• Accuracy: 0.8085
• Precision: 0.8038
• Recall: 0.8085
• F1: 0.8056
• Confusion Matrix: [[1731, 10, 163], [22, 180, 76], [227, 80, 530]]

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: 32
• 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
• lr_scheduler_warmup_steps: 100
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Confusion Matrix
0.4894	1.0	340	0.5121	0.7939	0.7896	0.7939	0.7792	[[712, 8, 42], [10, 92, 9], [143, 37, 155]]
0.3577	2.0	680	0.5073	0.7947	0.7854	0.7947	0.7822	[[718, 6, 38], [10, 71, 30], [147, 17, 171]]
0.2353	3.0	1020	0.5490	0.7997	0.7943	0.7997	0.7955	[[682, 10, 70], [6, 86, 19], [113, 24, 198]]
0.167	4.0	1360	0.6284	0.7980	0.7912	0.7980	0.7924	[[695, 5, 62], [8, 69, 34], [119, 16, 200]]
0.1297	5.0	1700	0.6984	0.8046	0.7976	0.8046	0.7992	[[697, 6, 59], [6, 76, 29], [117, 19, 199]]

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.2
• Tokenizers 0.22.1