youtube_comment_sentiment_kcbert_finetune

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

• Loss: 0.7068
• Accuracy: {'accuracy': 0.7691387559808612}
• F1 Score: {'f1': 0.7624618153347268}

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• 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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score
No log	1.0	74	0.7068	{'accuracy': 0.7691387559808612}	{'f1': 0.7624618153347268}
No log	2.0	148	0.7671	{'accuracy': 0.7787081339712919}	{'f1': 0.7808704180958316}
No log	3.0	222	0.8857	{'accuracy': 0.7942583732057417}	{'f1': 0.7945911253986201}
No log	4.0	296	0.9936	{'accuracy': 0.8038277511961722}	{'f1': 0.8042637284132086}
No log	5.0	370	1.0303	{'accuracy': 0.7990430622009569}	{'f1': 0.7997293347405633}

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1