bertweetB_15epoch

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

• Loss: 0.1645
• Accuracy: 0.77
• Precision: 0.2476
• Recall: 0.3173
• F1: 0.2757

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
• 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	Precision	Recall	F1
No log	1.0	217	0.1306	0.8571	0.0	0.0	0.0
No log	2.0	434	0.1295	0.8571	0.0	0.0	0.0
0.1937	3.0	651	0.1268	0.8571	0.0	0.0	0.0
0.1937	4.0	868	0.1227	0.8593	0.3712	0.0701	0.1179
0.1473	5.0	1085	0.1307	0.765	0.2292	0.4354	0.3003
0.1473	6.0	1302	0.1270	0.7964	0.2457	0.3469	0.2877
0.1018	7.0	1519	0.1398	0.7607	0.2276	0.4354	0.2978
0.1018	8.0	1736	0.1449	0.7821	0.2323	0.3506	0.2786
0.1018	9.0	1953	0.1408	0.7843	0.2681	0.3764	0.3127
0.0648	10.0	2170	0.1535	0.78	0.2455	0.2878	0.2634
0.0648	11.0	2387	0.1585	0.7593	0.2375	0.3911	0.2954
0.0396	12.0	2604	0.1591	0.7757	0.2642	0.3100	0.2809
0.0396	13.0	2821	0.1670	0.7614	0.2347	0.3432	0.2774
0.0284	14.0	3038	0.1623	0.7793	0.2561	0.3026	0.2745
0.0284	15.0	3255	0.1645	0.77	0.2476	0.3173	0.2757

Framework versions

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