roberta-base-downstream-build_rr

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

• Loss: 0.8610
• Precision-macro: 0.6015
• Recall-macro: 0.5642
• Macro-f1: 0.5742
• Precision-micro: 0.7871
• Recall-micro: 0.7871
• Micro-f1: 0.7871
• Accuracy: 0.7871

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 1
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision-macro	Recall-macro	Macro-f1	Precision-micro	Recall-micro	Micro-f1	Accuracy
No log	1.0	124	0.9703	0.5485	0.3447	0.3566	0.7155	0.7155	0.7155	0.7155
No log	2.0	248	0.8005	0.5181	0.5222	0.5080	0.7353	0.7353	0.7353	0.7353
No log	3.0	372	0.8156	0.5626	0.5322	0.5288	0.7454	0.7454	0.7454	0.7454
No log	4.0	496	0.7056	0.5881	0.5197	0.5180	0.7704	0.7704	0.7704	0.7704
1.0549	5.0	620	0.7526	0.5878	0.5906	0.5775	0.7642	0.7642	0.7642	0.7642
1.0549	6.0	744	0.7094	0.6336	0.5395	0.5649	0.7812	0.7812	0.7812	0.7812
1.0549	7.0	868	0.7391	0.6475	0.5339	0.5535	0.7808	0.7808	0.7808	0.7808
1.0549	8.0	992	0.7354	0.6169	0.5756	0.5881	0.7930	0.7930	0.7930	0.7930
0.545	9.0	1116	0.8143	0.5951	0.5963	0.5928	0.7805	0.7805	0.7805	0.7805
0.545	10.0	1240	0.8352	0.6029	0.5915	0.5918	0.7794	0.7794	0.7794	0.7794
0.545	11.0	1364	0.8610	0.6015	0.5642	0.5742	0.7871	0.7871	0.7871	0.7871

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1