ConPlag_1_codeberta_ep30_bs16_lr1e-05_l512_s42_ppy_f_beta_score_2

This model is a fine-tuned version of huggingface/CodeBERTa-small-v1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6386
• Accuracy: 0.9197
• Recall: 0.8333
• Precision: 0.8974
• F1: 0.8642
• F Beta Score: 0.8521

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: 1e-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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1	F Beta Score
No log	1.0	40	0.6090	0.7737	0.7381	0.6078	0.6667	0.6924
No log	2.0	80	0.5145	0.7372	0.9048	0.5429	0.6786	0.7508
0.6234	3.0	120	0.5056	0.8467	0.5714	0.8889	0.6957	0.6420
0.6234	4.0	160	0.4510	0.8832	0.6667	0.9333	0.7778	0.7309
0.3658	5.0	200	0.4820	0.8686	0.6190	0.9286	0.7429	0.6898
0.3658	6.0	240	0.4417	0.9051	0.7143	0.9677	0.8219	0.7769
0.3658	7.0	280	0.4399	0.8759	0.8571	0.7660	0.8090	0.8269
0.2126	8.0	320	0.5597	0.8978	0.7143	0.9375	0.8108	0.7708
0.2126	9.0	360	0.4830	0.9343	0.8333	0.9459	0.8861	0.8650
0.1306	10.0	400	0.5402	0.9270	0.8095	0.9444	0.8718	0.8467
0.1306	11.0	440	0.6326	0.9197	0.7857	0.9429	0.8571	0.8282
0.1306	12.0	480	0.5699	0.8978	0.8571	0.8182	0.8372	0.8448
0.0878	13.0	520	0.7407	0.8978	0.7619	0.8889	0.8205	0.7969
0.0878	14.0	560	0.6386	0.9197	0.8333	0.8974	0.8642	0.8521

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3