XMLRoberta_Lexical_Dataset59KCoDuoi

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

• Loss: 0.3668
• Accuracy: 0.9580
• F1: 0.9581

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.2558	200	0.2520	0.9109	0.9116
No log	0.5115	400	0.1839	0.9393	0.9398
No log	0.7673	600	0.2109	0.9362	0.9369
0.2271	1.0230	800	0.1567	0.9510	0.9512
0.2271	1.2788	1000	0.1477	0.9500	0.9502
0.2271	1.5345	1200	0.1551	0.9526	0.9529
0.2271	1.7903	1400	0.1419	0.9538	0.9542
0.1372	2.0460	1600	0.1607	0.9550	0.9554
0.1372	2.3018	1800	0.1590	0.9568	0.9568
0.1372	2.5575	2000	0.1415	0.9595	0.9596
0.1372	2.8133	2200	0.1473	0.9580	0.9582
0.1109	3.0691	2400	0.1644	0.9538	0.9542
0.1109	3.3248	2600	0.1300	0.9605	0.9607
0.1109	3.5806	2800	0.1664	0.9588	0.9591
0.1109	3.8363	3000	0.1395	0.9570	0.9572
0.0958	4.0921	3200	0.1602	0.9602	0.9603
0.0958	4.3478	3400	0.1566	0.9615	0.9616
0.0958	4.6036	3600	0.1413	0.9583	0.9586
0.0958	4.8593	3800	0.1973	0.9582	0.9582
0.083	5.1151	4000	0.1469	0.9591	0.9594
0.083	5.3708	4200	0.1541	0.9603	0.9605
0.083	5.6266	4400	0.1676	0.9585	0.9587
0.083	5.8824	4600	0.1687	0.9602	0.9604
0.0734	6.1381	4800	0.1865	0.9591	0.9592
0.0734	6.3939	5000	0.1723	0.9569	0.9569
0.0734	6.6496	5200	0.1761	0.9587	0.9589
0.0734	6.9054	5400	0.1596	0.9613	0.9614
0.0607	7.1611	5600	0.2193	0.9586	0.9588
0.0607	7.4169	5800	0.1984	0.9595	0.9596
0.0607	7.6726	6000	0.1745	0.9587	0.9589
0.0607	7.9284	6200	0.1939	0.9614	0.9615
0.0547	8.1841	6400	0.2081	0.9591	0.9592
0.0547	8.4399	6600	0.2048	0.9599	0.9601
0.0547	8.6957	6800	0.2260	0.9563	0.9565
0.0547	8.9514	7000	0.1786	0.9598	0.9600
0.047	9.2072	7200	0.2181	0.9596	0.9597
0.047	9.4629	7400	0.2120	0.9602	0.9603
0.047	9.7187	7600	0.2266	0.9597	0.9597
0.047	9.9744	7800	0.2128	0.9581	0.9583
0.0409	10.2302	8000	0.2207	0.9607	0.9608
0.0409	10.4859	8200	0.2375	0.9597	0.9599
0.0409	10.7417	8400	0.2241	0.9592	0.9593
0.0368	10.9974	8600	0.2181	0.9613	0.9613
0.0368	11.2532	8800	0.2574	0.9598	0.9599
0.0368	11.5090	9000	0.2598	0.9602	0.9602
0.0368	11.7647	9200	0.2448	0.9592	0.9594
0.0309	12.0205	9400	0.2521	0.9593	0.9594
0.0309	12.2762	9600	0.2824	0.9599	0.9601
0.0309	12.5320	9800	0.2606	0.9600	0.9602
0.0309	12.7877	10000	0.2841	0.9610	0.9612
0.0256	13.0435	10200	0.2662	0.9590	0.9591
0.0256	13.2992	10400	0.2839	0.9582	0.9582
0.0256	13.5550	10600	0.3053	0.9579	0.9580
0.0256	13.8107	10800	0.2697	0.9573	0.9574
0.0229	14.0665	11000	0.2741	0.9583	0.9584
0.0229	14.3223	11200	0.2881	0.9596	0.9597
0.0229	14.5780	11400	0.2921	0.9586	0.9588
0.0229	14.8338	11600	0.3162	0.9598	0.9600
0.0196	15.0895	11800	0.2989	0.9575	0.9576
0.0196	15.3453	12000	0.3267	0.9568	0.9570
0.0196	15.6010	12200	0.3113	0.9593	0.9594
0.0196	15.8568	12400	0.3198	0.9595	0.9597
0.0167	16.1125	12600	0.3355	0.9580	0.9582
0.0167	16.3683	12800	0.3525	0.9566	0.9569
0.0167	16.6240	13000	0.3337	0.9582	0.9584
0.0167	16.8798	13200	0.3105	0.9583	0.9585
0.0139	17.1355	13400	0.3348	0.9597	0.9599
0.0139	17.3913	13600	0.3290	0.9592	0.9593
0.0139	17.6471	13800	0.3476	0.9587	0.9589
0.0139	17.9028	14000	0.3498	0.9583	0.9584
0.0131	18.1586	14200	0.3483	0.9590	0.9590
0.0131	18.4143	14400	0.3386	0.9587	0.9588
0.0131	18.6701	14600	0.3512	0.9581	0.9582
0.0131	18.9258	14800	0.3627	0.9581	0.9582
0.01	19.1816	15000	0.3664	0.9572	0.9574
0.01	19.4373	15200	0.3688	0.9576	0.9578
0.01	19.6931	15400	0.3672	0.9579	0.9580
0.01	19.9488	15600	0.3668	0.9580	0.9581

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1