README.md

metadata

library_name: transformers
license: mit
base_model: nielsr/lilt-xlm-roberta-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: 100epoch_test
    results: []

100epoch_test

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

• Loss: 0.3027
• Precision: 0.9074
• Recall: 0.9128
• F1: 0.9101
• Accuracy: 0.9717

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.7937	100	0.1798	0.8560	0.8877	0.8716	0.9565
No log	1.5873	200	0.1305	0.8982	0.9032	0.9007	0.9679
No log	2.3810	300	0.1216	0.9019	0.9290	0.9152	0.9719
No log	3.1746	400	0.1334	0.8883	0.9062	0.8971	0.9676
0.2125	3.9683	500	0.1163	0.9103	0.9191	0.9147	0.9726
0.2125	4.7619	600	0.1171	0.9172	0.9002	0.9086	0.9712
0.2125	5.5556	700	0.1343	0.8646	0.9006	0.8822	0.9635
0.2125	6.3492	800	0.1397	0.9085	0.8956	0.9020	0.9690
0.2125	7.1429	900	0.1437	0.9166	0.8969	0.9067	0.9704
0.0606	7.9365	1000	0.1241	0.9075	0.9138	0.9106	0.9716
0.0606	8.7302	1100	0.1372	0.9112	0.9088	0.9100	0.9710
0.0606	9.5238	1200	0.1403	0.9151	0.9154	0.9153	0.9735
0.0606	10.3175	1300	0.1783	0.9169	0.9039	0.9103	0.9720
0.0606	11.1111	1400	0.1517	0.9148	0.9151	0.9149	0.9735
0.0374	11.9048	1500	0.1677	0.9148	0.9042	0.9095	0.9712
0.0374	12.6984	1600	0.1679	0.9178	0.9006	0.9091	0.9711
0.0374	13.4921	1700	0.1569	0.9178	0.8996	0.9086	0.9712
0.0374	14.2857	1800	0.2076	0.9066	0.8916	0.8991	0.9677
0.0374	15.0794	1900	0.1684	0.9001	0.9201	0.9100	0.9712
0.0231	15.8730	2000	0.1945	0.9163	0.9108	0.9135	0.9725
0.0231	16.6667	2100	0.1707	0.9099	0.9072	0.9085	0.9715
0.0231	17.4603	2200	0.2078	0.9167	0.9019	0.9092	0.9715
0.0231	18.2540	2300	0.2086	0.9068	0.9032	0.9050	0.9700
0.0231	19.0476	2400	0.2082	0.9110	0.8996	0.9053	0.9705
0.0155	19.8413	2500	0.1908	0.9109	0.9121	0.9115	0.9719
0.0155	20.6349	2600	0.1991	0.9094	0.9118	0.9106	0.9712
0.0155	21.4286	2700	0.2008	0.9053	0.9065	0.9059	0.9702
0.0155	22.2222	2800	0.2344	0.9097	0.9125	0.9111	0.9720
0.0155	23.0159	2900	0.2186	0.9076	0.9115	0.9095	0.9717
0.0097	23.8095	3000	0.2208	0.9045	0.9141	0.9093	0.9716
0.0097	24.6032	3100	0.1988	0.9033	0.9039	0.9036	0.9700
0.0097	25.3968	3200	0.2291	0.9197	0.9009	0.9102	0.9712
0.0097	26.1905	3300	0.2402	0.9011	0.9025	0.9018	0.9697
0.0097	26.9841	3400	0.2418	0.9095	0.9131	0.9113	0.9715
0.0066	27.7778	3500	0.2149	0.8997	0.9065	0.9031	0.9697
0.0066	28.5714	3600	0.2474	0.9016	0.9022	0.9019	0.9696
0.0066	29.3651	3700	0.2362	0.9143	0.9055	0.9099	0.9720
0.0066	30.1587	3800	0.2374	0.9058	0.9184	0.9121	0.9719
0.0066	30.9524	3900	0.2516	0.9032	0.9006	0.9019	0.9692
0.0048	31.7460	4000	0.2251	0.9038	0.9092	0.9065	0.9710
0.0048	32.5397	4100	0.2488	0.9062	0.9098	0.9080	0.9709
0.0048	33.3333	4200	0.2412	0.9034	0.9088	0.9061	0.9710
0.0048	34.1270	4300	0.2421	0.9041	0.9032	0.9037	0.9697
0.0048	34.9206	4400	0.2599	0.9038	0.9062	0.9050	0.9704
0.0042	35.7143	4500	0.2372	0.9002	0.9144	0.9072	0.9706
0.0042	36.5079	4600	0.2545	0.9020	0.9035	0.9028	0.9696
0.0042	37.3016	4700	0.2629	0.9034	0.8963	0.8998	0.9684
0.0042	38.0952	4800	0.2407	0.9069	0.9078	0.9074	0.9706
0.0042	38.8889	4900	0.2604	0.9115	0.9019	0.9067	0.9706
0.0039	39.6825	5000	0.2657	0.9091	0.9052	0.9071	0.9704
0.0039	40.4762	5100	0.2615	0.8995	0.9049	0.9022	0.9690
0.0039	41.2698	5200	0.2631	0.9115	0.9016	0.9065	0.9711
0.0039	42.0635	5300	0.2645	0.9087	0.9105	0.9096	0.9717
0.0039	42.8571	5400	0.2736	0.9027	0.9042	0.9034	0.9701
0.0031	43.6508	5500	0.2491	0.9064	0.9052	0.9058	0.9702
0.0031	44.4444	5600	0.2556	0.9110	0.9095	0.9102	0.9712
0.0031	45.2381	5700	0.2768	0.9009	0.9009	0.9009	0.9689
0.0031	46.0317	5800	0.2580	0.9054	0.9045	0.9050	0.9702
0.0031	46.8254	5900	0.2524	0.9047	0.9068	0.9058	0.9709
0.003	47.6190	6000	0.2652	0.9083	0.9101	0.9092	0.9717
0.003	48.4127	6100	0.2741	0.9082	0.9148	0.9115	0.9717
0.003	49.2063	6200	0.2835	0.9065	0.9062	0.9063	0.9706
0.003	50.0	6300	0.2906	0.9011	0.9058	0.9035	0.9699
0.003	50.7937	6400	0.2738	0.9060	0.9101	0.9080	0.9709
0.0019	51.5873	6500	0.2730	0.9088	0.9085	0.9086	0.9715
0.0019	52.3810	6600	0.2718	0.9042	0.9012	0.9027	0.9696
0.0019	53.1746	6700	0.2862	0.9101	0.9101	0.9101	0.9712
0.0019	53.9683	6800	0.2816	0.8985	0.9095	0.9040	0.9695
0.0019	54.7619	6900	0.2931	0.9051	0.9016	0.9033	0.9697
0.0017	55.5556	7000	0.2644	0.9082	0.9082	0.9082	0.9710
0.0017	56.3492	7100	0.2815	0.9089	0.9068	0.9079	0.9715
0.0017	57.1429	7200	0.2566	0.9044	0.9068	0.9056	0.9709
0.0017	57.9365	7300	0.2709	0.9130	0.9088	0.9109	0.9722
0.0017	58.7302	7400	0.2699	0.9089	0.9092	0.9090	0.9715
0.0016	59.5238	7500	0.2742	0.9084	0.9072	0.9078	0.9707
0.0016	60.3175	7600	0.2549	0.9062	0.9101	0.9082	0.9714
0.0016	61.1111	7700	0.2714	0.9068	0.8963	0.9015	0.9695
0.0016	61.9048	7800	0.2801	0.9098	0.9062	0.9080	0.9715
0.0016	62.6984	7900	0.2818	0.9006	0.9072	0.9039	0.9705
0.0013	63.4921	8000	0.2923	0.9053	0.9068	0.9061	0.9711
0.0013	64.2857	8100	0.2944	0.9068	0.8900	0.8983	0.9686
0.0013	65.0794	8200	0.2941	0.9033	0.9075	0.9054	0.9706
0.0013	65.8730	8300	0.2801	0.9075	0.9006	0.9040	0.9699
0.0013	66.6667	8400	0.2822	0.9098	0.9131	0.9115	0.9722
0.0008	67.4603	8500	0.3013	0.9066	0.9016	0.9041	0.9701
0.0008	68.2540	8600	0.2670	0.9040	0.9144	0.9092	0.9719
0.0008	69.0476	8700	0.2941	0.9054	0.9012	0.9033	0.9701
0.0008	69.8413	8800	0.2911	0.9086	0.9065	0.9076	0.9717
0.0008	70.6349	8900	0.2783	0.9123	0.9111	0.9117	0.9726
0.0007	71.4286	9000	0.2877	0.9122	0.9022	0.9072	0.9715
0.0007	72.2222	9100	0.3021	0.9019	0.9138	0.9078	0.9706
0.0007	73.0159	9200	0.2869	0.9094	0.9118	0.9106	0.9721
0.0007	73.8095	9300	0.2928	0.9041	0.9095	0.9068	0.9706
0.0007	74.6032	9400	0.2896	0.9088	0.9088	0.9088	0.9716
0.0007	75.3968	9500	0.3008	0.9073	0.9118	0.9095	0.9716
0.0007	76.1905	9600	0.3019	0.9067	0.9082	0.9074	0.9710
0.0007	76.9841	9700	0.2924	0.9073	0.9148	0.9110	0.9717
0.0007	77.7778	9800	0.2856	0.9117	0.9138	0.9127	0.9726
0.0007	78.5714	9900	0.2924	0.9098	0.9098	0.9098	0.9719
0.0004	79.3651	10000	0.3100	0.9047	0.9121	0.9084	0.9715
0.0004	80.1587	10100	0.3055	0.9100	0.9082	0.9091	0.9719
0.0004	80.9524	10200	0.2990	0.9103	0.9125	0.9114	0.9725
0.0004	81.7460	10300	0.2980	0.9099	0.9039	0.9069	0.9714
0.0004	82.5397	10400	0.2954	0.9095	0.9128	0.9111	0.9724
0.0003	83.3333	10500	0.2993	0.9092	0.9125	0.9108	0.9722
0.0003	84.1270	10600	0.3036	0.9079	0.9118	0.9098	0.9720
0.0003	84.9206	10700	0.2919	0.9088	0.9121	0.9105	0.9722
0.0003	85.7143	10800	0.2948	0.9075	0.9105	0.9090	0.9719
0.0003	86.5079	10900	0.3037	0.9081	0.9078	0.9080	0.9717
0.0003	87.3016	11000	0.3039	0.9086	0.9125	0.9105	0.9724
0.0003	88.0952	11100	0.3019	0.9088	0.9115	0.9101	0.9721
0.0003	88.8889	11200	0.3064	0.9108	0.9111	0.9110	0.9724
0.0003	89.6825	11300	0.3017	0.9088	0.9115	0.9101	0.9721
0.0003	90.4762	11400	0.2813	0.9074	0.9098	0.9086	0.9717
0.0005	91.2698	11500	0.2895	0.9081	0.9078	0.9080	0.9716
0.0005	92.0635	11600	0.2950	0.9065	0.9098	0.9082	0.9716
0.0005	92.8571	11700	0.2880	0.9074	0.9134	0.9104	0.9720
0.0005	93.6508	11800	0.2947	0.9066	0.9134	0.9100	0.9717
0.0005	94.4444	11900	0.3003	0.9050	0.9068	0.9059	0.9706
0.0002	95.2381	12000	0.3018	0.9064	0.9115	0.9089	0.9714
0.0002	96.0317	12100	0.3008	0.9071	0.9131	0.9101	0.9717
0.0002	96.8254	12200	0.3011	0.9071	0.9131	0.9101	0.9717
0.0002	97.6190	12300	0.3007	0.9077	0.9134	0.9106	0.9719
0.0002	98.4127	12400	0.3017	0.9077	0.9134	0.9106	0.9719
0.0002	99.2063	12500	0.3024	0.9074	0.9128	0.9101	0.9717
0.0002	100.0	12600	0.3027	0.9074	0.9128	0.9101	0.9717

Framework versions

• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.1