100epoch_test_march19

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: 0.3000
• Precision: 0.9344
• Recall: 0.9317
• F1: 0.9331
• Accuracy: 0.9725

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	Accuracy	F1	Validation Loss	Precision	Recall
No log	0.7937	100	0.9349	0.8465	0.2505	0.8337	0.8596
No log	1.5873	200	0.9581	0.8996	0.1602	0.8918	0.9075
No log	2.3810	300	0.9663	0.9162	0.1457	0.8993	0.9336
No log	3.1746	400	0.9682	0.9204	0.1426	0.9291	0.9119
0.2582	3.9683	500	0.9675	0.9205	0.1340	0.9289	0.9123
0.2582	4.7619	600	0.9714	0.9300	0.1387	0.9318	0.9281
0.2582	5.5556	700	0.9702	0.9284	0.1456	0.9283	0.9285
0.2582	6.3492	800	0.9696	0.9247	0.1438	0.9278	0.9216
0.2582	7.1429	900	0.9687	0.9218	0.1592	0.9349	0.9090
0.0527	7.9365	1000	0.9714	0.9280	0.1470	0.9255	0.9306
0.0527	8.7302	1100	0.9706	0.9286	0.1570	0.9322	0.9250
0.0527	9.5238	1200	0.9718	0.9301	0.1627	0.9295	0.9308
0.0527	10.3175	1300	0.9687	0.9215	0.1997	0.9200	0.9229
0.0527	11.1111	1400	0.9723	0.9330	0.1864	0.9296	0.9365
0.0296	11.9048	1500	0.9699	0.9249	0.1912	0.9277	0.9222
0.0296	12.6984	1600	0.9710	0.9282	0.1834	0.9288	0.9277
0.0296	13.4921	1700	0.9696	0.9264	0.2040	0.9275	0.9252
0.0296	14.2857	1800	0.9686	0.9260	0.2004	0.9315	0.9207
0.0296	15.0794	1900	0.9689	0.9229	0.1982	0.9233	0.9226
0.0182	15.8730	2000	0.9700	0.9262	0.2130	0.9259	0.9266
0.0182	16.6667	2100	0.9708	0.9274	0.1958	0.9297	0.9252
0.0182	17.4603	2200	0.9693	0.9233	0.2237	0.9298	0.9168
0.0182	18.2540	2300	0.9691	0.9245	0.2224	0.9276	0.9214
0.0182	19.0476	2400	0.9728	0.9333	0.2082	0.9359	0.9308
0.0138	19.8413	2500	0.9709	0.9284	0.2172	0.9267	0.9302
0.0138	20.6349	2600	0.9718	0.9313	0.1997	0.9347	0.9279
0.0138	21.4286	2700	0.9687	0.9240	0.2395	0.9297	0.9184
0.0138	22.2222	2800	0.9698	0.9254	0.2342	0.9251	0.9256
0.0138	23.0159	2900	0.9709	0.9281	0.2079	0.9311	0.9250
0.0092	23.8095	3000	0.9693	0.9245	0.2388	0.9219	0.9271
0.0092	24.6032	3100	0.9687	0.9237	0.2426	0.9190	0.9285
0.0092	25.3968	3200	0.9706	0.9268	0.2179	0.9299	0.9237
0.0092	26.1905	3300	0.9704	0.9259	0.2348	0.9278	0.9241
0.0092	26.9841	3400	0.9696	0.9257	0.2414	0.9323	0.9191
0.0062	27.7778	3500	0.9693	0.9244	0.2320	0.9232	0.9256
0.0062	28.5714	3600	0.9700	0.9246	0.2327	0.9227	0.9266
0.0062	29.3651	3700	0.9691	0.9245	0.2459	0.9232	0.9258
0.0062	30.1587	3800	0.9709	0.9279	0.2185	0.9304	0.9254
0.0062	30.9524	3900	0.9719	0.9290	0.2355	0.9305	0.9275
0.0045	31.7460	4000	0.9725	0.9316	0.2271	0.9324	0.9308
0.0045	32.5397	4100	0.9710	0.9270	0.2464	0.9265	0.9275
0.0045	33.3333	4200	0.9693	0.9253	0.2522	0.9254	0.9252
0.0045	34.1270	4300	0.9693	0.9259	0.2315	0.9217	0.9300
0.0045	34.9206	4400	0.9700	0.9258	0.2301	0.9261	0.9254
0.0039	35.7143	4500	0.9694	0.9242	0.2511	0.9291	0.9193
0.0039	36.5079	4600	0.9690	0.9236	0.2425	0.9232	0.9241
0.0039	37.3016	4700	0.9664	0.9167	0.2968	0.9257	0.9079
0.0039	38.0952	4800	0.9684	0.9223	0.2725	0.9201	0.9245
0.0039	38.8889	4900	0.9690	0.9225	0.2881	0.9252	0.9199
0.0032	39.6825	5000	0.9678	0.9240	0.2804	0.9231	0.9249
0.0032	40.4762	5100	0.9687	0.9248	0.2689	0.9297	0.9201
0.0032	41.2698	5200	0.9706	0.9276	0.2727	0.9256	0.9296
0.0032	42.0635	5300	0.9714	0.9293	0.2650	0.9299	0.9287
0.0032	42.8571	5400	0.9699	0.9261	0.2633	0.9292	0.9231
0.003	43.6508	5500	0.9710	0.9289	0.2668	0.9327	0.9250
0.003	44.4444	5600	0.9701	0.9281	0.2654	0.9314	0.9249
0.003	45.2381	5700	0.9704	0.9275	0.2701	0.9351	0.9201
0.003	46.0317	5800	0.9730	0.9332	0.2541	0.9329	0.9334
0.003	46.8254	5900	0.9715	0.9298	0.2486	0.9310	0.9287
0.0026	47.6190	6000	0.9726	0.9318	0.2248	0.9392	0.9245
0.0026	48.4127	6100	0.9716	0.9305	0.2562	0.9297	0.9313
0.0026	49.2063	6200	0.9730	0.9341	0.2648	0.9345	0.9336
0.0026	50.0	6300	0.9714	0.9296	0.2561	0.9308	0.9285
0.0026	50.7937	6400	0.9733	0.9331	0.2531	0.9359	0.9304
0.0014	51.5873	6500	0.9712	0.9290	0.2676	0.9356	0.9226
0.0014	52.3810	6600	0.9680	0.9204	0.2842	0.9276	0.9134
0.0014	53.1746	6700	0.9717	0.9311	0.2731	0.9274	0.9350
0.0014	53.9683	6800	0.9719	0.9306	0.2665	0.9319	0.9292
0.0014	54.7619	6900	0.9719	0.9307	0.2690	0.9294	0.9319
0.0017	55.5556	7000	0.9727	0.9319	0.2668	0.9326	0.9311
0.0017	56.3492	7100	0.9707	0.9268	0.2867	0.9219	0.9317
0.0017	57.1429	7200	0.9721	0.9324	0.2771	0.9306	0.9342
0.0017	57.9365	7300	0.9723	0.9336	0.2570	0.9372	0.9300
0.0017	58.7302	7400	0.9735	0.9348	0.2649	0.9363	0.9332
0.0008	59.5238	7500	0.9724	0.9327	0.2780	0.9325	0.9329
0.0008	60.3175	7600	0.9706	0.9270	0.2728	0.9263	0.9277
0.0008	61.1111	7700	0.9718	0.9315	0.2703	0.9268	0.9363
0.0008	61.9048	7800	0.9715	0.9297	0.2799	0.9288	0.9306
0.0008	62.6984	7900	0.9709	0.9275	0.2793	0.9233	0.9317
0.0018	63.4921	8000	0.9716	0.9296	0.2685	0.9271	0.9321
0.0018	64.2857	8100	0.9715	0.9301	0.2779	0.9293	0.9308
0.0018	65.0794	8200	0.9725	0.9316	0.2849	0.9314	0.9319
0.0018	65.8730	8300	0.9712	0.9291	0.2952	0.9246	0.9336
0.0018	66.6667	8400	0.9717	0.9308	0.2932	0.9250	0.9367
0.0008	67.4603	8500	0.9727	0.9320	0.2614	0.9339	0.9300
0.0008	68.2540	8600	0.9723	0.9305	0.2751	0.9311	0.9300
0.0008	69.0476	8700	0.9724	0.9306	0.2829	0.9351	0.9262
0.0008	69.8413	8800	0.9726	0.9318	0.2835	0.9294	0.9342
0.0008	70.6349	8900	0.9720	0.9302	0.2687	0.9284	0.9321
0.0007	71.4286	9000	0.9718	0.9291	0.2631	0.9294	0.9289
0.0007	72.2222	9100	0.9727	0.9316	0.2608	0.9324	0.9308
0.0007	73.0159	9200	0.9728	0.9329	0.2705	0.9307	0.9352
0.0007	73.8095	9300	0.9724	0.9319	0.2822	0.9288	0.9350
0.0007	74.6032	9400	0.9739	0.9351	0.2753	0.9361	0.9340
0.0006	75.3968	9500	0.9719	0.9304	0.2880	0.9329	0.9279
0.0006	76.1905	9600	0.9726	0.9314	0.2954	0.9290	0.9338
0.0006	76.9841	9700	0.9731	0.9334	0.2967	0.9343	0.9325
0.0006	77.7778	9800	0.9733	0.9339	0.3013	0.9327	0.9352
0.0006	78.5714	9900	0.9740	0.9355	0.2980	0.9360	0.9350
0.0002	79.3651	10000	0.9735	0.9344	0.3020	0.9334	0.9353
0.0002	80.1587	10100	0.2857	0.9368	0.9357	0.9363	0.9736
0.0002	80.9524	10200	0.2948	0.9214	0.9346	0.9279	0.9712
0.0002	81.7460	10300	0.2918	0.9285	0.9270	0.9277	0.9711
0.0002	82.5397	10400	0.2947	0.9268	0.9323	0.9295	0.9718
0.0003	83.3333	10500	0.2976	0.9300	0.9323	0.9311	0.9724
0.0003	84.1270	10600	0.2980	0.9302	0.9310	0.9306	0.9722
0.0003	84.9206	10700	0.2978	0.9310	0.9344	0.9327	0.9729
0.0003	85.7143	10800	0.2882	0.9318	0.9327	0.9322	0.9727
0.0003	86.5079	10900	0.2918	0.9314	0.9319	0.9316	0.9725
0.0003	87.3016	11000	0.2932	0.9305	0.9325	0.9315	0.9724
0.0003	88.0952	11100	0.2964	0.9305	0.9323	0.9314	0.9723
0.0003	88.8889	11200	0.3069	0.9307	0.9294	0.9301	0.9716
0.0003	89.6825	11300	0.3022	0.9309	0.9321	0.9315	0.9724
0.0003	90.4762	11400	0.3088	0.9330	0.9289	0.9309	0.9715
0.0002	91.2698	11500	0.3008	0.9323	0.9292	0.9307	0.9715
0.0002	92.0635	11600	0.3019	0.9314	0.9304	0.9309	0.9723
0.0002	92.8571	11700	0.3039	0.9309	0.9302	0.9305	0.9721
0.0002	93.6508	11800	0.3017	0.9345	0.9287	0.9316	0.9722
0.0002	94.4444	11900	0.2980	0.9340	0.9311	0.9326	0.9724
0.0003	95.2381	12000	0.2973	0.9346	0.9302	0.9324	0.9725
0.0003	96.0317	12100	0.2981	0.9345	0.9304	0.9324	0.9724
0.0003	96.8254	12200	0.2990	0.9337	0.9292	0.9315	0.9721
0.0003	97.6190	12300	0.2993	0.9337	0.9298	0.9318	0.9721
0.0003	98.4127	12400	0.2993	0.9343	0.9310	0.9326	0.9724
0.0001	99.2063	12500	0.2998	0.9340	0.9315	0.9328	0.9724
0.0001	100.0	12600	0.3000	0.9344	0.9317	0.9331	0.9725

Framework versions

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