ht-stmpnet-cls-v5_ftis_noPretrain

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

• Loss: 3.0546
• Accuracy: 0.9066
• Macro F1: 0.7603

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 6731
• training_steps: 134625

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
57.2411	0.0010	134	48.1282	0.1502	0.0407
26.3117	1.0010	268	134.1827	0.2457	0.0627
7.8677	2.0010	402	298.0198	0.5171	0.1341
6.4453	3.0010	536	241.7373	0.5618	0.1465
5.7397	4.0010	670	247.9332	0.5857	0.1514
4.9305	5.0010	804	215.6535	0.6001	0.1522
4.3836	6.0010	938	147.2826	0.5896	0.1585
3.6308	7.0009	1072	97.0444	0.6152	0.1682
2.9268	8.0009	1206	72.0529	0.6139	0.1697
2.7668	9.0009	1340	53.0436	0.5893	0.1672
2.4926	10.0009	1474	38.8328	0.6293	0.1739
2.4013	11.0009	1608	27.0263	0.6272	0.1815
2.3357	12.0009	1742	22.0645	0.6472	0.1957
2.2034	13.0009	1876	16.5040	0.6499	0.2254
2.0563	14.0009	2010	13.6849	0.6659	0.2370
2.0094	15.0009	2144	10.8709	0.6706	0.2457
2.1506	16.0009	2278	9.1181	0.6747	0.2460
1.9314	17.0009	2412	8.4932	0.6747	0.2538
1.8992	18.0009	2546	7.0325	0.6633	0.2605
1.8535	19.0009	2680	5.6560	0.7038	0.3071
1.6335	20.0008	2814	5.3615	0.6776	0.2932
1.6941	21.0008	2948	4.9919	0.7186	0.3186
1.5084	22.0008	3082	3.9574	0.7520	0.3696
1.4719	23.0008	3216	4.5916	0.7297	0.3547
1.3212	24.0008	3350	3.6468	0.7594	0.3713
1.307	25.0008	3484	3.5237	0.7587	0.4024
1.2214	26.0008	3618	3.4575	0.7792	0.4304
1.1814	27.0008	3752	3.1667	0.7846	0.4262
1.1466	28.0008	3886	3.3405	0.7968	0.4724
0.9577	29.0008	4020	3.1583	0.8071	0.4819
0.9267	30.0008	4154	3.0275	0.8058	0.4879
0.8851	31.0008	4288	3.5392	0.7986	0.5143
0.882	32.0008	4422	3.5224	0.8246	0.5507
0.8407	33.0008	4556	3.0531	0.8092	0.5276
0.8248	34.0007	4690	4.0564	0.8231	0.5266
0.7755	35.0007	4824	3.5834	0.8190	0.5663
0.7592	36.0007	4958	3.4173	0.8171	0.5549
0.6661	37.0007	5092	4.2197	0.8166	0.5710
0.6603	38.0007	5226	3.9674	0.8343	0.6002
0.6384	39.0007	5360	4.4058	0.8252	0.5720
0.5631	40.0007	5494	4.3746	0.8314	0.5936
0.5528	41.0007	5628	4.7360	0.8341	0.6066
0.5039	42.0007	5762	5.0756	0.8155	0.6062
0.4904	43.0007	5896	6.0930	0.8363	0.6133
0.5101	44.0007	6030	5.2076	0.8365	0.6182
0.4392	45.0007	6164	4.9944	0.8417	0.6172
0.4116	46.0007	6298	6.3981	0.8488	0.6288
0.4168	47.0006	6432	7.5452	0.8294	0.6278
0.4136	48.0006	6566	8.0047	0.8293	0.6256
0.3775	49.0006	6700	8.4309	0.8466	0.6341
0.3729	50.0006	6834	8.7770	0.8467	0.6450
0.3654	51.0006	6968	7.8225	0.8575	0.6527
0.3043	52.0006	7102	9.8550	0.8492	0.6278
0.3225	53.0006	7236	10.5254	0.8502	0.6454
0.2717	54.0006	7370	10.5405	0.8583	0.6569
0.2745	55.0006	7504	8.6422	0.8634	0.6624
0.252	56.0006	7638	9.4328	0.8654	0.6664
0.2307	57.0006	7772	10.4982	0.8560	0.6586
0.2361	58.0006	7906	11.1792	0.8631	0.6675
0.2304	59.0006	8040	10.1655	0.8653	0.6652
0.2344	60.0005	8174	9.3294	0.8667	0.6753
0.1947	61.0005	8308	11.4413	0.8581	0.6602
0.1962	62.0005	8442	13.8753	0.8653	0.6728
0.1785	63.0005	8576	12.1937	0.8702	0.6782
0.183	64.0005	8710	10.9347	0.8686	0.6805
0.1541	65.0005	8844	11.0915	0.8728	0.6777
0.1626	66.0005	8978	9.4917	0.8706	0.6839
0.1488	67.0005	9112	8.4089	0.8756	0.6876
0.1582	68.0005	9246	11.5104	0.8746	0.6877
0.1478	69.0005	9380	9.4014	0.8762	0.6853
0.1274	70.0005	9514	7.7970	0.8743	0.6859
0.1433	71.0005	9648	9.0243	0.8763	0.6928
0.1304	72.0005	9782	8.3959	0.8717	0.6854
0.1069	73.0005	9916	10.1263	0.8731	0.6948
0.1047	74.0004	10050	10.3105	0.8758	0.6964
0.1029	75.0004	10184	10.1299	0.8736	0.6971
0.0976	76.0004	10318	11.6986	0.8775	0.6949
0.1021	77.0004	10452	7.3299	0.8780	0.6969
0.098	78.0004	10586	7.0581	0.8832	0.7053
0.0869	79.0004	10720	7.9583	0.8735	0.6948
0.086	80.0004	10854	6.9579	0.8787	0.7042
0.0888	81.0004	10988	6.4546	0.8804	0.6985
0.0846	82.0004	11122	8.3227	0.8802	0.6994
0.0794	83.0004	11256	6.2773	0.8795	0.6974
0.074	84.0004	11390	6.5176	0.8774	0.7037
0.0683	85.0004	11524	7.1247	0.8812	0.7045
0.0706	86.0004	11658	6.2593	0.8714	0.7016
0.0752	87.0003	11792	6.0435	0.8832	0.6999
0.066	88.0003	11926	5.3773	0.8828	0.7046
0.0617	89.0003	12060	5.4743	0.8816	0.7058
0.0683	90.0003	12194	4.1337	0.8873	0.7096
0.0617	91.0003	12328	4.4415	0.8842	0.7129
0.0577	92.0003	12462	5.8869	0.8797	0.7109
0.0605	93.0003	12596	4.3467	0.8875	0.7150
0.0579	94.0003	12730	4.8355	0.8823	0.7074
0.0617	95.0003	12864	4.2854	0.8832	0.7124
0.0538	96.0003	12998	5.4158	0.8858	0.7140
0.0486	97.0003	13132	4.5717	0.8788	0.7052
0.05	98.0003	13266	4.6220	0.8854	0.7060
0.0465	99.0003	13400	4.7468	0.8809	0.7097
0.0426	100.0003	13534	4.7911	0.8902	0.7248
0.0405	101.0002	13668	4.4921	0.8850	0.7180
0.0457	102.0002	13802	4.4764	0.8821	0.7110
0.0551	103.0002	13936	4.4050	0.8822	0.7120
0.0513	104.0002	14070	5.5184	0.8842	0.7167
0.049	105.0002	14204	3.7660	0.8862	0.7144
0.0359	106.0002	14338	3.8898	0.8882	0.7146
0.0429	107.0002	14472	3.8001	0.8853	0.7162
0.0459	108.0002	14606	4.2187	0.8864	0.7200
0.0368	109.0002	14740	3.2633	0.8868	0.7217
0.0399	110.0002	14874	4.0445	0.8883	0.7236
0.0412	111.0002	15008	3.0764	0.8842	0.7163
0.0367	112.0002	15142	2.9909	0.8888	0.7191
0.0447	113.0002	15276	3.3596	0.8845	0.7173
0.0498	114.0001	15410	3.2358	0.8893	0.7204
0.0434	115.0001	15544	3.6316	0.8892	0.7216
0.0303	116.0001	15678	3.4604	0.8870	0.7231
0.0324	117.0001	15812	4.2350	0.8878	0.7268
0.0285	118.0001	15946	3.4594	0.8906	0.7243
0.0278	119.0001	16080	3.2241	0.8901	0.7287
0.0321	120.0001	16214	3.4182	0.8889	0.7240
0.0378	121.0001	16348	3.3749	0.8900	0.7288
0.0408	122.0001	16482	3.4872	0.8838	0.7135
0.0356	123.0001	16616	3.1642	0.8859	0.7186
0.0387	124.0001	16750	4.4308	0.8884	0.7207
0.0298	125.0001	16884	3.2343	0.8904	0.7264
0.0308	126.0001	17018	3.5812	0.8891	0.7264
0.0367	127.0001	17152	3.2693	0.8886	0.7186
0.0369	128.0000	17286	3.5343	0.8885	0.7210
0.0297	129.0000	17420	3.2665	0.8926	0.7275
0.0253	130.0000	17554	4.0442	0.8898	0.7250
0.0237	131.0000	17688	3.6591	0.8914	0.7252
0.0307	132.0000	17822	3.5956	0.8930	0.7306
0.0298	133.0000	17956	3.1877	0.8909	0.7278
0.0335	133.0010	18090	3.3527	0.8823	0.6996
0.0461	134.0010	18224	2.4156	0.8863	0.7207
0.0446	135.0010	18358	2.8521	0.8891	0.7226
0.0768	136.0010	18492	3.4625	0.8871	0.7241
0.0382	137.0010	18626	2.8181	0.8899	0.7262
0.0274	138.0010	18760	2.5205	0.8906	0.7313
0.0254	139.0010	18894	2.6763	0.8966	0.7323
0.0226	140.0010	19028	2.9371	0.8940	0.7292
0.0194	141.0009	19162	2.7370	0.8967	0.7310
0.0201	142.0009	19296	2.9556	0.8918	0.7349
0.0331	143.0009	19430	2.6614	0.8902	0.7274
0.0238	144.0009	19564	2.6757	0.8964	0.7377
0.0182	145.0009	19698	2.9333	0.8965	0.7392
0.0182	146.0009	19832	2.9219	0.8942	0.7350
0.0235	147.0009	19966	2.5359	0.8970	0.7363
0.0208	148.0009	20100	2.9597	0.8960	0.7329
0.0245	149.0009	20234	3.4899	0.8935	0.7274
0.0205	150.0009	20368	2.9046	0.8966	0.7321
0.03	151.0009	20502	2.9476	0.8875	0.7229
0.0274	152.0009	20636	2.9819	0.8968	0.7323
0.0251	153.0009	20770	2.6874	0.8946	0.7261
0.0245	154.0008	20904	2.7395	0.8964	0.7360
0.0254	155.0008	21038	2.8113	0.8959	0.7355
0.0252	156.0008	21172	2.6783	0.8957	0.7328
0.0314	157.0008	21306	3.2216	0.8958	0.7358
0.0236	158.0008	21440	2.5577	0.8938	0.7295
0.0194	159.0008	21574	2.6508	0.8977	0.7344
0.0171	160.0008	21708	2.7768	0.8980	0.7420
0.0139	161.0008	21842	3.0620	0.8997	0.7415
0.0155	162.0008	21976	3.3701	0.8983	0.7426
0.0273	163.0008	22110	2.5980	0.8959	0.7362
0.0273	164.0008	22244	2.3159	0.8902	0.7321
0.0429	165.0008	22378	3.0674	0.8856	0.7293
0.0454	166.0008	22512	2.6672	0.8862	0.7267
0.0362	167.0008	22646	2.7183	0.8916	0.7321
0.0218	168.0007	22780	2.8746	0.8947	0.7373
0.0174	169.0007	22914	2.4957	0.8923	0.7407
0.0318	170.0007	23048	2.4608	0.8945	0.7366
0.0168	171.0007	23182	2.8124	0.8911	0.7337
0.0192	172.0007	23316	2.2944	0.8921	0.7386
0.0114	173.0007	23450	2.3511	0.8986	0.7446
0.0163	174.0007	23584	2.5095	0.8969	0.7432
0.012	175.0007	23718	2.7702	0.8973	0.7427
0.0152	176.0007	23852	2.0697	0.8955	0.7392
0.0148	177.0007	23986	2.3427	0.8878	0.7304
0.0285	178.0007	24120	2.8657	0.8847	0.7339
0.0222	179.0007	24254	2.4395	0.8926	0.7376
0.0296	180.0007	24388	2.8879	0.8809	0.7351
0.0239	181.0006	24522	2.4314	0.8928	0.7345
0.0173	182.0006	24656	3.1864	0.8950	0.7380
0.0247	183.0006	24790	3.0219	0.8964	0.7439
0.0179	184.0006	24924	2.8811	0.8951	0.7400
0.0145	185.0006	25058	3.4297	0.8935	0.7408
0.0126	186.0006	25192	2.9333	0.8968	0.7429
0.0165	187.0006	25326	2.6480	0.8962	0.7407
0.0122	188.0006	25460	2.8965	0.8972	0.7458
0.0145	189.0006	25594	2.9822	0.8853	0.7354
0.0181	190.0006	25728	3.5766	0.8958	0.7448
0.0263	191.0006	25862	2.6186	0.8899	0.7314
0.0289	192.0006	25996	2.3478	0.8864	0.7319
0.0246	193.0006	26130	2.8031	0.8840	0.7311
0.0202	194.0005	26264	3.4395	0.8979	0.7428
0.0178	195.0005	26398	2.3771	0.8984	0.7491
0.0105	196.0005	26532	3.2311	0.8988	0.7382
0.0107	197.0005	26666	2.9176	0.9009	0.7454
0.0096	198.0005	26800	2.3241	0.9008	0.7452
0.0089	199.0005	26934	2.7654	0.9015	0.7478
0.0122	200.0005	27068	2.7759	0.8994	0.7448
0.01	201.0005	27202	2.9339	0.8991	0.7450
0.0217	202.0005	27336	2.4087	0.8947	0.7331
0.0313	203.0005	27470	2.3820	0.8943	0.7324
0.0177	204.0005	27604	2.4881	0.8945	0.7451
0.0147	205.0005	27738	3.0517	0.8993	0.7452
0.0096	206.0005	27872	3.0411	0.8988	0.7461
0.0119	207.0005	28006	2.7581	0.8992	0.7438
0.0092	208.0004	28140	3.5028	0.9002	0.7470
0.0124	209.0004	28274	2.6177	0.8998	0.7395
0.0098	210.0004	28408	2.5279	0.9016	0.7478
0.0132	211.0004	28542	2.2788	0.9003	0.7486
0.0128	212.0004	28676	2.3972	0.8991	0.7439
0.0117	213.0004	28810	2.4534	0.8987	0.7414
0.0157	214.0004	28944	2.9153	0.8976	0.7415
0.0133	215.0004	29078	2.8110	0.8997	0.7496
0.0099	216.0004	29212	3.1402	0.8946	0.7403
0.014	217.0004	29346	2.4582	0.8964	0.7393
0.0398	218.0004	29480	2.6382	0.8910	0.7294
0.0335	219.0004	29614	2.2036	0.8903	0.7326
0.0268	220.0004	29748	2.4391	0.8893	0.7387
0.0173	221.0003	29882	2.6495	0.9002	0.7470
0.0159	222.0003	30016	2.5093	0.9008	0.7475
0.008	223.0003	30150	2.7068	0.9016	0.7484
0.0087	224.0003	30284	3.0752	0.8995	0.7509
0.0073	225.0003	30418	2.4908	0.9020	0.7507
0.0055	226.0003	30552	3.2953	0.9026	0.7533
0.0058	227.0003	30686	3.1263	0.9030	0.7535
0.0054	228.0003	30820	2.8419	0.9029	0.7541
0.0051	229.0003	30954	3.5094	0.9033	0.7565
0.0062	230.0003	31088	3.2330	0.9014	0.7478
0.0064	231.0003	31222	2.7211	0.9024	0.7509
0.0203	232.0003	31356	2.5700	0.8961	0.7335
0.017	233.0003	31490	3.0242	0.8997	0.7455
0.0141	234.0003	31624	2.6718	0.8954	0.7435
0.0129	235.0002	31758	2.9305	0.9039	0.7550
0.0152	236.0002	31892	2.5959	0.8957	0.7507
0.0118	237.0002	32026	3.1503	0.8980	0.7539
0.01	238.0002	32160	3.5122	0.9012	0.7565
0.0084	239.0002	32294	3.5436	0.9038	0.7568
0.0057	240.0002	32428	3.1792	0.9043	0.7575
0.0045	241.0002	32562	3.7850	0.9045	0.7583
0.0087	242.0002	32696	3.7095	0.9041	0.7563
0.0049	243.0002	32830	2.8775	0.9047	0.7569
0.0058	244.0002	32964	3.1835	0.9036	0.7531
0.0092	245.0002	33098	2.6343	0.9019	0.7536
0.0066	246.0002	33232	3.0736	0.9030	0.7530
0.0184	247.0002	33366	3.3067	0.9036	0.7510
0.0225	248.0001	33500	2.7005	0.8858	0.7356
0.0301	249.0001	33634	2.3220	0.9004	0.7441
0.0187	250.0001	33768	2.0436	0.9010	0.7439
0.0149	251.0001	33902	2.9553	0.8985	0.7480
0.0102	252.0001	34036	2.8283	0.8982	0.7502
0.007	253.0001	34170	3.1807	0.9035	0.7526
0.0053	254.0001	34304	2.9726	0.9063	0.7584
0.0038	255.0001	34438	3.1863	0.9066	0.7603
0.0042	256.0001	34572	3.2792	0.9061	0.7569
0.0039	257.0001	34706	3.2983	0.9075	0.7601
0.0039	258.0001	34840	3.5063	0.9072	0.7598
0.0036	259.0001	34974	3.3959	0.9055	0.7573
0.004	260.0001	35108	3.2935	0.9065	0.7600
0.0038	261.0001	35242	2.7151	0.9065	0.7600
0.0046	262.0000	35376	2.6936	0.9044	0.7531
0.0088	263.0000	35510	2.8098	0.9016	0.7504
0.0181	264.0000	35644	3.3416	0.9008	0.7456
0.0097	265.0000	35778	3.0239	0.9035	0.7473
0.0158	266.0000	35912	4.8047	0.8987	0.7512
0.0223	267.0000	36046	3.7809	0.8936	0.7489
0.0157	267.0010	36180	4.0004	0.9007	0.7558
0.0135	268.0010	36314	4.0651	0.9015	0.7562
0.0063	269.0010	36448	3.8184	0.8999	0.7565
0.0074	270.0010	36582	2.8200	0.8997	0.7549
0.0051	271.0010	36716	4.4006	0.9007	0.7525
0.0042	272.0010	36850	5.9740	0.9028	0.7575
0.0031	273.0010	36984	3.8698	0.9009	0.7576
0.0134	274.0010	37118	3.6286	0.9013	0.7537

Framework versions

• Transformers 4.46.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.1