ht-finbert-cls-v5_ftis_noPretrain_tdso

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

• Loss: 2.7215
• Accuracy: 0.8960
• Macro F1: 0.7522

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: 6725
• training_steps: 134500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
54.1267	1.0002	100	42.4539	0.0693	0.0340
20.8339	2.0005	200	75.1536	0.1443	0.0557
8.6862	4.0002	300	105.8607	0.4400	0.1239
6.9579	5.0004	400	132.3561	0.5393	0.1544
6.3137	7.0002	500	163.3538	0.5817	0.1790
5.7884	8.0004	600	149.1104	0.6043	0.1961
5.238	10.0001	700	149.0267	0.6232	0.2087
4.6053	11.0004	800	122.6960	0.6450	0.2278
3.9789	13.0001	900	88.3309	0.6608	0.2490
3.4193	14.0004	1000	70.7087	0.6655	0.2566
2.9712	16.0001	1100	53.3170	0.6815	0.2796
2.5305	17.0003	1200	36.2476	0.7014	0.3222
2.3318	19.0001	1300	26.5832	0.7201	0.3406
2.04	20.0003	1400	19.7882	0.7461	0.3732
1.8635	22.0000	1500	18.2478	0.7632	0.4002
1.7301	23.0003	1600	12.3389	0.7779	0.4322
1.5551	24.0005	1700	9.3268	0.7815	0.4542
1.4405	26.0002	1800	9.2790	0.7954	0.4642
1.4124	27.0005	1900	7.5042	0.7902	0.4744
1.265	29.0002	2000	6.2492	0.7913	0.4995
1.2012	30.0004	2100	6.2005	0.8002	0.5169
1.1371	32.0002	2200	5.8517	0.8080	0.5313
1.0514	33.0004	2300	5.5730	0.8110	0.5412
1.0293	35.0001	2400	4.6457	0.8153	0.5503
0.9164	36.0004	2500	4.8986	0.8244	0.5791
0.8696	38.0001	2600	5.2707	0.8242	0.5853
0.8678	39.0004	2700	5.4791	0.8199	0.5702
0.817	41.0001	2800	5.8519	0.8276	0.5881
0.7781	42.0003	2900	5.7369	0.8361	0.6065
0.7561	44.0001	3000	6.8293	0.8322	0.5961
0.7016	45.0003	3100	6.5144	0.8343	0.6086
0.7073	47.0000	3200	6.8592	0.8387	0.6133
0.6632	48.0003	3300	6.8253	0.8402	0.6156
0.6415	49.0005	3400	7.3926	0.8440	0.6237
0.6284	51.0002	3500	8.3232	0.8490	0.6307
0.6271	52.0005	3600	8.3714	0.8462	0.6354
0.6024	54.0002	3700	7.8233	0.8496	0.6374
0.5716	55.0004	3800	8.6780	0.8491	0.6344
0.5656	57.0002	3900	8.8122	0.8523	0.6446
0.5519	58.0004	4000	7.9019	0.8485	0.6443
0.5383	60.0001	4100	8.2911	0.8525	0.6447
0.5225	61.0004	4200	8.9799	0.8580	0.6595
0.5166	63.0001	4300	9.7411	0.8562	0.6523
0.5067	64.0004	4400	10.2261	0.8612	0.6616
0.5083	66.0001	4500	10.1360	0.8607	0.6669
0.4822	67.0003	4600	10.8119	0.8633	0.6696
0.478	69.0001	4700	12.0300	0.8601	0.6650
0.468	70.0003	4800	10.7509	0.8601	0.6679
0.4786	72.0000	4900	11.1814	0.8606	0.6688
0.4542	73.0003	5000	11.7393	0.8686	0.6788
0.4471	74.0005	5100	11.8721	0.8661	0.6778
0.4425	76.0002	5200	12.2049	0.8681	0.6829
0.4378	77.0005	5300	10.3071	0.8684	0.6814
0.441	79.0002	5400	12.7758	0.8656	0.6875
0.4218	80.0004	5500	13.9862	0.8695	0.6838
0.4237	82.0002	5600	11.4036	0.8681	0.6824
0.4261	83.0004	5700	12.3810	0.8685	0.6849
0.4189	85.0001	5800	10.9566	0.8703	0.6840
0.4163	86.0004	5900	11.9092	0.8658	0.6862
0.4062	88.0001	6000	10.6187	0.8748	0.7032
0.4054	89.0004	6100	11.1740	0.8753	0.6960
0.3989	91.0001	6200	10.6474	0.8748	0.6953
0.4013	92.0003	6300	11.6176	0.8754	0.7027
0.3905	94.0001	6400	10.2454	0.8753	0.7045
0.3826	95.0003	6500	12.9530	0.8729	0.6963
0.3971	97.0000	6600	11.5349	0.8716	0.6941
0.3878	98.0003	6700	10.0413	0.8744	0.7073
0.3875	99.0005	6800	9.9055	0.8734	0.7015
0.3761	101.0002	6900	10.3418	0.8789	0.7061
0.3774	102.0005	7000	10.1067	0.8806	0.7090
0.3825	104.0002	7100	8.3686	0.8806	0.7144
0.3876	105.0004	7200	6.3883	0.8724	0.6980
0.3852	107.0002	7300	9.0841	0.8739	0.7034
0.3767	108.0004	7400	7.3012	0.8813	0.7106
0.3718	110.0001	7500	8.5764	0.8828	0.7159
0.3696	111.0004	7600	8.4104	0.8832	0.7154
0.3759	113.0001	7700	8.6043	0.8832	0.7155
0.3668	114.0004	7800	7.4298	0.8847	0.7246
0.3684	116.0001	7900	9.5902	0.8848	0.7244
0.363	117.0003	8000	6.8931	0.8838	0.7222
0.357	119.0001	8100	6.9210	0.8831	0.7257
0.3575	120.0003	8200	5.8508	0.8858	0.7287
0.35	122.0000	8300	6.6739	0.8851	0.7277
0.3518	123.0003	8400	6.6337	0.8850	0.7269
0.3539	124.0005	8500	6.4691	0.8851	0.7266
0.3478	126.0002	8600	5.6397	0.8865	0.7300
0.3451	127.0005	8700	5.6356	0.8880	0.7322
0.3415	129.0002	8800	4.4881	0.8871	0.7312
0.3425	130.0004	8900	4.5283	0.8854	0.7296
0.3484	132.0002	9000	4.5845	0.8811	0.7249
0.3767	133.0004	9100	4.6531	0.8795	0.7151
0.3704	135.0001	9200	3.2323	0.8747	0.7109
0.3663	136.0004	9300	4.3291	0.8773	0.7187
0.3553	138.0001	9400	3.9327	0.8841	0.7259
0.3423	139.0004	9500	4.4498	0.8860	0.7305
0.3433	141.0001	9600	4.7134	0.8794	0.7298
0.3424	142.0003	9700	4.5364	0.8863	0.7271
0.3576	144.0001	9800	3.2682	0.8843	0.7313
0.3422	145.0003	9900	2.8321	0.8844	0.7296
0.3412	147.0000	10000	4.3759	0.8890	0.7307
0.349	148.0003	10100	3.9332	0.8851	0.7280
0.3391	149.0005	10200	4.5327	0.8893	0.7355
0.3335	151.0002	10300	4.9665	0.8898	0.7368
0.3369	152.0005	10400	3.4262	0.8880	0.7363
0.3314	154.0002	10500	3.5618	0.8893	0.7368
0.3277	155.0004	10600	3.4955	0.8914	0.7393
0.326	157.0002	10700	3.0240	0.8900	0.7380
0.3266	158.0004	10800	2.4971	0.8895	0.7372
0.3254	160.0001	10900	2.9598	0.8879	0.7346
0.3249	161.0004	11000	3.1897	0.8910	0.7416
0.3247	163.0001	11100	3.0436	0.8897	0.7377
0.331	164.0004	11200	3.0063	0.8884	0.7356
0.3372	166.0001	11300	3.3399	0.8830	0.7247
0.3365	167.0003	11400	3.2443	0.8871	0.7314
0.3404	169.0001	11500	2.7850	0.8745	0.7107
0.3669	170.0003	11600	2.0442	0.8828	0.7278
0.3399	172.0000	11700	2.4131	0.8859	0.7352
0.33	173.0003	11800	2.5376	0.8850	0.7363
0.3209	174.0005	11900	2.9024	0.8932	0.7460
0.3181	176.0002	12000	3.0534	0.8935	0.7446
0.3167	177.0005	12100	2.8713	0.8931	0.7458
0.3149	179.0002	12200	3.1409	0.8911	0.7397
0.3168	180.0004	12300	2.7827	0.8927	0.7423
0.3154	182.0002	12400	2.9169	0.8938	0.7436
0.3143	183.0004	12500	2.7046	0.8927	0.7427
0.3175	185.0001	12600	3.0517	0.8904	0.7388
0.3473	186.0004	12700	2.5254	0.8668	0.7191
0.373	188.0001	12800	1.9765	0.8802	0.7184
0.335	189.0004	12900	2.3713	0.8882	0.7289
0.3308	191.0001	13000	2.3606	0.8891	0.7436
0.3272	192.0003	13100	2.3548	0.8907	0.7420
0.3138	194.0001	13200	2.3770	0.8858	0.7389
0.3148	195.0003	13300	2.4873	0.8950	0.7485
0.3104	197.0000	13400	2.6665	0.8949	0.7487
0.308	198.0003	13500	2.7137	0.8960	0.7522
0.3089	199.0005	13600	2.8533	0.8954	0.7502
0.3083	201.0002	13700	2.7738	0.8949	0.7473
0.3058	202.0005	13800	2.8489	0.8945	0.7490
0.3171	204.0002	13900	2.4588	0.8926	0.7478
0.312	205.0004	14000	2.1815	0.8930	0.7438
0.3102	207.0002	14100	2.3426	0.8928	0.7459
0.3074	208.0004	14200	2.4680	0.8914	0.7424
0.305	210.0001	14300	2.3430	0.8940	0.7440
0.3055	211.0004	14400	2.6109	0.8942	0.7492
0.3065	213.0001	14500	3.2677	0.8936	0.7400
0.3038	214.0004	14600	2.6056	0.8951	0.7500
0.3022	216.0001	14700	3.7801	0.8929	0.7465
0.3045	217.0003	14800	3.1071	0.8881	0.7453
0.3045	219.0001	14900	3.9928	0.8911	0.7454
0.3066	220.0003	15000	2.8048	0.8893	0.7415
0.3083	222.0000	15100	2.8299	0.8808	0.7369
0.3089	223.0003	15200	2.6550	0.8845	0.7384
0.3122	224.0005	15300	2.8834	0.8858	0.7403
0.3318	226.0002	15400	1.9369	0.8823	0.7328
0.3205	227.0005	15500	2.7083	0.8886	0.7431

Framework versions

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