scenario-kd-scr-ner-full_data-univner_full55

This model is a fine-tuned version of haryoaw/scenario-TCR-NER_data-univner_full on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.1087
• Precision: 0.6202
• Recall: 0.5509
• F1: 0.5835
• Accuracy: 0.9594

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 55
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
2.7888	0.2910	500	2.4429	0.2941	0.0087	0.0168	0.9242
2.1116	0.5821	1000	2.1355	0.2870	0.1215	0.1707	0.9277
1.9109	0.8731	1500	2.0239	0.2850	0.1580	0.2033	0.9306
1.771	1.1641	2000	1.9268	0.4072	0.1535	0.2230	0.9319
1.633	1.4552	2500	1.8866	0.2862	0.2658	0.2756	0.9336
1.5872	1.7462	3000	1.7527	0.3273	0.2818	0.3028	0.9372
1.4833	2.0373	3500	1.7643	0.3558	0.2464	0.2912	0.9377
1.3965	2.3283	4000	1.6664	0.3675	0.3353	0.3507	0.9394
1.3237	2.6193	4500	1.6537	0.3445	0.3481	0.3463	0.9376
1.2791	2.9104	5000	1.5552	0.3960	0.3758	0.3857	0.9431
1.2023	3.2014	5500	1.5571	0.4338	0.3855	0.4083	0.9442
1.1456	3.4924	6000	1.4999	0.4258	0.3998	0.4124	0.9457
1.1315	3.7835	6500	1.4824	0.4244	0.3741	0.3977	0.9458
1.0693	4.0745	7000	1.4836	0.4407	0.3842	0.4105	0.9461
1.0052	4.3655	7500	1.4413	0.4322	0.4275	0.4298	0.9472
0.9737	4.6566	8000	1.4101	0.4634	0.4161	0.4385	0.9491
0.9521	4.9476	8500	1.3865	0.4476	0.4214	0.4341	0.9491
0.8818	5.2386	9000	1.4115	0.4612	0.4232	0.4414	0.9494
0.8396	5.5297	9500	1.3702	0.4645	0.4470	0.4556	0.9501
0.8456	5.8207	10000	1.3441	0.5076	0.4252	0.4627	0.9508
0.829	6.1118	10500	1.3357	0.4922	0.4718	0.4818	0.9518
0.7611	6.4028	11000	1.3320	0.5100	0.4548	0.4808	0.9522
0.7475	6.6938	11500	1.3570	0.4852	0.4953	0.4902	0.9531
0.7362	6.9849	12000	1.3154	0.5039	0.4929	0.4983	0.9529
0.6776	7.2759	12500	1.3044	0.5099	0.4884	0.4989	0.9534
0.6701	7.5669	13000	1.2921	0.5229	0.4675	0.4936	0.9541
0.6586	7.8580	13500	1.2670	0.5185	0.5067	0.5126	0.9548
0.6284	8.1490	14000	1.2752	0.5346	0.4979	0.5156	0.9548
0.6025	8.4400	14500	1.2738	0.5270	0.4884	0.5070	0.9545
0.5955	8.7311	15000	1.2564	0.5340	0.4895	0.5108	0.9552
0.5784	9.0221	15500	1.2502	0.5406	0.5035	0.5214	0.9546
0.5479	9.3132	16000	1.2339	0.5418	0.5203	0.5308	0.9566
0.54	9.6042	16500	1.2380	0.5473	0.5175	0.5320	0.9564
0.5368	9.8952	17000	1.2403	0.5726	0.5044	0.5363	0.9568
0.5151	10.1863	17500	1.2152	0.5516	0.5445	0.5480	0.9571
0.4959	10.4773	18000	1.2323	0.5657	0.5359	0.5504	0.9570
0.4946	10.7683	18500	1.2150	0.5679	0.5236	0.5449	0.9575
0.499	11.0594	19000	1.2119	0.5637	0.5372	0.5501	0.9576
0.462	11.3504	19500	1.2289	0.5736	0.5294	0.5506	0.9578
0.4631	11.6414	20000	1.2106	0.5661	0.5435	0.5546	0.9576
0.464	11.9325	20500	1.2292	0.5886	0.5087	0.5458	0.9576
0.4463	12.2235	21000	1.2135	0.5823	0.5465	0.5639	0.9578
0.4339	12.5146	21500	1.2098	0.5890	0.5208	0.5528	0.9578
0.4386	12.8056	22000	1.1906	0.5754	0.5387	0.5565	0.9573
0.4249	13.0966	22500	1.1972	0.5873	0.5379	0.5615	0.9580
0.4076	13.3877	23000	1.1994	0.5680	0.5585	0.5632	0.9576
0.4122	13.6787	23500	1.2129	0.5894	0.5331	0.5598	0.9580
0.4156	13.9697	24000	1.1865	0.5779	0.5485	0.5628	0.9580
0.3926	14.2608	24500	1.1828	0.5974	0.5397	0.5671	0.9589
0.3966	14.5518	25000	1.1764	0.5959	0.5390	0.5660	0.9586
0.3861	14.8428	25500	1.1769	0.5869	0.5307	0.5574	0.9581
0.3847	15.1339	26000	1.1997	0.5829	0.5406	0.5610	0.9581
0.3703	15.4249	26500	1.1809	0.5736	0.5543	0.5638	0.9582
0.3747	15.7159	27000	1.1896	0.5871	0.5320	0.5582	0.9577
0.3713	16.0070	27500	1.1700	0.5965	0.5422	0.5681	0.9589
0.3558	16.2980	28000	1.1922	0.5970	0.5416	0.5680	0.9586
0.3582	16.5891	28500	1.1507	0.5831	0.5470	0.5644	0.9586
0.3571	16.8801	29000	1.1405	0.5899	0.5418	0.5648	0.9584
0.3522	17.1711	29500	1.1610	0.6046	0.5517	0.5769	0.9588
0.3414	17.4622	30000	1.1670	0.6042	0.5485	0.5750	0.9590
0.3488	17.7532	30500	1.1502	0.5904	0.5624	0.5761	0.9586
0.34	18.0442	31000	1.1595	0.6091	0.5304	0.5670	0.9585
0.3336	18.3353	31500	1.1553	0.6025	0.5439	0.5717	0.9589
0.3295	18.6263	32000	1.1683	0.5916	0.5337	0.5611	0.9580
0.3345	18.9173	32500	1.1478	0.5825	0.5536	0.5677	0.9585
0.3263	19.2084	33000	1.1415	0.6093	0.5369	0.5708	0.9589
0.3206	19.4994	33500	1.1410	0.5888	0.5637	0.5760	0.9593
0.3234	19.7905	34000	1.1371	0.6072	0.5490	0.5766	0.9591
0.3212	20.0815	34500	1.1401	0.6006	0.5478	0.5730	0.9587
0.3154	20.3725	35000	1.1505	0.6165	0.5400	0.5758	0.9591
0.3081	20.6636	35500	1.1512	0.5977	0.5393	0.5670	0.9591
0.3137	20.9546	36000	1.1477	0.6185	0.5357	0.5741	0.9590
0.3048	21.2456	36500	1.1344	0.6070	0.5416	0.5724	0.9593
0.3028	21.5367	37000	1.1308	0.6192	0.5481	0.5815	0.9594
0.3039	21.8277	37500	1.1492	0.6167	0.5318	0.5711	0.9591
0.3013	22.1187	38000	1.1340	0.6139	0.5393	0.5742	0.9592
0.2966	22.4098	38500	1.1176	0.6073	0.5561	0.5806	0.9594
0.2956	22.7008	39000	1.1156	0.6100	0.5627	0.5854	0.9593
0.2982	22.9919	39500	1.1282	0.6162	0.5553	0.5842	0.9596
0.2915	23.2829	40000	1.1359	0.6048	0.5510	0.5766	0.9593
0.2882	23.5739	40500	1.1194	0.6075	0.5517	0.5783	0.9592
0.2906	23.8650	41000	1.1256	0.6058	0.5442	0.5734	0.9590
0.2852	24.1560	41500	1.1115	0.6143	0.5465	0.5785	0.9596
0.2864	24.4470	42000	1.1214	0.6103	0.5441	0.5753	0.9594
0.2829	24.7381	42500	1.1333	0.6267	0.5346	0.5770	0.9592
0.2836	25.0291	43000	1.1195	0.6067	0.5550	0.5797	0.9591
0.2795	25.3201	43500	1.1260	0.6332	0.5315	0.5779	0.9593
0.2779	25.6112	44000	1.1119	0.6164	0.5457	0.5789	0.9597
0.2787	25.9022	44500	1.1094	0.6103	0.5640	0.5862	0.9600
0.2765	26.1932	45000	1.1104	0.6166	0.5474	0.5799	0.9596
0.2743	26.4843	45500	1.1164	0.6172	0.5553	0.5846	0.9596
0.2731	26.7753	46000	1.1246	0.6158	0.5578	0.5854	0.9594
0.2705	27.0664	46500	1.1110	0.6153	0.5468	0.5790	0.9593
0.2707	27.3574	47000	1.1101	0.6207	0.5586	0.5880	0.9602
0.2713	27.6484	47500	1.1131	0.6203	0.5455	0.5805	0.9596
0.2704	27.9395	48000	1.1122	0.6193	0.5494	0.5823	0.9596
0.2669	28.2305	48500	1.1127	0.6139	0.5519	0.5812	0.9596
0.2696	28.5215	49000	1.1148	0.6233	0.5449	0.5815	0.9597
0.2658	28.8126	49500	1.1130	0.6182	0.5451	0.5794	0.9597
0.2663	29.1036	50000	1.1070	0.6170	0.5475	0.5802	0.9593
0.2625	29.3946	50500	1.1055	0.6172	0.5498	0.5816	0.9599
0.2652	29.6857	51000	1.1010	0.6332	0.5516	0.5896	0.9603
0.2662	29.9767	51500	1.1087	0.6202	0.5509	0.5835	0.9594

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1