scenario-kd-po-ner-half_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: 0.3977
• Precision: 0.8322
• Recall: 0.8250
• F1: 0.8286
• Accuracy: 0.9824

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
0.9599	0.2910	500	0.7292	0.7506	0.7468	0.7487	0.9756
0.5565	0.5821	1000	0.6142	0.7746	0.7775	0.7761	0.9783
0.5029	0.8731	1500	0.5821	0.7811	0.7804	0.7807	0.9786
0.4411	1.1641	2000	0.5625	0.7944	0.7860	0.7902	0.9793
0.3891	1.4552	2500	0.5504	0.7850	0.7990	0.7919	0.9791
0.3893	1.7462	3000	0.5271	0.7792	0.8126	0.7955	0.9791
0.3654	2.0373	3500	0.5237	0.7939	0.8031	0.7985	0.9797
0.3332	2.3283	4000	0.5146	0.7948	0.8204	0.8074	0.9803
0.3191	2.6193	4500	0.5136	0.7956	0.8100	0.8027	0.9801
0.3158	2.9104	5000	0.5068	0.7983	0.8136	0.8059	0.9803
0.2954	3.2014	5500	0.5005	0.7992	0.8158	0.8074	0.9804
0.282	3.4924	6000	0.4871	0.8028	0.8204	0.8115	0.9804
0.2806	3.7835	6500	0.4767	0.8069	0.8104	0.8087	0.9803
0.2693	4.0745	7000	0.4729	0.7969	0.8150	0.8058	0.9801
0.2524	4.3655	7500	0.4877	0.8169	0.8038	0.8103	0.9806
0.2499	4.6566	8000	0.4748	0.8036	0.8134	0.8085	0.9803
0.2513	4.9476	8500	0.4685	0.8154	0.8080	0.8117	0.9806
0.2311	5.2386	9000	0.4729	0.8205	0.8065	0.8134	0.9808
0.228	5.5297	9500	0.4614	0.8142	0.8126	0.8134	0.9812
0.2281	5.8207	10000	0.4755	0.8128	0.8124	0.8126	0.9808
0.2231	6.1118	10500	0.4550	0.8135	0.8160	0.8148	0.9810
0.2096	6.4028	11000	0.4833	0.8254	0.7918	0.8082	0.9802
0.2114	6.6938	11500	0.4562	0.8091	0.8091	0.8091	0.9807
0.2085	6.9849	12000	0.4630	0.8141	0.8083	0.8112	0.9807
0.1921	7.2759	12500	0.4601	0.8149	0.8201	0.8175	0.9809
0.1954	7.5669	13000	0.4489	0.8231	0.8117	0.8174	0.9810
0.1945	7.8580	13500	0.4572	0.8169	0.8038	0.8103	0.9806
0.1889	8.1490	14000	0.4604	0.8202	0.8091	0.8146	0.9807
0.1821	8.4400	14500	0.4510	0.8178	0.8204	0.8191	0.9810
0.1815	8.7311	15000	0.4462	0.8179	0.8270	0.8224	0.9814
0.1795	9.0221	15500	0.4396	0.8125	0.8160	0.8143	0.9810
0.1711	9.3132	16000	0.4452	0.8179	0.8139	0.8159	0.9811
0.1703	9.6042	16500	0.4469	0.8178	0.8140	0.8159	0.9813
0.1705	9.8952	17000	0.4361	0.8110	0.8234	0.8172	0.9811
0.1662	10.1863	17500	0.4399	0.8241	0.8204	0.8222	0.9816
0.1618	10.4773	18000	0.4344	0.8123	0.8247	0.8184	0.9817
0.1601	10.7683	18500	0.4435	0.8184	0.8116	0.8150	0.9810
0.16	11.0594	19000	0.4434	0.8417	0.8029	0.8218	0.9814
0.1511	11.3504	19500	0.4432	0.8323	0.8191	0.8256	0.9816
0.1531	11.6414	20000	0.4323	0.8223	0.8194	0.8208	0.9816
0.1532	11.9325	20500	0.4339	0.8201	0.8189	0.8195	0.9811
0.1481	12.2235	21000	0.4232	0.8148	0.8238	0.8193	0.9817
0.1455	12.5146	21500	0.4300	0.8196	0.8260	0.8228	0.9819
0.1469	12.8056	22000	0.4328	0.8244	0.8182	0.8213	0.9813
0.1428	13.0966	22500	0.4262	0.8236	0.8300	0.8268	0.9817
0.1402	13.3877	23000	0.4295	0.8181	0.8198	0.8190	0.9812
0.1407	13.6787	23500	0.4317	0.8180	0.8224	0.8202	0.9813
0.1413	13.9697	24000	0.4279	0.8240	0.8160	0.8200	0.9816
0.1342	14.2608	24500	0.4316	0.8184	0.8153	0.8169	0.9812
0.1371	14.5518	25000	0.4215	0.8278	0.8254	0.8266	0.9817
0.1337	14.8428	25500	0.4395	0.8251	0.8120	0.8185	0.9811
0.1343	15.1339	26000	0.4257	0.8211	0.8235	0.8223	0.9814
0.131	15.4249	26500	0.4299	0.8345	0.8072	0.8206	0.9812
0.1315	15.7159	27000	0.4187	0.8298	0.8199	0.8248	0.9817
0.1305	16.0070	27500	0.4220	0.8269	0.8175	0.8222	0.9817
0.1268	16.2980	28000	0.4201	0.8257	0.8162	0.8209	0.9814
0.1258	16.5891	28500	0.4210	0.8204	0.8286	0.8245	0.9819
0.1275	16.8801	29000	0.4212	0.8279	0.8169	0.8224	0.9813
0.1256	17.1711	29500	0.4155	0.8269	0.8166	0.8217	0.9814
0.1218	17.4622	30000	0.4119	0.8236	0.8218	0.8227	0.9816
0.1234	17.7532	30500	0.4223	0.8295	0.8194	0.8244	0.9818
0.1216	18.0442	31000	0.4184	0.8353	0.8218	0.8285	0.9820
0.1189	18.3353	31500	0.4211	0.8323	0.8184	0.8253	0.9820
0.1201	18.6263	32000	0.4168	0.8302	0.8198	0.8250	0.9818
0.1209	18.9173	32500	0.4124	0.8303	0.8204	0.8253	0.9817
0.1187	19.2084	33000	0.4209	0.8304	0.8087	0.8194	0.9814
0.1178	19.4994	33500	0.4143	0.8271	0.8199	0.8235	0.9817
0.1168	19.7905	34000	0.4179	0.8304	0.8169	0.8236	0.9815
0.118	20.0815	34500	0.4162	0.8254	0.8181	0.8217	0.9816
0.1161	20.3725	35000	0.4174	0.8287	0.8150	0.8218	0.9817
0.1147	20.6636	35500	0.4129	0.8278	0.8246	0.8262	0.9821
0.114	20.9546	36000	0.4091	0.8303	0.8217	0.8260	0.9819
0.1133	21.2456	36500	0.4167	0.8248	0.8224	0.8236	0.9816
0.1126	21.5367	37000	0.4105	0.8314	0.8176	0.8245	0.9817
0.1126	21.8277	37500	0.4052	0.8302	0.8225	0.8264	0.9820
0.1111	22.1187	38000	0.4073	0.8298	0.8221	0.8259	0.9822
0.1108	22.4098	38500	0.4100	0.8324	0.8257	0.8291	0.9820
0.1103	22.7008	39000	0.4092	0.8296	0.8160	0.8228	0.9816
0.1114	22.9919	39500	0.4029	0.8322	0.8243	0.8282	0.9821
0.1096	23.2829	40000	0.4051	0.8292	0.8238	0.8265	0.9820
0.1088	23.5739	40500	0.4020	0.8275	0.8195	0.8235	0.9818
0.1091	23.8650	41000	0.4063	0.8256	0.8194	0.8224	0.9815
0.1079	24.1560	41500	0.4015	0.8284	0.8218	0.8251	0.9819
0.108	24.4470	42000	0.4058	0.8349	0.8217	0.8282	0.9822
0.1081	24.7381	42500	0.4035	0.8301	0.8207	0.8254	0.9817
0.1079	25.0291	43000	0.4029	0.8306	0.8235	0.8271	0.9820
0.1065	25.3201	43500	0.4053	0.8309	0.8235	0.8272	0.9818
0.1062	25.6112	44000	0.4046	0.8276	0.8188	0.8232	0.9818
0.1063	25.9022	44500	0.4050	0.8323	0.8197	0.8259	0.9820
0.1054	26.1932	45000	0.4000	0.8310	0.8267	0.8289	0.9820
0.1048	26.4843	45500	0.4063	0.8311	0.8227	0.8269	0.9819
0.105	26.7753	46000	0.3994	0.8315	0.8233	0.8274	0.9821
0.104	27.0664	46500	0.4046	0.8320	0.8198	0.8259	0.9820
0.1038	27.3574	47000	0.4042	0.8312	0.8243	0.8277	0.9820
0.1042	27.6484	47500	0.4006	0.8340	0.8276	0.8308	0.9821
0.1041	27.9395	48000	0.4000	0.8344	0.8244	0.8294	0.9823
0.1036	28.2305	48500	0.4027	0.8325	0.8220	0.8272	0.9819
0.1039	28.5215	49000	0.3972	0.8339	0.8257	0.8298	0.9823
0.1023	28.8126	49500	0.3960	0.8310	0.8267	0.8289	0.9824
0.103	29.1036	50000	0.4038	0.8320	0.8188	0.8253	0.9817
0.102	29.3946	50500	0.4010	0.8287	0.8211	0.8249	0.9818
0.1025	29.6857	51000	0.3960	0.8361	0.8277	0.8319	0.9823
0.1025	29.9767	51500	0.3977	0.8322	0.8250	0.8286	0.9824

Framework versions

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