scenario-non-kd-scr-ner-half_data-univner_full44

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: 0.1359
• Precision: 0.8579
• Recall: 0.8661
• F1: 0.8620
• Accuracy: 0.9848

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: 44
• 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
0.0095	0.2910	500	0.0842	0.8414	0.8590	0.8501	0.9842
0.0115	0.5821	1000	0.0851	0.8363	0.8626	0.8493	0.9836
0.0108	0.8731	1500	0.0812	0.8414	0.8664	0.8537	0.9844
0.0086	1.1641	2000	0.0967	0.8277	0.8729	0.8497	0.9829
0.0083	1.4552	2500	0.0794	0.8469	0.8652	0.8560	0.9841
0.0082	1.7462	3000	0.0808	0.8391	0.8712	0.8548	0.9843
0.0089	2.0373	3500	0.0862	0.8488	0.8645	0.8566	0.9846
0.0063	2.3283	4000	0.0889	0.8455	0.8735	0.8593	0.9847
0.0068	2.6193	4500	0.0900	0.8463	0.8638	0.8550	0.9842
0.0071	2.9104	5000	0.0883	0.8362	0.8691	0.8524	0.9837
0.0055	3.2014	5500	0.0945	0.8433	0.8588	0.8510	0.9838
0.0055	3.4924	6000	0.0951	0.8428	0.8687	0.8556	0.9840
0.0059	3.7835	6500	0.0975	0.8563	0.8551	0.8557	0.9842
0.0054	4.0745	7000	0.1009	0.8422	0.8673	0.8546	0.9841
0.0047	4.3655	7500	0.0967	0.8556	0.8589	0.8572	0.9848
0.0048	4.6566	8000	0.0989	0.8506	0.8694	0.8599	0.9847
0.0048	4.9476	8500	0.0959	0.8526	0.8713	0.8619	0.9844
0.0036	5.2386	9000	0.1014	0.8476	0.8706	0.8589	0.9841
0.0042	5.5297	9500	0.1147	0.8068	0.8613	0.8332	0.9814
0.0048	5.8207	10000	0.1065	0.8392	0.8691	0.8539	0.9844
0.0041	6.1118	10500	0.1076	0.8417	0.8758	0.8584	0.9843
0.0035	6.4028	11000	0.1029	0.8505	0.8732	0.8617	0.9848
0.0036	6.6938	11500	0.0929	0.8460	0.8719	0.8587	0.9849
0.0038	6.9849	12000	0.1019	0.8494	0.8631	0.8562	0.9846
0.0031	7.2759	12500	0.1073	0.8563	0.8575	0.8569	0.9845
0.0031	7.5669	13000	0.1013	0.8431	0.8696	0.8561	0.9847
0.0034	7.8580	13500	0.1058	0.8533	0.8596	0.8565	0.9845
0.0027	8.1490	14000	0.1154	0.8431	0.8719	0.8572	0.9845
0.0027	8.4400	14500	0.1030	0.8404	0.8785	0.8591	0.9845
0.0028	8.7311	15000	0.1132	0.8559	0.8510	0.8534	0.9846
0.003	9.0221	15500	0.1106	0.8514	0.8648	0.8581	0.9848
0.0022	9.3132	16000	0.1136	0.8586	0.8657	0.8621	0.9852
0.0025	9.6042	16500	0.1128	0.8494	0.8697	0.8594	0.9848
0.002	9.8952	17000	0.1139	0.8453	0.8600	0.8526	0.9841
0.0024	10.1863	17500	0.1124	0.8541	0.8658	0.8599	0.9849
0.002	10.4773	18000	0.1154	0.8368	0.8663	0.8513	0.9842
0.0019	10.7683	18500	0.1182	0.8457	0.8629	0.8542	0.9844
0.0023	11.0594	19000	0.1140	0.8531	0.8596	0.8563	0.9846
0.0018	11.3504	19500	0.1194	0.8526	0.8683	0.8604	0.9851
0.0016	11.6414	20000	0.1198	0.8527	0.8652	0.8589	0.9848
0.0021	11.9325	20500	0.1169	0.8592	0.8654	0.8623	0.9852
0.0016	12.2235	21000	0.1229	0.8605	0.8626	0.8616	0.9848
0.0021	12.5146	21500	0.1198	0.8484	0.8697	0.8589	0.9846
0.0019	12.8056	22000	0.1177	0.8535	0.8600	0.8568	0.9844
0.0013	13.0966	22500	0.1190	0.8436	0.8716	0.8574	0.9844
0.0016	13.3877	23000	0.1227	0.8475	0.8665	0.8569	0.9847
0.0012	13.6787	23500	0.1237	0.8513	0.8676	0.8594	0.9848
0.0015	13.9697	24000	0.1198	0.8407	0.8709	0.8555	0.9843
0.0012	14.2608	24500	0.1239	0.8516	0.8689	0.8602	0.9850
0.0014	14.5518	25000	0.1261	0.8432	0.8634	0.8532	0.9843
0.0015	14.8428	25500	0.1220	0.8451	0.8716	0.8582	0.9849
0.0013	15.1339	26000	0.1209	0.8608	0.8598	0.8603	0.9847
0.0011	15.4249	26500	0.1261	0.8457	0.8637	0.8546	0.9847
0.0011	15.7159	27000	0.1273	0.8510	0.8616	0.8563	0.9846
0.0022	16.0070	27500	0.1282	0.8431	0.8738	0.8582	0.9847
0.001	16.2980	28000	0.1357	0.8451	0.8628	0.8539	0.9842
0.0013	16.5891	28500	0.1301	0.8465	0.8658	0.8561	0.9843
0.0008	16.8801	29000	0.1335	0.8533	0.8678	0.8605	0.9845
0.0011	17.1711	29500	0.1338	0.8572	0.8654	0.8613	0.9846
0.0006	17.4622	30000	0.1368	0.8561	0.8628	0.8594	0.9847
0.0008	17.7532	30500	0.1359	0.8579	0.8661	0.8620	0.9848

Framework versions

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