scenario-non-kd-po-ner-full-mdeberta_data-univner_en55

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

• Loss: 0.1441
• Precision: 0.8306
• Recall: 0.8427
• F1: 0.8366
• Accuracy: 0.9853

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
0.0022	1.2755	500	0.1340	0.8474	0.7878	0.8165	0.9844
0.0015	2.5510	1000	0.1233	0.8365	0.8261	0.8312	0.9849
0.0021	3.8265	1500	0.1064	0.8406	0.8188	0.8296	0.9851
0.0018	5.1020	2000	0.1190	0.8403	0.8116	0.8257	0.9850
0.0009	6.3776	2500	0.1223	0.8570	0.8251	0.8407	0.9857
0.0014	7.6531	3000	0.1144	0.8218	0.8261	0.8240	0.9844
0.0011	8.9286	3500	0.1146	0.8199	0.8437	0.8316	0.9847
0.0005	10.2041	4000	0.1218	0.8255	0.8375	0.8314	0.9845
0.0007	11.4796	4500	0.1291	0.8147	0.8147	0.8147	0.9839
0.0006	12.7551	5000	0.1328	0.8319	0.8251	0.8285	0.9844
0.0005	14.0306	5500	0.1282	0.8026	0.8292	0.8157	0.9841
0.0005	15.3061	6000	0.1312	0.8217	0.8395	0.8305	0.9849
0.0004	16.5816	6500	0.1298	0.8407	0.8302	0.8354	0.9857
0.0003	17.8571	7000	0.1336	0.8103	0.8489	0.8291	0.9847
0.0003	19.1327	7500	0.1345	0.8162	0.8364	0.8262	0.9839
0.0002	20.4082	8000	0.1349	0.8372	0.8302	0.8337	0.9851
0.0002	21.6837	8500	0.1409	0.8281	0.8427	0.8353	0.9854
0.0001	22.9592	9000	0.1457	0.8304	0.8364	0.8334	0.9849
0.0002	24.2347	9500	0.1417	0.8340	0.8427	0.8383	0.9852
0.0001	25.5102	10000	0.1421	0.8369	0.8395	0.8382	0.9852
0.0001	26.7857	10500	0.1468	0.8345	0.8458	0.8401	0.9853
0.0001	28.0612	11000	0.1440	0.8376	0.8437	0.8406	0.9854
0.0001	29.3367	11500	0.1441	0.8306	0.8427	0.8366	0.9853

Framework versions

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