metadata

base_model: haryoaw/scenario-TCR-NER_data-univner_half
library_name: transformers
license: mit
metrics:
  - precision
  - recall
  - f1
  - accuracy
tags:
  - generated_from_trainer
model-index:
  - name: scenario-non-kd-pre-ner-full-xlmr_data-univner_half55
    results: []

scenario-non-kd-pre-ner-full-xlmr_data-univner_half55

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

Loss: 0.1450
Precision: 0.8516
Recall: 0.8357
F1: 0.8436
Accuracy: 0.9832

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
0.0106	0.5828	500	0.0804	0.8385	0.8489	0.8437	0.9838
0.0101	1.1655	1000	0.0863	0.8549	0.8432	0.8490	0.9838
0.009	1.7483	1500	0.0960	0.8282	0.8557	0.8418	0.9829
0.0077	2.3310	2000	0.1040	0.8320	0.8498	0.8408	0.9827
0.0077	2.9138	2500	0.0931	0.8461	0.8478	0.8469	0.9835
0.0054	3.4965	3000	0.0982	0.8482	0.8523	0.8502	0.9843
0.0061	4.0793	3500	0.1075	0.8450	0.8355	0.8402	0.9832
0.0053	4.6620	4000	0.1068	0.8409	0.8536	0.8472	0.9839
0.0047	5.2448	4500	0.1073	0.8372	0.8592	0.8480	0.9837
0.0047	5.8275	5000	0.1100	0.8451	0.8585	0.8517	0.9839
0.0034	6.4103	5500	0.1167	0.8350	0.8497	0.8422	0.9832
0.0034	6.9930	6000	0.1122	0.8405	0.8478	0.8441	0.9836
0.0035	7.5758	6500	0.1125	0.8424	0.8419	0.8421	0.9834
0.0032	8.1585	7000	0.1145	0.8454	0.8504	0.8479	0.9836
0.0035	8.7413	7500	0.1075	0.8499	0.8407	0.8453	0.9838
0.0027	9.3240	8000	0.1213	0.8493	0.8384	0.8438	0.9837
0.0031	9.9068	8500	0.1083	0.8551	0.8440	0.8495	0.9842
0.0027	10.4895	9000	0.1273	0.8329	0.8639	0.8482	0.9835
0.0024	11.0723	9500	0.1247	0.8478	0.8411	0.8444	0.9834
0.0021	11.6550	10000	0.1161	0.8487	0.8378	0.8432	0.9838
0.0019	12.2378	10500	0.1284	0.8316	0.8556	0.8434	0.9830
0.0021	12.8205	11000	0.1208	0.8492	0.8510	0.8501	0.9840
0.0015	13.4033	11500	0.1266	0.8374	0.8499	0.8436	0.9830
0.002	13.9860	12000	0.1236	0.8403	0.8530	0.8466	0.9832
0.0016	14.5688	12500	0.1313	0.8453	0.8409	0.8430	0.9833
0.0013	15.1515	13000	0.1362	0.8460	0.8482	0.8471	0.9835
0.0015	15.7343	13500	0.1246	0.8480	0.8511	0.8496	0.9840
0.0012	16.3170	14000	0.1335	0.8549	0.8423	0.8485	0.9837
0.0014	16.8998	14500	0.1265	0.8445	0.8433	0.8439	0.9833
0.0009	17.4825	15000	0.1450	0.8516	0.8357	0.8436	0.9832

Framework versions

Transformers 4.44.2
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.19.1