metadata

library_name: transformers
license: mit
base_model: haryoaw/scenario-TCR-NER_data-univner_en
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: scenario-kd-po-ner-full_data-univner_full66
    results: []

scenario-kd-po-ner-full_data-univner_full66

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: 0.5267
Precision: 0.7744
Recall: 0.7391
F1: 0.7564
Accuracy: 0.9807

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: 66
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
0.8089	1.2755	500	0.7185	0.7338	0.6791	0.7054	0.9767
0.4626	2.5510	1000	0.6447	0.7127	0.7319	0.7222	0.9787
0.3791	3.8265	1500	0.5975	0.7349	0.7288	0.7318	0.9794
0.3262	5.1020	2000	0.5889	0.7447	0.7277	0.7361	0.9797
0.2868	6.3776	2500	0.5714	0.7427	0.7381	0.7404	0.9799
0.2587	7.6531	3000	0.5688	0.7703	0.7257	0.7473	0.9807
0.2389	8.9286	3500	0.5610	0.7338	0.7246	0.7292	0.9791
0.2211	10.2041	4000	0.5571	0.7719	0.7495	0.7605	0.9800
0.2022	11.4796	4500	0.5692	0.776	0.7029	0.7376	0.9799
0.1903	12.7551	5000	0.5554	0.7711	0.7360	0.7532	0.9804
0.179	14.0306	5500	0.5411	0.7574	0.7371	0.7471	0.9803
0.1688	15.3061	6000	0.5353	0.7602	0.7516	0.7559	0.9804
0.1608	16.5816	6500	0.5383	0.7748	0.7267	0.75	0.9802
0.1552	17.8571	7000	0.5223	0.7716	0.7381	0.7545	0.9800
0.1489	19.1327	7500	0.5300	0.7721	0.7329	0.7520	0.9801
0.1439	20.4082	8000	0.5321	0.7634	0.7246	0.7435	0.9797
0.1391	21.6837	8500	0.5204	0.7798	0.7443	0.7617	0.9805
0.1351	22.9592	9000	0.5251	0.7489	0.7350	0.7419	0.9800
0.131	24.2347	9500	0.5164	0.7664	0.7505	0.7584	0.9808
0.1291	25.5102	10000	0.5216	0.7614	0.7236	0.7420	0.9798
0.1276	26.7857	10500	0.5257	0.7739	0.7371	0.7550	0.9804
0.1251	28.0612	11000	0.5156	0.7692	0.7453	0.7571	0.9808
0.1241	29.3367	11500	0.5267	0.7744	0.7391	0.7564	0.9807

Framework versions

Transformers 4.44.2
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.19.1