factual-consistency-classification-ja-avgpool

This model is a fine-tuned version of line-corporation/line-distilbert-base-japanese on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4881
• Accuracy: 0.8223

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 8
• seed: 42
• distributed_type: tpu
• 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	Accuracy
No log	1.0	306	0.6837	0.7402
0.7763	2.0	612	0.6102	0.7734
0.7763	3.0	918	0.5782	0.7832
0.657	4.0	1224	0.5698	0.7949
0.6267	5.0	1530	0.5743	0.7793
0.6267	6.0	1836	0.5465	0.8066
0.6082	7.0	2142	0.5474	0.8066
0.6082	8.0	2448	0.5488	0.7949
0.5976	9.0	2754	0.5359	0.8125
0.5845	10.0	3060	0.5236	0.8086
0.5845	11.0	3366	0.5240	0.8027
0.5769	12.0	3672	0.5120	0.8125
0.5769	13.0	3978	0.5105	0.8125
0.5742	14.0	4284	0.5282	0.7969
0.5631	15.0	4590	0.5026	0.8086
0.5631	16.0	4896	0.5120	0.8125
0.5529	17.0	5202	0.4996	0.8145
0.5525	18.0	5508	0.4928	0.8145
0.5525	19.0	5814	0.5143	0.8027
0.5471	20.0	6120	0.4859	0.8203
0.5471	21.0	6426	0.4923	0.8145
0.5397	22.0	6732	0.4874	0.8242
0.5404	23.0	7038	0.4926	0.8184
0.5404	24.0	7344	0.4913	0.8223
0.5375	25.0	7650	0.4914	0.8223
0.5375	26.0	7956	0.4960	0.8047
0.5301	27.0	8262	0.4883	0.8203
0.5313	28.0	8568	0.4890	0.8223
0.5313	29.0	8874	0.4918	0.8203
0.5318	30.0	9180	0.4881	0.8223

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.0+cu118
• Datasets 2.14.5
• Tokenizers 0.14.0