output

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: 0.3471
• Accuracy: 0.8672

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	0.3968	0.8594
0.4221	2.0	612	0.3889	0.8594
0.4221	3.0	918	0.3814	0.8594
0.4026	4.0	1224	0.3775	0.8594
0.396	5.0	1530	0.3724	0.8594
0.396	6.0	1836	0.3707	0.8594
0.392	7.0	2142	0.3721	0.8594
0.392	8.0	2448	0.3653	0.8594
0.3898	9.0	2754	0.3765	0.8613
0.3835	10.0	3060	0.3572	0.8594
0.3835	11.0	3366	0.3664	0.8613
0.3869	12.0	3672	0.3568	0.8613
0.3869	13.0	3978	0.3583	0.8613
0.3825	14.0	4284	0.3526	0.8613
0.3813	15.0	4590	0.3581	0.8613
0.3813	16.0	4896	0.3553	0.8613
0.3759	17.0	5202	0.3504	0.8613
0.3788	18.0	5508	0.3490	0.8613
0.3788	19.0	5814	0.3520	0.8613
0.3754	20.0	6120	0.3450	0.8613
0.3754	21.0	6426	0.3494	0.8633
0.3748	22.0	6732	0.3491	0.8633
0.3775	23.0	7038	0.3499	0.8633
0.3775	24.0	7344	0.3494	0.8633
0.3748	25.0	7650	0.3504	0.8672
0.3748	26.0	7956	0.3495	0.8672
0.3701	27.0	8262	0.3454	0.8633
0.3712	28.0	8568	0.3472	0.8633
0.3712	29.0	8874	0.3478	0.8672
0.3751	30.0	9180	0.3471	0.8672

Framework versions

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