distilroberta-rbm231k-ep20-op40-phrase5k

This model is a fine-tuned version of judy93536/distilroberta-rbm231k-ep20-op40 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1735
• Accuracy: 0.9459

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: 1.113335054745316e-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.28
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	250	1.0641	0.6156
1.0561	2.0	500	0.9262	0.6136
1.0561	3.0	750	0.8124	0.6136
0.8474	4.0	1000	0.7271	0.6476
0.8474	5.0	1250	0.6283	0.7237
0.6495	6.0	1500	0.5066	0.7758
0.6495	7.0	1750	0.3436	0.8979
0.3657	8.0	2000	0.2598	0.9189
0.3657	9.0	2250	0.2117	0.9299
0.2023	10.0	2500	0.2022	0.9279
0.2023	11.0	2750	0.1787	0.9329
0.1518	12.0	3000	0.1632	0.9429
0.1518	13.0	3250	0.1883	0.9369
0.1324	14.0	3500	0.1699	0.9409
0.1324	15.0	3750	0.1704	0.9439
0.1225	16.0	4000	0.1755	0.9429
0.1225	17.0	4250	0.1743	0.9439
0.119	18.0	4500	0.1726	0.9459
0.119	19.0	4750	0.1723	0.9459
0.1145	20.0	5000	0.1735	0.9459

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0