tiny_bert_25_intents

This model is a fine-tuned version of prajjwal1/bert-tiny on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2718
• Accuracy: 0.9238
• F1: 0.9238

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	472	1.9552	0.6143	0.6143
2.6425	2.0	944	1.1750	0.7952	0.7952
1.5608	3.0	1416	0.7802	0.8667	0.8667
0.9473	4.0	1888	0.5623	0.8976	0.8976
0.6217	5.0	2360	0.4360	0.9167	0.9167
0.4313	6.0	2832	0.3584	0.9143	0.9143
0.3165	7.0	3304	0.3150	0.9238	0.9238
0.237	8.0	3776	0.3006	0.9214	0.9214
0.1674	9.0	4248	0.2799	0.9214	0.9214
0.1691	10.0	4720	0.2647	0.9238	0.9238
0.1263	11.0	5192	0.2742	0.9238	0.9238
0.1201	12.0	5664	0.2745	0.9214	0.9214
0.1132	13.0	6136	0.2641	0.9214	0.9214
0.1045	14.0	6608	0.2728	0.9238	0.9238
0.0918	15.0	7080	0.2718	0.9238	0.9238

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3