nosql-identifier-electra

This model is a fine-tuned version of bhadresh-savani/electra-base-emotion on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2416
• Accuracy: 0.925

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: 2e-05
• 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
No log	1.0	40	0.5918	0.725
No log	2.0	80	0.5009	0.85
No log	3.0	120	0.3760	0.9
No log	4.0	160	0.4031	0.8
No log	5.0	200	0.3600	0.875
No log	6.0	240	0.2552	0.925
No log	7.0	280	0.2876	0.9
No log	8.0	320	0.2684	0.9
No log	9.0	360	0.2454	0.95
No log	10.0	400	0.2540	0.925
No log	11.0	440	0.2622	0.9
No log	12.0	480	0.2397	0.95
0.3893	13.0	520	0.2355	0.925
0.3893	14.0	560	0.2432	0.925
0.3893	15.0	600	0.2416	0.925

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.11.0