nosql-identifier-albert

This model is a fine-tuned version of textattack/albert-base-v2-imdb on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4351
• Accuracy: 0.9

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	0.4383	0.8
No log	2.0	80	0.3771	0.875
No log	3.0	120	0.2954	0.9
No log	4.0	160	0.3746	0.85
No log	5.0	200	0.4419	0.875
No log	6.0	240	0.4913	0.825
No log	7.0	280	0.4602	0.875
No log	8.0	320	0.4185	0.925
No log	9.0	360	0.5439	0.875
No log	10.0	400	0.4071	0.9
No log	11.0	440	0.5948	0.85
No log	12.0	480	0.5412	0.875
0.2697	13.0	520	0.5105	0.9
0.2697	14.0	560	0.4590	0.9
0.2697	15.0	600	0.4351	0.9

Framework versions

• Transformers 4.31.0
• Pytorch 1.12.1+cu113
• Datasets 2.13.1
• Tokenizers 0.11.0