BERT_large_with_preprocessing_grid_search

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

• Loss: 1.0365
• Precision: 0.8410
• Recall: 0.8308
• F1: 0.8352
• Accuracy: 0.8753

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.9616	1.0	510	0.6482	0.7704	0.8009	0.7781	0.8360
0.4395	2.0	1020	0.7507	0.8422	0.7993	0.8157	0.8552
0.2995	3.0	1530	0.7064	0.8445	0.8213	0.8287	0.8684
0.2117	4.0	2040	0.7889	0.8262	0.8325	0.8245	0.8679
0.1805	5.0	2550	0.9295	0.8406	0.8161	0.8271	0.8670
0.1225	6.0	3060	0.9491	0.8429	0.8260	0.8333	0.8758
0.0983	7.0	3570	0.9901	0.8444	0.8299	0.8359	0.8773
0.0869	8.0	4080	1.0300	0.8377	0.8278	0.8319	0.8719
0.0745	9.0	4590	1.0220	0.8439	0.8341	0.8379	0.8773
0.0591	10.0	5100	1.0365	0.8410	0.8308	0.8352	0.8753

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3