fine-tune-vanilla-bert-base-uncased-ch9

This model is a fine-tuned version of omersubasi/bert-base-uncased-issues-128 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1664
• Micro f1: 0.7308
• Macro f1: 0.6418

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: 3e-05
• train_batch_size: 4
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Micro f1	Macro f1
0.3943	1.0	56	0.3426	0.0	0.0
0.3165	2.0	112	0.3111	0.2857	0.1010
0.2701	3.0	168	0.2531	0.5	0.2266
0.2019	4.0	224	0.2155	0.6196	0.3375
0.1544	5.0	280	0.2094	0.6064	0.4363
0.1135	6.0	336	0.1829	0.7030	0.5914
0.0823	7.0	392	0.1774	0.6970	0.5956
0.0619	8.0	448	0.1781	0.6965	0.6130
0.0491	9.0	504	0.1695	0.7327	0.6402
0.0419	10.0	560	0.1664	0.7308	0.6418

Framework versions

• Transformers 4.16.2
• Pytorch 2.1.0+cu118
• Datasets 1.16.1
• Tokenizers 0.15.0