KlemPear/bert-base-uncased-git-issues-300-classifier

results

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1328
• Micro f1: 0.8436
• Macro f1: 0.6883

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Micro f1	Macro f1
0.3843	1.0	53	0.3015	0.2520	0.0889
0.2851	2.0	106	0.2373	0.5660	0.2041
0.2167	3.0	159	0.2121	0.4969	0.3009
0.1565	4.0	212	0.1534	0.7273	0.4905
0.1062	5.0	265	0.1417	0.7812	0.6161
0.0817	6.0	318	0.1360	0.7761	0.6147
0.0607	7.0	371	0.1204	0.8190	0.6776
0.0462	8.0	424	0.1240	0.8263	0.6727
0.0382	9.0	477	0.1207	0.8246	0.6682
0.032	10.0	530	0.1292	0.8093	0.6602
0.0266	11.0	583	0.1214	0.8491	0.6921
0.0229	12.0	636	0.1233	0.8451	0.6856
0.0207	13.0	689	0.1204	0.8396	0.6773
0.0181	14.0	742	0.1203	0.8436	0.6817
0.0161	15.0	795	0.1197	0.8372	0.6776
0.0148	16.0	848	0.1278	0.8357	0.6771
0.0135	17.0	901	0.1319	0.8230	0.6520
0.0129	18.0	954	0.1345	0.8491	0.6905
0.0114	19.0	1007	0.1309	0.8436	0.6794
0.0105	20.0	1060	0.1328	0.8436	0.6883

Framework versions

• Transformers 4.32.1
• Pytorch 2.2.0.dev20231111
• Datasets 2.14.6
• Tokenizers 0.13.3