bert-base-uncased-finetuned-math_punctuation-25-01-two_linear_layers-frozen_bert

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

• Loss: 0.2150
• Micro f1: 0.8910
• Macro f1: 0.2672
• Weighted f1: 0.8495

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Micro f1	Macro f1	Weighted f1
0.193	0.62	500	0.2146	0.8937	0.2360	0.8435
0.1936	1.23	1000	0.2130	0.8937	0.2360	0.8435
0.1924	1.85	1500	0.2119	0.8937	0.2361	0.8435
0.1911	2.47	2000	0.2128	0.8936	0.2369	0.8437
0.1909	3.09	2500	0.2114	0.8937	0.2369	0.8437
0.1904	3.7	3000	0.2137	0.8935	0.2407	0.8445
0.1935	4.32	3500	0.2138	0.8934	0.2469	0.8458
0.1874	4.94	4000	0.2118	0.8929	0.2561	0.8479
0.1908	5.56	4500	0.2134	0.8925	0.2588	0.8483
0.1877	6.17	5000	0.2135	0.8918	0.2628	0.8490
0.1881	6.79	5500	0.2133	0.8931	0.2554	0.8478
0.1902	7.41	6000	0.2137	0.8922	0.2603	0.8485
0.1883	8.02	6500	0.2155	0.8914	0.2655	0.8493
0.19	8.64	7000	0.2154	0.8914	0.2647	0.8490
0.1881	9.26	7500	0.2149	0.8915	0.2645	0.8492
0.1876	9.88	8000	0.2141	0.8911	0.2671	0.8496
0.1879	10.49	8500	0.2155	0.8897	0.2722	0.8501
0.1897	11.11	9000	0.2156	0.8910	0.2670	0.8494
0.1883	11.73	9500	0.2150	0.8910	0.2672	0.8495

Framework versions

• Transformers 4.25.1
• Pytorch 2.0.0.dev20230111
• Datasets 2.8.0
• Tokenizers 0.13.2