nruigrok/local_politics

my_awesome_model

This model is a fine-tuned version of GroNLP/bert-base-dutch-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4758
• F1: 0.9017

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: 48
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	220	0.1520	0.8825
No log	2.0	440	0.2150	0.9041
0.1543	3.0	660	0.2199	0.9104
0.1543	4.0	880	0.3506	0.8988
0.0418	5.0	1100	0.4256	0.8930
0.0418	6.0	1320	0.3492	0.9037
0.0178	7.0	1540	0.3244	0.9092
0.0178	8.0	1760	0.4016	0.9044
0.0178	9.0	1980	0.4155	0.8985
0.0135	10.0	2200	0.4235	0.9008
0.0135	11.0	2420	0.4109	0.9084
0.0084	12.0	2640	0.4400	0.9032
0.0084	13.0	2860	0.4518	0.9064
0.0046	14.0	3080	0.4740	0.9017
0.0046	15.0	3300	0.4758	0.9017

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3