robbert_seed36_1311

This model is a fine-tuned version of pdelobelle/robbert-v2-dutch-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3538
• Precisions: 0.8351
• Recall: 0.8079
• F-measure: 0.8173
• Accuracy: 0.9422

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: 7.5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 36
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Precisions	Recall	F-measure	Accuracy
0.4364	1.0	236	0.2547	0.8525	0.7285	0.7372	0.9231
0.2196	2.0	472	0.2772	0.8456	0.7521	0.7718	0.9291
0.1273	3.0	708	0.2681	0.8056	0.7798	0.7897	0.9315
0.0799	4.0	944	0.2971	0.8835	0.7898	0.8158	0.9393
0.0541	5.0	1180	0.3302	0.8515	0.7815	0.8016	0.9373
0.0358	6.0	1416	0.3291	0.8140	0.7901	0.7994	0.9385
0.0217	7.0	1652	0.3538	0.8351	0.8079	0.8173	0.9422
0.0145	8.0	1888	0.3622	0.8331	0.8000	0.8113	0.9431
0.0092	9.0	2124	0.3782	0.8190	0.8098	0.8116	0.9402
0.0091	10.0	2360	0.4023	0.8499	0.7967	0.8149	0.9422
0.0068	11.0	2596	0.3932	0.8293	0.8062	0.8154	0.9409
0.0053	12.0	2832	0.3894	0.8415	0.7942	0.8108	0.9412
0.0023	13.0	3068	0.3910	0.8379	0.7987	0.8127	0.9426
0.0035	14.0	3304	0.3919	0.8349	0.7990	0.8110	0.9422

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1