robbert_dataaugmentation

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: 0.7814
• Precisions: 0.8515
• Recall: 0.8094
• F-measure: 0.8265
• Accuracy: 0.9039

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: 42
• 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
0.5813	1.0	285	0.4311	0.7695	0.7413	0.7537	0.8704
0.2533	2.0	570	0.4952	0.8339	0.7501	0.7745	0.8801
0.1216	3.0	855	0.5067	0.8403	0.7968	0.8148	0.8932
0.0685	4.0	1140	0.6121	0.8041	0.7972	0.7963	0.8886
0.0478	5.0	1425	0.6603	0.8239	0.7820	0.7983	0.8893
0.0294	6.0	1710	0.7029	0.8190	0.8029	0.8083	0.8954
0.0147	7.0	1995	0.7219	0.8332	0.8198	0.8227	0.8991
0.0142	8.0	2280	0.7702	0.8330	0.7953	0.8109	0.8961
0.0099	9.0	2565	0.7670	0.8340	0.7943	0.8086	0.8972
0.0044	10.0	2850	0.8132	0.8434	0.8026	0.8193	0.9025
0.0058	11.0	3135	0.7757	0.8468	0.8100	0.8253	0.9033
0.0046	12.0	3420	0.7814	0.8515	0.8094	0.8265	0.9039
0.0029	13.0	3705	0.8057	0.8494	0.8046	0.8229	0.9029
0.0012	14.0	3990	0.7994	0.8492	0.8047	0.8230	0.9031

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1