README.md

metadata

license: mit
base_model: pdelobelle/robbert-v2-dutch-base
tags:
  - generated_from_trainer
metrics:
  - recall
  - accuracy
model-index:
  - name: robbert_dataaugmentation
    results: []

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: 0.6304
• Precisions: 0.8565
• Recall: 0.7966
• F-measure: 0.8182
• Accuracy: 0.9056

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
0.5972	1.0	284	0.4291	0.7808	0.7248	0.7363	0.8652
0.2753	2.0	568	0.4249	0.7837	0.7521	0.7570	0.8811
0.1379	3.0	852	0.5021	0.8379	0.7750	0.7955	0.8815
0.0776	4.0	1136	0.6344	0.8567	0.7657	0.7907	0.8842
0.0401	5.0	1420	0.6621	0.8442	0.7622	0.7856	0.8884
0.0319	6.0	1704	0.6013	0.8435	0.7870	0.8010	0.8969
0.0205	7.0	1988	0.6304	0.8565	0.7966	0.8182	0.9056
0.0138	8.0	2272	0.6804	0.8538	0.7732	0.7896	0.9030
0.0096	9.0	2556	0.7395	0.8274	0.7696	0.7862	0.8923
0.005	10.0	2840	0.7293	0.8531	0.7846	0.8054	0.8967
0.0034	11.0	3124	0.7385	0.8621	0.7929	0.8105	0.9022
0.0047	12.0	3408	0.7428	0.8575	0.7953	0.8155	0.9061
0.004	13.0	3692	0.7524	0.8617	0.7954	0.8152	0.9024
0.0023	14.0	3976	0.7515	0.8636	0.7957	0.8174	0.9041

Framework versions

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