bpmn-task-extractor

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

• Loss: 0.0970
• Precision: 0.95
• Recall: 0.95
• F1: 0.9500
• Accuracy: 0.9888

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: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	1	1.0813	0.3077	0.2	0.2424	0.6404
No log	2.0	2	0.7296	0.4783	0.55	0.5116	0.7191
No log	3.0	3	0.5097	0.6111	0.55	0.5789	0.8090
No log	4.0	4	0.3683	0.7059	0.6	0.6486	0.8652
No log	5.0	5	0.2926	0.75	0.6	0.6667	0.8539
No log	6.0	6	0.2268	0.7647	0.65	0.7027	0.8764
No log	7.0	7	0.1699	0.7778	0.7	0.7368	0.9101
No log	8.0	8	0.1273	0.8	0.8	0.8000	0.9438
No log	9.0	9	0.1061	0.95	0.95	0.9500	0.9888
No log	10.0	10	0.0970	0.95	0.95	0.9500	0.9888

Framework versions

• Transformers 4.21.3
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1