metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
widget:
  - text: >-
      The process starts when the customer enters the shop. The customer then
      takes the product from the shelf. The customer then pays for the product
      and leaves the store.
    example_title: Example 1
  - text: >-
      The process begins when the HR department hires the new employee. Next,
      the new employee completes necessary paperwork and provides documentation
      to the HR department. After the initial task, the HR department performs
      a  decision to determine the employee's role and department assignment.
      The employee is trained on the company's sales processes and systems by
      the Sales department. After the training, the Sales department assigns the
      employee a sales quota and performance goals. Finally, the process ends
      with an 'End' event, when the employee begins their role in the Sales
      department.
    example_title: Example 2
  - text: >-
      The process begins with a 'Start' event, when a customer places an order
      for a product on the company's website. Next, the customer service
      department checks the availability of the product and confirms the order
      with the customer. After the initial task, the warehouse  processes the
      order. If the order is eligible for same-day shipping, the warehouse staff
      picks and packs the order, and it is sent to the shipping department.
      After the order is packed, the shipping department arranges for the order
      to be delivered to the customer. Finally, the process ends with an 'End'
      event, when the customer receives their order.
    example_title: Example 3
base_model: bert-base-cased
model-index:
  - name: bert-finetuned-bpmn
    results: []

bert-finetuned-bpmn

This model is a fine-tuned version of bert-base-cased on a dataset containing textual process descriptions.

The dataset contains 2 target labels:

AGENT
TASK

The dataset (and the notebook used for training) can be found on the following GitHub repo: https://github.com/jtlicardo/bert-finetuned-bpmn

Update: a model trained on 5 BPMN-specific labels can be found here: https://huggingface.co/jtlicardo/bpmn-information-extraction

The model achieves the following results on the evaluation set:

Loss: 0.2656
Precision: 0.7314
Recall: 0.8366
F1: 0.7805
Accuracy: 0.8939

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: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	10	0.8437	0.1899	0.3203	0.2384	0.7005
No log	2.0	20	0.4967	0.5421	0.7582	0.6322	0.8417
No log	3.0	30	0.3403	0.6719	0.8431	0.7478	0.8867
No log	4.0	40	0.2821	0.6923	0.8235	0.7522	0.8903
No log	5.0	50	0.2656	0.7314	0.8366	0.7805	0.8939

Framework versions

Transformers 4.25.1
Pytorch 1.13.0+cu116
Datasets 2.7.1
Tokenizers 0.13.2