Process Sequence

Directory structure:

project
│   README.md
│   requirement.txt: to run experiment in python 2.7
|   requirement2.txt: to run in python 3.5   
│   report.pptx
|   performance. xlsx
|
└───code: code of reference 
│   │   train.py
|   |   evaluate_next_activity_and_time.py
│   │   evaluate_suffix_and_remaining_time.py
│   │   calculate_accuracy_on_next_event.py
|   |
│   └───output_files
| 
└───data
|   │   helpdesk.csv: full data for analysis
|
└───experiment
|   |   create_df.ipynb
|   |   prepare_data.ipynb
|
└───data_preprocessing
|   |   create_df.ipynb
|   |   prepare_input.ipynb
|
└───input
|
└───experiment
|
└───benchmark
|
└───reference

Reference

The code of reference model was implemented in Python 2.7. To run:

1. Install requirement

pip install -r requirements.txt

2. Create dirs for output:

mkdir code/output_files/folds

mkdir code/output_files/models

mkdir code/output_files/results

3. Run python train.py
4. Run python evaluate_next_activity_and_time.py and python evaluate_suffix_and_remaining_time.py
5. Run python calculate_accuracy_on_next_event.py