Metadata-Version: 2.4
Name: himap
Version: 1.4.0
Summary: HiMAP: Hidden Markov models for Advanced Prognostics!
Author-email: Thanos Kontogiannis <a.kontogiannis@tudelft.nl>, Mariana Salinas-Camus <m.salinascamus@tudelft.nl>, Nick Eleftheroglou <n.eleftheroglou@tudelft.nl>
Maintainer-email: Thanos Kontogiannis <a.kontogiannis@tudelft.nl>, Mariana Salinas-Camus <m.salinascamus@tudelft.nl>
License-Expression: Apache-2.0
Project-URL: Documentation, https://himap.readthedocs.io/en/latest/
Project-URL: Repository, https://github.com/GroupiSP/himap/
Keywords: Hidden Markov Models,Prognostics
Classifier: Development Status :: 5 - Production/Stable
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Requires-Python: <3.13,>=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: matplotlib>=3.7.1
Requires-Dist: numpy>=1.23.5
Requires-Dist: scikit_learn>=1.3.0
Requires-Dist: scipy>=1.10.0
Requires-Dist: tqdm>=4.65.0
Requires-Dist: numba>=0.60
Requires-Dist: pandas>=1.5.3
Requires-Dist: Cython>=0.29.35
Provides-Extra: test
Requires-Dist: pytest>=8.0; extra == "test"
Dynamic: license-file

# HiMAP:  Hidden Markov models for Advanced Prognostics

HiMAP is a Python package for implementing hidden Markov Models (HMMs) and hidden semi-Markov Models (HSMMs) tailored for prognostic applications. It provides a probabilistic framework for predicting Remaining Useful Life (RUL) and modeling complex degradation processes without requiring labeled datasets.

## Table of Contents

- [Citing](#citing)
- [Configuration and Installation](#installation)
- [Data Structure](#structure)
- [Example](#example)
- [Contributors](#contributors)

## Citing
[![DOI](https://zenodo.org/badge/932810938.svg)](https://doi.org/10.5281/zenodo.18418215)

If you use **HiMAP**, please cite the software using the DOI for the release you used (recommended for reproducibility). Zenodo also provides a concept DOI that represents all versions.

- Version DOI (recommended): https://doi.org/10.5281/zenodo.18418216 
- Concept DOI (all versions): https://doi.org/10.5281/zenodo.18418215 

You can also use the citation information provided by GitHub via the **“Cite this repository”** button (generated from `CITATION.cff`).

## Requirements
> [!WARNING]
> A C++ compiler is required to build the .pyx files.

### Windows users:
Microsoft Visual C/C++ (MSVC) 14.0 or higher is required to build the .pyx files.

https://visualstudio.microsoft.com/visual-cpp-build-tools/ 

(Download Build Tool - After Visual Studio Installer is ready, choose Desktop development with C++)

### Linux users:
The GNU C Compiler (gcc) is usually present. Next to a C compiler, Cython requires the Python header files. 
On Ubuntu or Debian run the following command:

```
sudo apt-get install build-essential python3-dev
```

>[!Note]
>For more information refer to the Cython package documentation:
>
>https://cython.readthedocs.io/en/latest/src/quickstart/install.html

## Installation
You can install HiMAP in two ways:

### Option 1: Install via `pip`

The easiest way to install HiMAP is through `pip`. Simply run the following command:

```bash
pip install himap
```

>[!Note]
>To install the package, you need `Python>=3.9`


### Option 2: Install from Source Code

If you prefer to install HiMAP directly from the source, follow these steps:

1. Create a virtual environment and activate it. (This example will be demonstrated with Anaconda, but it is not required.)

  Step 1a

```
conda create -n himap_env python=3.12
```

  Step 1b

```
conda activate himap_env
```


2. This repository can be directly pulled through GitHub by the following commands:

  Step 2a
```
conda install git
```

  Step 2b
```
git clone https://github.com/GroupiSP/himap.git
```


  Step 2c
```
cd himap
```

3. The dependencies can be installed using the requirements.txt file
```
pip install -r requirements.txt
```

4. To compile the Cython code, run the following commands:
```
python setup_cython.py build_ext --inplace
```

>[!Note]
>For detailed usage instructions, guides, and API references, please visit our comprehensive documentation: [Read the docs](https://himap.readthedocs.io/en/latest/)

## Structure


```
../root/
      └── LICENSE
      └── README.md
      └── requirements.txt
      └── ...

    
      ├── himap/                                                        -- Required
          └── ab.py                                                     -- Required
          └── base.py                                                   -- Required
          └── main.py                                                   -- Required
          └── plot.py                                                   -- Required
          └── smoothed.pyd                                              -- Required
          └── utils.py                                                  -- Required
          
          ├── cython_build/                                             -- Required
              └── __init__.py                                           -- Required      
              └── fwd_bwd.pyx                                           -- Required
              └── setup.py                                              -- Required


          ├── example_data/                                             -- Required      
              └── test_FD001_disc_20_mod.csv                            -- Required
              └── train_FD001_disc_20_mod.csv                           -- Required

          ├── results/                                                  -- Automatically generated      
              ├── dictionaries                                          -- Automatically generated
              ├── figures/                                              -- Automatically generated
              ├── models/                                               -- Automatically generated
```

## Example


### Run from the rebuild repo

To describe how to train and use the HMM and HSMM models, we show an example below. To run the code from the terminal with default values, use the following command from the root directory of the repository:

```
python -m himap.main
```

This runs the HMM model for the C-MAPSS dataset by default and fits the best model utilizing the Bayesian Information Criterion.

If you want to fit the HSMM model to the C-MAPSS data run the command:

```
python -m himap.main --hsmm True 
```

If you want to run the example utilizing Monte Carlo Sampling generated data run the command:

```
python -m himap.main --mc_sampling True
```

See the `main.py` file for different existing variables and options.

### Run from the installed package

The example can be also run via the distribution. After installing the package in the virtual environment you can run

```
python -m himap.main
```
and use the same arguments as previously for the different example options.

>[!Note]
>The CLI is used only for demonstration purposes. For more control and customization, you can directly import 
> the relevant functions and classes from the `himap` package in your own scripts or Jupyter notebooks.

## Results

The results are saved inside the directory `root/himap/results/`

## Contributors

- [Thanos Kontogiannis](https://github.com/thanoskont)
- [Mariana Salinas-Camus](https://github.com/mariana-sc)
- [Nick Eleftheroglou](https://www.tudelft.nl/staff/n.eleftheroglou/)

## License
As of v1.1.0 (2025-08-25), **HiMAP** is licensed under the **Apache License 2.0**.
Prior releases remain under their original licenses.

See [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE).
