Skip to main content

a package for multi-label classify

Project description

# multi-label-learn

mlleran is a python library for multi-label classification bulti on scikit-learn and numpy.

## Implementation
The implementation is based on the paper [A Review on Multi-Label Learning Algorithms](https://ieeexplore.ieee.org/document/6471714/), and the implementated algorithms include:

**Problem Transformation**

- [x] Binary Relevance
- [x] Classifier Chains
- [x] Calibrated Label Ranking
- [x] Random k-Labelsets

**Algorithm Adaptation**

- [x] Multi-Label k-Nearest Neighbor
- [x] Multi-Label Decision Tree
- [ ] Ranking Support Vector Machine
- [ ] Collective Multi-Label Classifier

## Installation
```bash
pip install mllearn
```
**Note: Support Python3 only.**

## Data Format
All data type should be `ndarray`, especially y should be the binary format. For example, if your dataset totally have 5 labels and one of your samples has only first and last labels, then the corresponding output should be `[1, 0, 0, 0, 1]`.
```python
samples, features = X_train.shape
samples, labels = y_train.shape
samples_test, features = X_test.shape
samples_test, labels = y_test.shape
```
You can also find multi-label dataset provided by Mulan [here](http://mulan.sourceforge.net/datasets-mlc.html).

## Example Usage
This library includes 2 parts, algorithms and metrics.
```python
from mllearn.problem_transform import BinaryRelevance

classif = BinaryRelevance()
classif.fit(X_train, y_train)
predictions = classif.predict(X_test)
```

```python
from mllearn.metrics import subset_acc
acc = subset_acc(y_test, predictions)
```


Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for mllearn, version 1.2.3
Filename, size File type Python version Upload date Hashes
Filename, size mllearn-1.2.3-py3-none-any.whl (14.9 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size mllearn-1.2.3.tar.gz (10.8 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page