Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/kentonishi/blanksort
The Python package for the BlankSort keyword extraction algorithm.
https://github.com/kentonishi/blanksort
Last synced: 7 days ago
JSON representation
The Python package for the BlankSort keyword extraction algorithm.
- Host: GitHub
- URL: https://github.com/kentonishi/blanksort
- Owner: KentoNishi
- License: other
- Created: 2020-03-09T15:29:03.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2020-09-30T18:45:16.000Z (about 4 years ago)
- Last Synced: 2024-10-12T04:28:29.717Z (about 1 month ago)
- Language: Python
- Homepage: https://pypi.org/project/blanksort/
- Size: 19.5 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# [BlankSort](https://pypi.org/project/blanksort/)
The Python package for the BlankSort keyword extraction algorithm.
## Installation
Use the following command to install from TestPyPI:
```shell
pip install --upgrade blanksort
```
## Usage
BlankSort can be imported like any other Python package:
```python
import blanksort
```
After importing, you can create a `BlankSort` object.
```python
algo = blanksort.BlankSort()
```
Alternatively, you can use the following code to create the `BlankSort` object.
```python
from BlankSort import *
algo = BlankSort()
```
The `BlankSort` constructor accepts the following arguments:
* `binary_path` [**optional**]
* The path to the binary folder that contains `blanksort.database` and `stopwords-en.txt`. If not specified, the default binary files will be downloaded.
* `preloadVectors` [**optional keyword argument**]
* A boolean (default `False`) that specifies if word vectors should be pre-loaded into memory.
* `saveGeneratedVectors` [**optional keyword argument**]
* A boolean (default `False`) that specifies if word vectors generated on-the-fly should be saved to `blanksort.database`.
### Example
```python
import blanksort
algo = blanksort.BlankSort()
keywords = algo.rank("[input text]")
```
## Binary Files
The BlankSort package requires several files, collectively referred to as "binaries".
| File name | Description |
|:---------------------|:---------------------------------------------------|
| `blanksort.database` | A database file created using [SQLiteDict](https://github.com/RaRe-Technologies/sqlitedict). The database contains word vectors and n-grams imported from [FastText](https://fasttext.cc/).
| `stopwords-en.txt` | A text file containing a list of stop words for pre-processing. Each line should contain a single word. |
If the files are not found in the `binary_path`, the package will download the default binary folder from the latest version of `binaries.zip` attached to a release.