https://github.com/neurodata/preproject_sporf
Pre-project SPORF
https://github.com/neurodata/preproject_sporf
Last synced: 9 months ago
JSON representation
Pre-project SPORF
- Host: GitHub
- URL: https://github.com/neurodata/preproject_sporf
- Owner: neurodata
- License: apache-2.0
- Created: 2019-11-08T18:58:16.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2019-11-08T20:38:58.000Z (over 6 years ago)
- Last Synced: 2025-01-07T23:25:40.300Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 59.6 KB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SPORF Preprocessing
This repo holds code to preprocess your data in a sporf-like manner.
DISCLAMER: This will increase the size of your data.
## To compile:
Run
First, make virtualenv and activate
```
cd sporf_transform
pip install sklearn
pip install -e .
```
## To Use:
Run
```python
from sporf_transform import sporf_ternaryColumns
"""
Suppose X_{n \times p} is your data matrix.
p is the number of features in your data matrix.
d is the number of new features created from sparse-oblique projections.
lam is the parameter in a Pois(lam) + 1 that determins the sparsity.
"""
p = 12
d = 12**4
lam = 1
A = sporf_ternaryColumns(p,d,lam)
### Create the projected data
# Xt = X @ A
```
## [Example01](sporf_transform/docs/Example01.ipynb)