An open API service indexing awesome lists of open source software.

https://github.com/vaasudevans/svc-support-vector-classification

Classification Algorithm in SVM (Support Vector Machine) 🐍
https://github.com/vaasudevans/svc-support-vector-classification

python3 sklearn svc svm

Last synced: about 1 month ago
JSON representation

Classification Algorithm in SVM (Support Vector Machine) 🐍

Awesome Lists containing this project

README

          

# Support Vector Classification
Python script to predict Occupancy using SVM in [sklearn](https://scikit-learn.org/stable/modules/svm.html) library.

**Note:**
This script uses 60% Training and 40% testing of the data file.
Random seed used in model_selection.train_test_split is *100.*
Tested in python: 3.5, Pandas: 0.23.4, scikit-learn: 0.20.2
License: *GNU-GPL*

**Installing Dependencies**
```
pip3 install pandas scikit-learn
```
**Data-Insight**
```
$ head Data/430pm_Managed.txt
201611071630 CP:C96GB CHAdeMO Occ
201611071630 CP:C96GB FastAC43 Free
201611071630 CP:C96GB ComboCCS Occ
201611071630 CP:C9QKW CHAdeMO Occ
201611071630 CP:C9QKW FastAC43 Occ
201611071630 CP:C9QKW ComboCCS Occ
201611071630 CP:RC13 CHAdeMO Free
201611071630 CP:RC13 FastAC43 Free
201611071630 CP:RC13 ComboCCS Occ
201611071630 CP:C85WS CHAdeMO Free
```
**Usage**
```
>>> python3 PredictOccupancy.py Data/430pm_Managed.txt

Accuracy: 0.714516129032258
Report:
precision recall f1-score support

Occupied 0.51 0.19 0.28 178
Free 0.74 0.93 0.82 442

micro avg 0.71 0.71 0.71 620
macro avg 0.62 0.56 0.55 620
weighted avg 0.67 0.71 0.67 620

ConfusionMatrix:
[[ 34 144]
[ 33 409]]
```