Code Monkey home page Code Monkey logo

likarajo / currencynote_authenticity Goto Github PK

View Code? Open in Web Editor NEW
1.0 2.0 0.0 38 KB

Classification to predict whether a bank currency note is authentic or not based on variance of the image wavelet transformed image, skewness, entropy, and curtosis of the image using Machine Learning classifiers.

Jupyter Notebook 100.00%
classification random-forest-classifier grid-search python3 jupyter-notebook decision-tree-classifier support-vector-classifier

currencynote_authenticity's Introduction

Currency Note Authenticity

Predict whether a bank currency note is authentic or not based on variance, skewness, and curtosis of the wavelet transformed image, and entropy of the image using classifiers.

Background

Machine Learning classifiers: classifiers.md

Goal

  • Binary Classification problem.
  • Use a Decision Tree Classifier.
  • Use a Random Forest Classifier.
  • Use a Support Vector Machine.

Dependencies

  • Pandas
  • Scikit-learn

pip install -r requirements.txt

Dataset

Bank note authetication dataset from UCI archive: https://archive.ics.uci.edu/ml/datasets/banknote+authentication
Saved in: data/banknote_authentication.txt

  1. Variance of Wavelet Transformed image (continuous)
  2. Skewness of Wavelet Transformed image (continuous)
  3. Curtosis of Wavelet Transformed image (continuous)
  4. Entropy of image (continuous)
  5. Class (integer)

Data Preprocessing

  • Preprare features and labels
    • Features = Variance, Skewness, Curtosis, Entropy
    • Label = Class
  • Split data into Training and Test sets
    • Test size = 20%
  • Scale the features

Create, Train and Evaluate model

  • Use Cross Validation training models with different hyperparameter values.
  • Use Grid Search to select the best model with the best accuracy.

Decision Tree Classifier

authentication_DT.ipynb

Grid Search scores:

0.977 (+/-0.023) for {'criterion': 'gini'}
0.981 (+/-0.023) for {'criterion': 'entropy'}

Best parameters:
 {'criterion': 'entropy'}

Training accuracy: 98.08592777085927 %

Test Accuracy: 98.54545454545455 %

Confusion matrix:
 [[154   3]
 [  1 117]]

Classification report:
               precision    recall  f1-score   support

           0       0.99      0.98      0.99       157
           1       0.97      0.99      0.98       118

    accuracy                           0.99       275
   macro avg       0.98      0.99      0.99       275
weighted avg       0.99      0.99      0.99       275

Random Forest Classifier

authentication_RF.ipynb

Grid Search scores:

0.984 (+/-0.033) for {'bootstrap': True, 'criterion': 'gini', 'n_estimators': 5}
0.986 (+/-0.019) for {'bootstrap': True, 'criterion': 'gini', 'n_estimators': 10}
0.993 (+/-0.014) for {'bootstrap': True, 'criterion': 'gini', 'n_estimators': 20}
0.991 (+/-0.011) for {'bootstrap': True, 'criterion': 'gini', 'n_estimators': 40}
0.985 (+/-0.018) for {'bootstrap': True, 'criterion': 'entropy', 'n_estimators': 5}
0.986 (+/-0.016) for {'bootstrap': True, 'criterion': 'entropy', 'n_estimators': 10}
0.986 (+/-0.016) for {'bootstrap': True, 'criterion': 'entropy', 'n_estimators': 20}
0.985 (+/-0.023) for {'bootstrap': True, 'criterion': 'entropy', 'n_estimators': 40}
0.985 (+/-0.022) for {'bootstrap': False, 'criterion': 'gini', 'n_estimators': 5}
0.989 (+/-0.012) for {'bootstrap': False, 'criterion': 'gini', 'n_estimators': 10}
0.992 (+/-0.007) for {'bootstrap': False, 'criterion': 'gini', 'n_estimators': 20}
0.991 (+/-0.015) for {'bootstrap': False, 'criterion': 'gini', 'n_estimators': 40}
0.988 (+/-0.016) for {'bootstrap': False, 'criterion': 'entropy', 'n_estimators': 5}
0.989 (+/-0.014) for {'bootstrap': False, 'criterion': 'entropy', 'n_estimators': 10}
0.989 (+/-0.014) for {'bootstrap': False, 'criterion': 'entropy', 'n_estimators': 20}
0.988 (+/-0.012) for {'bootstrap': False, 'criterion': 'entropy', 'n_estimators': 40}

Best parameters:
 {'bootstrap': True, 'criterion': 'gini', 'n_estimators': 20}

Training accuracy: 99.27023661270236 %

Test Accuracy: 98.9090909090909 %

Confusion matrix:
 [[155   2]
 [  1 117]]

Classification report:
               precision    recall  f1-score   support

           0       0.99      0.99      0.99       157
           1       0.98      0.99      0.99       118

    accuracy                           0.99       275
   macro avg       0.99      0.99      0.99       275
weighted avg       0.99      0.99      0.99       275

Support Vector Machine

authentication_SVM.ipynb

Grid Search scores:

0.974 (+/-0.021) for {'C': 0.01, 'degree': 1, 'gamma': 'auto', 'kernel': 'linear'}
0.941 (+/-0.027) for {'C': 0.01, 'degree': 1, 'gamma': 'auto', 'kernel': 'poly'}
0.924 (+/-0.019) for {'C': 0.01, 'degree': 1, 'gamma': 'auto', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 1, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 1, 'gamma': 'scale', 'kernel': 'linear'}
0.941 (+/-0.027) for {'C': 0.01, 'degree': 1, 'gamma': 'scale', 'kernel': 'poly'}
0.923 (+/-0.021) for {'C': 0.01, 'degree': 1, 'gamma': 'scale', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 1, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 2, 'gamma': 'auto', 'kernel': 'linear'}
0.582 (+/-0.018) for {'C': 0.01, 'degree': 2, 'gamma': 'auto', 'kernel': 'poly'}
0.924 (+/-0.019) for {'C': 0.01, 'degree': 2, 'gamma': 'auto', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 2, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 2, 'gamma': 'scale', 'kernel': 'linear'}
0.582 (+/-0.018) for {'C': 0.01, 'degree': 2, 'gamma': 'scale', 'kernel': 'poly'}
0.923 (+/-0.021) for {'C': 0.01, 'degree': 2, 'gamma': 'scale', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 2, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 3, 'gamma': 'auto', 'kernel': 'linear'}
0.715 (+/-0.023) for {'C': 0.01, 'degree': 3, 'gamma': 'auto', 'kernel': 'poly'}
0.924 (+/-0.019) for {'C': 0.01, 'degree': 3, 'gamma': 'auto', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 3, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 3, 'gamma': 'scale', 'kernel': 'linear'}
0.714 (+/-0.025) for {'C': 0.01, 'degree': 3, 'gamma': 'scale', 'kernel': 'poly'}
0.923 (+/-0.021) for {'C': 0.01, 'degree': 3, 'gamma': 'scale', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 3, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 4, 'gamma': 'auto', 'kernel': 'linear'}
0.649 (+/-0.028) for {'C': 0.01, 'degree': 4, 'gamma': 'auto', 'kernel': 'poly'}
0.924 (+/-0.019) for {'C': 0.01, 'degree': 4, 'gamma': 'auto', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 4, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.974 (+/-0.021) for {'C': 0.01, 'degree': 4, 'gamma': 'scale', 'kernel': 'linear'}
0.649 (+/-0.028) for {'C': 0.01, 'degree': 4, 'gamma': 'scale', 'kernel': 'poly'}
0.923 (+/-0.021) for {'C': 0.01, 'degree': 4, 'gamma': 'scale', 'kernel': 'rbf'}
0.907 (+/-0.029) for {'C': 0.01, 'degree': 4, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 1, 'gamma': 'auto', 'kernel': 'linear'}
0.978 (+/-0.018) for {'C': 0.1, 'degree': 1, 'gamma': 'auto', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 1, 'gamma': 'auto', 'kernel': 'rbf'}
0.854 (+/-0.029) for {'C': 0.1, 'degree': 1, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 1, 'gamma': 'scale', 'kernel': 'linear'}
0.978 (+/-0.018) for {'C': 0.1, 'degree': 1, 'gamma': 'scale', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 1, 'gamma': 'scale', 'kernel': 'rbf'}
0.856 (+/-0.027) for {'C': 0.1, 'degree': 1, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 2, 'gamma': 'auto', 'kernel': 'linear'}
0.603 (+/-0.035) for {'C': 0.1, 'degree': 2, 'gamma': 'auto', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 2, 'gamma': 'auto', 'kernel': 'rbf'}
0.854 (+/-0.029) for {'C': 0.1, 'degree': 2, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 2, 'gamma': 'scale', 'kernel': 'linear'}
0.603 (+/-0.035) for {'C': 0.1, 'degree': 2, 'gamma': 'scale', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 2, 'gamma': 'scale', 'kernel': 'rbf'}
0.856 (+/-0.027) for {'C': 0.1, 'degree': 2, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 3, 'gamma': 'auto', 'kernel': 'linear'}
0.932 (+/-0.045) for {'C': 0.1, 'degree': 3, 'gamma': 'auto', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 3, 'gamma': 'auto', 'kernel': 'rbf'}
0.854 (+/-0.029) for {'C': 0.1, 'degree': 3, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 3, 'gamma': 'scale', 'kernel': 'linear'}
0.933 (+/-0.033) for {'C': 0.1, 'degree': 3, 'gamma': 'scale', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 3, 'gamma': 'scale', 'kernel': 'rbf'}
0.856 (+/-0.027) for {'C': 0.1, 'degree': 3, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 4, 'gamma': 'auto', 'kernel': 'linear'}
0.680 (+/-0.024) for {'C': 0.1, 'degree': 4, 'gamma': 'auto', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 4, 'gamma': 'auto', 'kernel': 'rbf'}
0.854 (+/-0.029) for {'C': 0.1, 'degree': 4, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.980 (+/-0.016) for {'C': 0.1, 'degree': 4, 'gamma': 'scale', 'kernel': 'linear'}
0.680 (+/-0.024) for {'C': 0.1, 'degree': 4, 'gamma': 'scale', 'kernel': 'poly'}
0.988 (+/-0.018) for {'C': 0.1, 'degree': 4, 'gamma': 'scale', 'kernel': 'rbf'}
0.856 (+/-0.027) for {'C': 0.1, 'degree': 4, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 1, 'gamma': 'auto', 'kernel': 'linear'}
0.982 (+/-0.019) for {'C': 1.0, 'degree': 1, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 1, 'gamma': 'auto', 'kernel': 'rbf'}
0.785 (+/-0.050) for {'C': 1.0, 'degree': 1, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 1, 'gamma': 'scale', 'kernel': 'linear'}
0.982 (+/-0.019) for {'C': 1.0, 'degree': 1, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 1, 'gamma': 'scale', 'kernel': 'rbf'}
0.785 (+/-0.046) for {'C': 1.0, 'degree': 1, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 2, 'gamma': 'auto', 'kernel': 'linear'}
0.762 (+/-0.025) for {'C': 1.0, 'degree': 2, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 2, 'gamma': 'auto', 'kernel': 'rbf'}
0.785 (+/-0.050) for {'C': 1.0, 'degree': 2, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 2, 'gamma': 'scale', 'kernel': 'linear'}
0.760 (+/-0.021) for {'C': 1.0, 'degree': 2, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 2, 'gamma': 'scale', 'kernel': 'rbf'}
0.785 (+/-0.046) for {'C': 1.0, 'degree': 2, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 3, 'gamma': 'auto', 'kernel': 'linear'}
0.986 (+/-0.021) for {'C': 1.0, 'degree': 3, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 3, 'gamma': 'auto', 'kernel': 'rbf'}
0.785 (+/-0.050) for {'C': 1.0, 'degree': 3, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 3, 'gamma': 'scale', 'kernel': 'linear'}
0.986 (+/-0.021) for {'C': 1.0, 'degree': 3, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 3, 'gamma': 'scale', 'kernel': 'rbf'}
0.785 (+/-0.046) for {'C': 1.0, 'degree': 3, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 4, 'gamma': 'auto', 'kernel': 'linear'}
0.768 (+/-0.036) for {'C': 1.0, 'degree': 4, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 4, 'gamma': 'auto', 'kernel': 'rbf'}
0.785 (+/-0.050) for {'C': 1.0, 'degree': 4, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.984 (+/-0.017) for {'C': 1.0, 'degree': 4, 'gamma': 'scale', 'kernel': 'linear'}
0.764 (+/-0.034) for {'C': 1.0, 'degree': 4, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 1.0, 'degree': 4, 'gamma': 'scale', 'kernel': 'rbf'}
0.785 (+/-0.046) for {'C': 1.0, 'degree': 4, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 1, 'gamma': 'auto', 'kernel': 'linear'}
0.985 (+/-0.017) for {'C': 10, 'degree': 1, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 1, 'gamma': 'auto', 'kernel': 'rbf'}
0.763 (+/-0.057) for {'C': 10, 'degree': 1, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 1, 'gamma': 'scale', 'kernel': 'linear'}
0.985 (+/-0.016) for {'C': 10, 'degree': 1, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 1, 'gamma': 'scale', 'kernel': 'rbf'}
0.763 (+/-0.063) for {'C': 10, 'degree': 1, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 2, 'gamma': 'auto', 'kernel': 'linear'}
0.796 (+/-0.025) for {'C': 10, 'degree': 2, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 2, 'gamma': 'auto', 'kernel': 'rbf'}
0.763 (+/-0.057) for {'C': 10, 'degree': 2, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 2, 'gamma': 'scale', 'kernel': 'linear'}
0.795 (+/-0.027) for {'C': 10, 'degree': 2, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 2, 'gamma': 'scale', 'kernel': 'rbf'}
0.763 (+/-0.063) for {'C': 10, 'degree': 2, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 3, 'gamma': 'auto', 'kernel': 'linear'}
0.988 (+/-0.014) for {'C': 10, 'degree': 3, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 3, 'gamma': 'auto', 'kernel': 'rbf'}
0.763 (+/-0.057) for {'C': 10, 'degree': 3, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 3, 'gamma': 'scale', 'kernel': 'linear'}
0.988 (+/-0.014) for {'C': 10, 'degree': 3, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 3, 'gamma': 'scale', 'kernel': 'rbf'}
0.763 (+/-0.063) for {'C': 10, 'degree': 3, 'gamma': 'scale', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 4, 'gamma': 'auto', 'kernel': 'linear'}
0.863 (+/-0.019) for {'C': 10, 'degree': 4, 'gamma': 'auto', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 4, 'gamma': 'auto', 'kernel': 'rbf'}
0.763 (+/-0.057) for {'C': 10, 'degree': 4, 'gamma': 'auto', 'kernel': 'sigmoid'}
0.988 (+/-0.016) for {'C': 10, 'degree': 4, 'gamma': 'scale', 'kernel': 'linear'}
0.862 (+/-0.017) for {'C': 10, 'degree': 4, 'gamma': 'scale', 'kernel': 'poly'}
1.000 (+/-0.000) for {'C': 10, 'degree': 4, 'gamma': 'scale', 'kernel': 'rbf'}
0.763 (+/-0.063) for {'C': 10, 'degree': 4, 'gamma': 'scale', 'kernel': 'sigmoid'}

Best parameters:
 {'C': 1.0, 'degree': 1, 'gamma': 'auto', 'kernel': 'rbf'}

Training accuracy: 100.0 %

Test Accuracy: 100.0 %

Confusion matrix:
 [[157   0]
 [  0 118]]

Classification report:
               precision    recall  f1-score   support

           0       1.00      1.00      1.00       157
           1       1.00      1.00      1.00       118

    accuracy                           1.00       275
   macro avg       1.00      1.00      1.00       275
weighted avg       1.00      1.00      1.00       275

K Nearest Neighbors

Grid Search scores:

0.999 (+/-0.004) for {'n_neighbors': 5}
0.999 (+/-0.004) for {'n_neighbors': 10}
0.991 (+/-0.012) for {'n_neighbors': 20}

Best parameters:
 {'n_neighbors': 5}

Training accuracy: 99.90867579908675 %

Test Accuracy: 99.63636363636364 %

Confusion matrix:
 [[153   1]
 [  0 121]]

Classification report:
               precision    recall  f1-score   support

           0       1.00      0.99      1.00       154
           1       0.99      1.00      1.00       121

    accuracy                           1.00       275
   macro avg       1.00      1.00      1.00       275
weighted avg       1.00      1.00      1.00       275

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.