Example of using LBP with different classifiers

Example image classification using local binary patterns (LBP) for feature extraction and different classifiers, including SVM, Random Forest, and KNeighbors, using GridSearchCV to find the best hyperparameters for each classifier.

Requirements

The following packages are required

• numpy
• opencv-python
• scikit-image
• scikit-learn
• matplotlib
• seaborn

The dataset must be a folder with a subfolder for each class containing images

To create a dataset from the source code of a project, you use the minimaps.py

npm install @tensorflow/tfjs-node

pip install --upgrade tensorflowjs tensorflowjs_converter --input_format=tf_saved_model
--output_format=tfjs_graph_model
--skip_op_check
/home/munif-gebara-junior/ml/learn/codeminimap/modelos/_home_munif-gebara-junior_ml_learn_codeminimap_dataset_novo_encrypted_68b36a01/saved_model
modelos_js/

tensorflowjs_converter --input_format=tf_saved_model
--output_format=tfjs_graph_model
/home/munif-gebara-junior/ml/learn/codeminimap/modelos/_home_munif-gebara-junior_ml_learn_codeminimap_dataset_novo_encrypted_68b36a01/saved_model
modelos_js/

tensorflowjs_converter --input_format keras --skip_op_check modelos/meu_modelo.h5 modelos_js/