ANN Classifier Hub

Welcome to this repository! This repository contains various artificial neural network models for different tasks, including the Brain Tumor Classifier, Chess Piece Classifier model and Pepsi OR Coca Cola Classifier.

Models

Airline Passenger Satisfacton Classifier

• Description : The Airline Passenger Satisfacton Classifier model is designed to classify whether a passenger is satisfied from his/her experience in the airline or not. It has been trained on a large dataset and utilizes deep learning techniques to achieve accurate classification.

• Accuracy Score : 0.9199645826917154

• Precision : 0.9186422413793104

• Recall : 0.897132333596422

• F1 Score : 0.9077598828696926

• Model : Available in the folder itself

Brain Tumor Classifier

• Description : The Brain Tumor Classifier model is designed to classify whether a patient is suffering from brain tumor or not. It has been trained on a large dataset of brain images and utilizes deep learning techniques to achieve accurate classification.

• Accuracy Score : 0.832

• Precision : 0.8725490196078431

• Recall : 0.7542372881355932

• Model : Available on Google Drive

Chess Piece Classifier

• Description : The Chess Piece Classifier model is designed to classify images of chess pieces into their respective categories. It has been trained on a large dataset of labeled chess piece images and utilizes deep learning techniques to achieve accurate classification. The model can be used to identify the type of chess piece present in an image, such as King, Queen, Bishop, Knight, Rook, or Pawn.

• Accuracy Score : 0.8253968253968254

• Precision : 0.8398689721270367

• Recall : 0.8253968253968254

• Model : Available on Google Drive

Pepsi OR Coca Cola Classifier

• Description : The Pepsi OR Coca Cola Classifier model is designed to classify a drink as Pepsi or Coca Cola.

• Accuracy Score : 0.8833333333333333

• Precision : 0.96

• Recall : 0.8

• Model : Available on Google Drive

Vehicle Classifier

• Description : The Vehicle Classifier model is designed to classify images of vehicles into different categories, such as car, truck, bike, cycle, bus, helicopter, scooty and plane. It has been trained on a diverse dataset of labeled vehicle images. The model can identify the type of vehicle present in an image, making it useful for applications like traffic analysis, object detection, and more.

• Accuracy Score : 0.3888888888888889

Tomato OR Apple Classifier

• Description : The Tomato OR Apple Classifier model is designed to classify tomatoes and apples.

• Accuracy Score : 0.6804123711340206

• Model : Available on Google Drive

Repository Structure

The repository is organized as follows:

- /Airline Passenger Satisfacton Classifier
  - /data                # data for training and testing
  - /model               # saved model
  - /notebook.ipynb      # jupyter notebook

- /Brain Tumor Classifier
  - /data                # data for training and testing
  - /notebook.ipynb      # jupyter notebook
  - /Brain Tumor.csv     # CSV file

- /Chess Piece Classifier
  - /data                # data for training and testing
  - /notebook.ipynb      # jupyter notebook

- /Pepsi OR Coca Cola Classifier
  - /data                # data for training and testing
  - /notebook.ipynb      # jupyter notebook

- /Tomato OR Apple Classifier
  - /data                # data for training and testing
  - /notebook.ipynb      # jupyter notebook

- /Vehicle Classifier
  - /data                # data for training and testing
  - /notebook.ipynb      # jupyter notebook

- LICENSE                # license

- README.md              # Overview and instructions for using the repository

- requirements.txt       # requirements.txt

Requirements

To use the models, the following libraries must be installed:

torch            [v2.1.0+cu118]
torchaudio       [v2.1.0+cu118]
torchdata        [v0.7.0]
torchsummary     [v1.5.1]
torchtext        [v0.16.0]
torchvision      [v0.16.0+cu118]
matplotlib       [v3.4.3]
OpenCV           [v4.5.3.0]
numpy            [v1.21.2]
scikit-learn     [v1.0]
pandas           [v1.3.3]
joblib           [v1.0.1]
jupyter notebook [v1.0.0]

Run the following command in the terminal to install the libraries mentioned above:

pip3 install -r requirements.txt

Usage

To run and use the models, follow the steps given below:

1) Run Jupyter Notebook using the following command

jupyter notebook

2) Go to the appropriate directory and open the notebook.ipynb file

3) Run all the cells

4) A file savedModel.pkl or *****.pt will be created(or download it from the link provided above in this README.md file), this is the saved model.

Credits

The notebooks and models available in this repository are created and trained by Omanshu.