Analysis of Images and Text

Image and text analysis performed using various classification and anomaly detection techniques.

Table of Contents

1. About The Project
   • Structure
   • Built With
2. Getting Started
   • Prerequisites
   • Installation
3. Usage
4. License
5. Contact

About The Project

The project carried out for the course "Machine and Deep Learning," consists of analyzing two types of datasets: images and text, using different machine learning algorithms.

Structure

• models_images: Folder with pre-trained models for images.
• models_text: Folder with pre-trained models for texts.
• notebook: Folder with notebooks used in the development of the project.
• accuracy_images_script.py: script to compute accuracy on custom images dataset with the pretrained models.
• accuracy_text_script.py: script to compute accuracy on custom texts dataset with the pretrained models

Built With

• scikit learn
• Keras
• TensorFlow

Getting Started

To get a local copy up and running follow these simple example steps.

Prerequisites

Install the following libraries.

• requirements.txt

    pip install -r requirements.txt

Installation

1. Clone the repo

   git clone https://github.com/GiuseTripodi/Image_Text_Analysis.git

Usage

Run one of the two scripts:

• accuracy_images_script.py: If you want to calculate accuracy on images with pretrained models.
• accuracy_text_script.py: If you want to calculate accuracy on text with pretrained models.

Information about script.

  usage: accuracy_images_script.py [-h] [-s SCALE] [--version] images_path models_path

  positional arguments:
    images_path           path of the directory with all the images
    models_path           path of the directory with all the images

  optional arguments:
    -h, --help            show this help message and exit
    -s SCALE, --scale SCALE
                      Percentage of the data to get, is a value between (0,1)
    --version             show program's version number and exit

To run the script:

  python <script> [-s] <images_path> <models_path>

Example:

   python accuracy_images_script.py --scale 0.05  "/home/Dataset/images" "/home/models_images"

License

Distributed under the MIT License. See LICENSE.txt for more information.

Contact

Giuseppe Tripodi - @giuseppetripod3 - giuseppetripodi1@outlook.it - LinkedIn