Iris Flower Classification

Iris flower classification using Machine learning, also referred as Hello World for Machine Learning. It is very basic classification problem which helps understand basic concept of Machine Learning for beginners.

Three class for classification are as follows:

• Iris-setosa
• Iris-versicolor
• Iris-virginica

Implementation in tensorflow
Implementation in scikit-learn will be released soon.

Requirement

To install this package, python3, pip, and virtual environment are required.

For Windows users:
Install python3 and pip
Install virtual environment

For Linux users:

sudo apt-get update
sudo apt-get install python python3-pip
sudo pip3 install virtualenv 

Installation

Navigate to the cloned directory.

cd iris-flower-classification

Create virtual environment for Python.

python -m venv tutorial-env

For Linux users:

python3 -m venv tutorial-env

Here, tutorial-env is the name of the virtual environment, you can name it as you like.

Activate virtual environment by the following command:
To activate on windows:

Scripts\activate

On Linux based on OS:

. tutorial-env/bin/activate

Install pip packages from requirements.txt

pip install -r requirements.txt

Run

On Windows run:

python classification.py

On Linux run:

python3 classification.py

Data Set

The dataset is downloaded from Iris Data Set.

Model

Tensorflow

This classification model is build using deep neural network (DNN) layers. Architecture of DNN is as follows:

• Input layer also known as Layer 0. Size: N * 4

• First deep learning layer, Size: 4 * 20

• Second deep learning layer, Size: 20 * 20

• Output layer, Size: 20 * 3 (Output is 3 because our classification problem has three classes)

Scikit-learn

Coming soon. Please wait!!

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

MIT