A post describing the whole process is available on Medium

We build a supervised classification model with multiple layer perceptron on iris flowers dataset. This model must predict which flowers are likely to belong to one of the three specific flower species with 95% or greater accuracy. Besides the feedforward neural network on scikit, we use the same network with more hidden layers on keras, a decision tree and a logistic regression to make a benchmarking on different model prediction accuracy.

Iris is a genus of flowering plants species. It takes its name from the goddess of rainbow in Greek mythology.

It has got three classes where one class is linearly separable from the other 2; the latter are not linearly separable from each other. Each class refers to a type of iris plant and contains 50 instances.

• Data Set Characteristics: Multivariate
• Number of Instances:150
• Number of Features: 4
• Number of Classes: 3
• Features Characteristics: Real
• Classes Characteristics: Nominal
• Area: Life
• Associated Tasks: Classification
• Missing Values? No

To run the project, it is required that the following are installed in your system:

• anaconda
• Pyhton version: "^2.7"
• NumPy version: "^1.11.3"
• Scikit learn version: "^0.19.1"
• Seaborn version: "^0.8.1"
• Tensorflow version: "^1.4"
• Keras version: "^2.0.7"