rpongweb / mlfruit

Machine Learning using open source python libraries. (TensorFlow & Scikit Learn)

MLFruit

Machine Learning using open source python libraries. (TensorFlow & Scikit Learn)

Supervised Learning : Create a classifier by finding patterns in examples.

To code this , we'll need to work with scikit learn!

Download and install the library here : http://scikit-learn.org/stable/install.html

Once it's installed , you can test it to see if you have it installed successfully by importing sklearn. (import sklearn)

Training Data consists of - texture , weight, and label of the various fruits. Write down your training data in code. The more training data you have , the better classifier you can create!

Our two variables will be features and labels. Features contains the texture and weight while labels contain the name of the fruit.

e.g - Apple , Oranges , and Grapes would be labels. e.g - smooth , bumpy, and hard would be Features

Features are like the input to the classifer and labels as the output.

e.g if we enter the weight and texture of a spcific fruit , we should expect the output to be a resulting fruit.

We'll use 0 for bumpy and 1 for smooth. Do the same for your labels. By using numbers to represent our data (fruits) it can boost the simplicity of the program.

We'll use a decision tree to get appropriate results.

Enter "import tree" to do so. clf means classifier. In scikit the training algorithim is included in the classifier object and it's called FIT. (Find patterns in Data)

At this point , we have a trained classifier so you can modify the program as you wish for other variables.

Have Fun!