Recognizing Handwritten Digits

Recognizing handwritten text is a problem that can be traced back to the first automatic 

machines that needed to recognize individual characters in handwritten documents. 

Think about, for example, the ZIP codes on letters at the post office and the automation 

needed to recognize these five digits. Perfect recognition of these codes is necessary in 

order to sort mail automatically and efficiently

libraries:    scikitlearn, numpy, pandas

language:   python(100%)

dataset:      kaggle, minist, sklearn, dataset 

dataset link:    https://www.kaggle.com/oddrationale/mnist-in-csv

    

            

The scikit-learn library provides numerous datasets that are useful for testing many 

problems of data analysis and prediction of the results. Also in this case there is a dataset 

of images called Digits

 

conclusion:

• The principles of supervised machine learning for classification,
• How to install and use the scientific python suite for machine learning,
• How to investigate about your input dataset,
• How to train a neural network for image recognition, reaching an accuracy larger than 90% for digit classification.

It can be seen in above picture  i have used 2 validation sets for and the accuracy score is quite different. we can use linear regression and decision tree also instead of SVM classifier.

In this way i have completed my internship project by using algorithm provided by scikit-learn 

library for machine learning.