Skip to content
/ OCR-lite Public

Flask app for digit recognition, using CNN and Logistic Regression.

Notifications You must be signed in to change notification settings

wxo15/OCR-lite

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

41 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

OCR-lite

A simple Python Flask app to allow users to write a digit on a canvas, and a model will analyze and output the predicted number. Amazon S3 is used for storage of models and parameters.

Web app taken down due to Heroku ending their free plan.

Models available:

  1. Convoluted Neural Network (CNN) built using TensorFlow's Keras Layers, made up of:
    1. Conv2D(filters=16, kernel_size=(5, 5), padding='same', activation='relu', input_shape=(28, 28, 1))
    2. MaxPooling2D(pool_size=(2, 2))
    3. Conv2D(filters=36, kernel_size=(5, 5), padding='same', activation='relu')
    4. MaxPooling2D(pool_size=(2, 2)),
    5. Dropout(0.25)
    6. Flatten()
    7. Dense(128, activation='relu')
    8. Dropout(0.5)
    9. Dense(10, activation='softmax')
  2. sklearn's Logistic Regression

screenshots

About

Flask app for digit recognition, using CNN and Logistic Regression.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published