Kannada-Digit-Recognizer-v1

Using a Convolutional Neural Network (CNN) to identify the Kannada numerical digits.
Tensorflow (Keras) is used to create, train and load the neural network model.
CustomTKinter/TKinter are used to provide the GUI and OpenCV is used to read input from the GUI.

New version

Requirements -

pip install -r requirements.txt

• Tensorflow (Keras)
• OpenCV
• Pillow (PIL)
• Pandas
• Numpy
• CustomTkinter

Instructions -

• Run either of the Jupyter notebooks - GUI_CustomTkinter.ipynb / GUI_Tkinter.ipynb

CustomTKinter GUI

TKinter GUI

Data - https://www.kaggle.com/competitions/Kannada-MNIST/data

Notebook used as reference - https://www.kaggle.com/code/shahules/indian-way-to-learn-cnn