Skip to content

artem-8178/NN_from_let_rec

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NN_from_let_rec

Languages and Tools:

opencv python pytorch

20 epochs were passed for a dataset of 370K letters.

loss 0.1561

Management

left-click = draw

right-click = clean up

scroll-wheel +/- = +/- brush size

Structure

Net(
  (c1): Linear(in_features=784, out_features=800, bias=True)
  (c2): Linear(in_features=800, out_features=500, bias=True)
  (c3): Linear(in_features=500, out_features=500, bias=True)
  (c4): Linear(in_features=500, out_features=400, bias=True)
  (c5): Linear(in_features=400, out_features=26, bias=True)
)

ReLU activation function

Installation dependencies and run

Clone the project

  git clone https://github.com/artem-8178/NN_from_let_rec

Go to the project directory

  cd NN_from_let_rec

Install dependencies

  pip install -r requirements.txt

Run

  python main.py

About

Neural network for letter recognition.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages