Tarun Kumar tarunk04

## digit-recognition-tutorial-cnn-99-67-accuracy.ipynb

      
              1 file
            
          
              0 forks
            
          
                0 comments
              
            
              0 stars
            
          
                tarunk04
                / digit-recognition-tutorial-cnn-99-67-accuracy.ipynb
            
            
              Last active
              April 27, 2020 21:15
            
              
                MNIST data set tutorial. In this notebook, I have covered the necessary steps to approach any Machine Learning Classification Problem. Included Image Visualization for better understanding. Quick Links to the functions I have used to explore it in depth. Basic techniques such as Confusion Matrix, Image Augmentation, etc.
              
          
      Loading

      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## validate.py
    def validate_cell(self, sudoku_solved, val, i, j):
        # getting index for block ( 9 blocks of 3x3 size)
        block_i = i // 3
        block_j = j // 3

        # check block
        for b_i in range(block_i * 3, block_i * 3 + 3):
            for b_j in range(block_j * 3, block_j * 3 + 3):
                if sudoku_solved[b_i][b_j] == val:
                    return False

## solve_sudoku.py
    def __solve_backtrack__(self, i, j, visualize=False):
        # Changing row if col index reaches end of the row
        if j > 8:
            j = 0
            i += 1
            # end condition
            if i > 8:
                if visualize == False:
                    eel.draw_sudoku(self.sudoku_solved)
                return True

## generate_solved_sudoku.py
    def __generator__(self, grid, i, j):
        number = [1, 2, 3, 4, 5, 6, 7, 8, 9]
        if j > 8:
            j = 0
            i += 1
            # end condition
            if i > 8:
                return True

        # skipping cell if the cell has already a number

## generate_solved_sudoku.py
    def __generator__(self, grid, i, j):
        number = [1, 2, 3, 4, 5, 6, 7, 8, 9]
        if j > 8:
            j = 0
            i += 1
            # end condition
            if i > 8:
                return True

        # skipping cell if the cell has already a number

## generating_puzzle.py
def generate_sudoku(self, level=2):
        # generate solved sudoku
        self.generate_solved_sudoku()
        if level == 1:
            remove_cell = random.randint(32, 38)
        if level == 2:
            remove_cell = random.randint(40, 46)
        if level == 3:
            remove_cell = random.randint(48, 52)
        if level == 4:

## script.js

// JavaScript functions exposed to python
eel.expose(draw_sudoku)
eel.expose(update_sudoku)
// Function for drawing sudoku
function draw_sudoku(data) {
    for (i = 0; i < 81; i++) {
        // row and col values for corresponding i values
        row = ~~(i / 9);
        col = i % 9;

## App.py
# python functions exposed to javascript
@eel.expose
def generate_new_sudoku(level):
    global current_sudoku_onboard
    # generating sudoku for required level
    current_sudoku_onboard = Sudoku().generate_sudoku(int(level))
    # drawing sudoku by calling javascript function from python
    eel.draw_sudoku(current_sudoku_onboard)


## example1-code.py
#Importing the progress bar library
from progress import Progress as P
import time

images = 2478 # number of images to process
# Create a progress bar instance
p = P(images,mode = 'no-bar')

# Creating one element for image counter
counter = P.Element("Images Processed", 0)

## example2-code.py
#Importing the progress bar library
from progress import Progress as P
import time
import random


p = P(70, mode='bar')  # progress bar object

# defining progress bar elements
epoch = P.Element("Epoch", 0)
	def validate_cell(self, sudoku_solved, val, i, j):
	# getting index for block ( 9 blocks of 3x3 size)
	block_i = i // 3
	block_j = j // 3

	# check block
	for b_i in range(block_i * 3, block_i * 3 + 3):
	for b_j in range(block_j * 3, block_j * 3 + 3):
	if sudoku_solved[b_i][b_j] == val:
	return False
	def __solve_backtrack__(self, i, j, visualize=False):
	# Changing row if col index reaches end of the row
	if j > 8:
	j = 0
	i += 1
	# end condition
	if i > 8:
	if visualize == False:
	eel.draw_sudoku(self.sudoku_solved)
	return True
	def __generator__(self, grid, i, j):
	number = [1, 2, 3, 4, 5, 6, 7, 8, 9]
	if j > 8:
	j = 0
	i += 1
	# end condition
	if i > 8:
	return True

	# skipping cell if the cell has already a number
	def generate_sudoku(self, level=2):
	# generate solved sudoku
	self.generate_solved_sudoku()
	if level == 1:
	remove_cell = random.randint(32, 38)
	if level == 2:
	remove_cell = random.randint(40, 46)
	if level == 3:
	remove_cell = random.randint(48, 52)
	if level == 4:

	// JavaScript functions exposed to python
	eel.expose(draw_sudoku)
	eel.expose(update_sudoku)
	// Function for drawing sudoku
	function draw_sudoku(data) {
	for (i = 0; i < 81; i++) {
	// row and col values for corresponding i values
	row = ~~(i / 9);
	col = i % 9;
	# python functions exposed to javascript
	@eel.expose
	def generate_new_sudoku(level):
	global current_sudoku_onboard
	# generating sudoku for required level
	current_sudoku_onboard = Sudoku().generate_sudoku(int(level))
	# drawing sudoku by calling javascript function from python
	eel.draw_sudoku(current_sudoku_onboard)
	#Importing the progress bar library
	from progress import Progress as P
	import time

	images = 2478 # number of images to process
	# Create a progress bar instance
	p = P(images,mode = 'no-bar')

	# Creating one element for image counter
	counter = P.Element("Images Processed", 0)