sudoku_data.py

# -*- coding: utf-8 -*-
"""sudoku_data.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1Nm6VofL0tUgUUvymSsMmvZgB-vfMXtwJ
"""

import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split

def normalize(game):
  return game/9-0.5
def denormalize(game):
  return (game+0.5)*9

def getSudokuData(file):
  dataset = pd.read_csv(file, compression='zip', header=0, sep=',', quotechar='"').to_numpy()
  features_dataset = dataset[0:,0]
  targets_dataset = dataset[0:,1]
  features = []
  targets = []
  for i in range(len(dataset)):
    features.append([int(j) for j in dataset[i][0]])
    targets.append( [int(j) for j in dataset[i][1]])
  features = normalize(np.array(features).reshape(len(dataset),9,9,1))
  targets = np.array(targets).reshape(len(dataset),81,1)-1
  X_train, X_test, y_train, y_test = train_test_split(features, targets, test_size=0.2)
  return (X_train,y_train), (X_test,y_test)