diff --git a/Preprocessing/expand.py b/Preprocessing/expand.py
new file mode 100644
index 0000000..e059ec8
--- /dev/null
+++ b/Preprocessing/expand.py
@@ -0,0 +1,51 @@
+import cv2
+import numpy as np
+import os
+
+
+def show_image(img):
+    cv2.imshow("preview", img)
+    cv2.waitKey(0)
+    cv2.destroyAllWindows()
+
+
+def rotate_symbol(img, angle):
+    _, img_binary = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY_INV)
+    h, w = img.shape[:2]
+
+    center = (w / 2, h / 2)
+
+    rot_matrix = cv2.getRotationMatrix2D(center, angle, 1.0)
+    img_rotated = cv2.warpAffine(img_binary, rot_matrix, (w, h))
+
+    img_final = cv2.bitwise_not(img_rotated)
+
+    return img_final
+    # show_image(img_final)
+
+
+def expand_dataset(folder):
+    images = os.listdir(folder)
+    num_augmented = 6000 - len(images)
+    print(num_augmented)
+    i = 0
+
+    for filename in images:
+        img = cv2.imread(os.path.join(folder, filename), cv2.IMREAD_GRAYSCALE)
+
+        img_rotate_10 = rotate_symbol(img, 10)
+        # img_flip_lr = cv2.flip(img, 1)
+        img_rotate_180 = cv2.rotate(img, cv2.ROTATE_180)
+
+        # print(folder + '/augmented_' + str(i) + '_flip_lr.jpg')
+        # cv2.imwrite(folder + '/augmented_' + str(i) + '_flip_lr.jpg', img_flip_lr)
+        cv2.imwrite(folder + '/augmented_' + str(i) + '_rotate_180.jpg', img_rotate_180)
+        cv2.imwrite(folder + '/augmented_' + str(i) + '_rotate_10.jpg', img_rotate_10)
+        i += 2
+
+        if i > num_augmented+2:
+            break
+
+
+if __name__ == '__main__':
+    expand_dataset("../Data/int")
diff --git a/Preprocessing/solver_data.py b/Preprocessing/solver_data.py
new file mode 100644
index 0000000..c036447
--- /dev/null
+++ b/Preprocessing/solver_data.py
@@ -0,0 +1,76 @@
+from keras.datasets import mnist
+import cv2
+import numpy as np
+import pandas as pd
+import os
+
+
+def show_image(img):
+    cv2.imshow("preview", img)
+    cv2.waitKey(0)
+    cv2.destroyAllWindows()
+
+
+def load_symbol_dataset(folder):
+    data = list()
+    for filename in os.listdir(folder):
+        img = cv2.imread(os.path.join(folder, filename), cv2.IMREAD_GRAYSCALE)
+        img = ~img
+        if img is not None:
+            _, thresh = cv2.threshold(img, 127, 255, cv2.THRESH_BINARY)
+            # show_image(thresh)
+
+            img_resize = cv2.resize(thresh, (28, 28))
+            # show_image(img_resize)
+
+            _, img_resize_thresh = cv2.threshold(img_resize, 30, 255, cv2.THRESH_BINARY)
+            # show_image(img_resize_thresh)
+            #
+            # kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (2, 2))
+            # img_dilated = cv2.dilate(img_resize_thresh, kernel, iterations=1)
+            # show_image(img_dilated)
+
+            img_resize_thresh = np.reshape(img_resize_thresh, (784, ))
+
+            data.append(img_resize_thresh)
+    return data
+
+# '+' - 10
+data_plus = load_symbol_dataset('../Data/+')
+data_plus = np.insert(data_plus, 784, 10, axis=1)
+print(data_plus.shape)
+
+# '-' - 11
+data_minus = load_symbol_dataset('../Data/-')
+data_minus = np.insert(data_minus, 784, 11, axis=1)
+print(data_minus.shape)
+
+# '*' - 12
+data_times = load_symbol_dataset('../Data/times')
+data_times = np.insert(data_times, 784, 12, axis=1)
+print(data_times.shape)
+
+# '=' - 13
+data_equal = load_symbol_dataset('../Data/=')
+data_equal = np.insert(data_equal, 784, 13, axis=1)
+print(data_equal.shape)
+
+# 'x' = 14
+data_x = load_symbol_dataset('../Data/X')
+data_x = np.insert(data_x, 784, 14, axis=1)
+print(data_x.shape)
+
+# 'y' = 15
+data_y = load_symbol_dataset('../Data/Y')
+data_y = np.insert(data_y, 784, 15, axis=1)
+print(data_y.shape)
+
+train_data = np.concatenate((data_plus, data_minus, data_times, data_equal, data_x, data_y))
+
+for num in range(10):
+    data_num = load_symbol_dataset(f'../Data/{num}')
+    data_num = np.insert(data_num, 784, num, axis=1)
+    train_data = np.concatenate((train_data, data_num))
+
+df = pd.DataFrame(train_data, index=None)
+df.to_csv('solver_train_data.csv', index=False)