diff --git a/Autoencoder.py b/Autoencoder.py
new file mode 100644
index 0000000..1f29067
--- /dev/null
+++ b/Autoencoder.py
@@ -0,0 +1,46 @@
+import tensorflow as tf
+import numpy as np
+
+from Encoder import * 
+from Decoder import *
+
+class Autoencoder(tf.keras.Model):
+    def __init__(self):
+        super(Autoencoder, self).__init__()
+
+        self.encoder = Encoder()
+        self.decoder = Decoder()
+
+        self.loss_function = tf.keras.losses.MeanSquaredError()
+
+        self.optimizer = tf.keras.optimizers.Adam(learning_rate=0.01)
+
+
+    @tf.function
+    def call(self, x, training=False):
+        embedding = self.encoder(x, training)
+        decoded = self.decoder(embedding, training)
+        return decoded
+
+    @tf.function
+    def train_step(self, input, target):
+
+        with tf.GradientTape() as tape:
+            prediction = self(input, training=True)
+            loss = self.loss_function(target, prediction)
+
+        gradients = tape.gradient(loss, self.trainable_variables)
+        self.optimizer.apply_gradients(zip(gradients, self.trainable_variables))
+        return loss
+
+    def test(self, test_data):
+        # test over complete test data
+        test_loss_aggregator = []
+        for input, target in test_data: # ignore label            
+            prediction = self(input)
+            
+            sample_test_loss = self.loss_function(target, prediction)
+            test_loss_aggregator.append(sample_test_loss.numpy())
+
+        test_loss = tf.reduce_mean(test_loss_aggregator)
+        return test_loss
\ No newline at end of file
diff --git a/Decoder.py b/Decoder.py
new file mode 100644
index 0000000..3d9ffa2
--- /dev/null
+++ b/Decoder.py
@@ -0,0 +1,42 @@
+import tensorflow as tf
+
+class Decoder(tf.keras.Model): # <-- Needed to make parameters trainable and to be callable
+    def __init__(self):
+
+        super(Decoder, self).__init__()
+        self.layer_list = [
+            
+            tf.keras.layers.Conv2DTranspose(105, kernel_size=(3,3), strides=2, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+
+            tf.keras.layers.Conv2DTranspose(90, kernel_size=(3,3), strides=2, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+
+            tf.keras.layers.Conv2DTranspose(75, kernel_size=(3,3), strides=2, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+
+            # bottleneck to RGB
+
+            tf.keras.layers.Conv2DTranspose(2, kernel_size=(1,1), strides=1, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+            
+        ]
+    
+    @tf.function
+    def call(self, x, training):
+        
+        #print("decoder:")
+        for layer in self.layer_list:
+            #print(x.shape)
+            if isinstance(layer, tf.keras.layers.BatchNormalization):
+                x = layer(x,training)
+            else: 
+                x = layer(x)
+        # print(x.shape) 
+        # print("-------------")
+        # exit()
+        return x
\ No newline at end of file
diff --git a/Encoder.py b/Encoder.py
new file mode 100644
index 0000000..fc75cc1
--- /dev/null
+++ b/Encoder.py
@@ -0,0 +1,45 @@
+import tensorflow as tf
+
+class Encoder(tf.keras.Model): # <-- Needed to make parameters trainable and to be callable
+    def __init__(self):
+        super(Encoder, self).__init__()
+        self.layer_list = [
+            # input (243,243) 
+
+            tf.keras.layers.Conv2D(75, kernel_size=(3, 3), strides=2, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+            # -> (81, 81, 32)   
+
+            tf.keras.layers.Conv2D(90, kernel_size=(3, 3), strides=2, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+            # -> (27, 27, 64)
+
+            tf.keras.layers.Conv2D(105, kernel_size=(3, 3), strides=2, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+
+
+            # bottleneck
+            tf.keras.layers.Conv2D(3, kernel_size=(1, 1), strides=1, padding='same'),
+            tf.keras.layers.BatchNormalization(),
+            tf.keras.layers.Activation(tf.nn.tanh),
+            
+        ]
+    
+    @tf.function
+    def call(self, x, training):
+        #print("encoder:")
+        for layer in self.layer_list: 
+            #print(x.shape)
+            if isinstance(layer, tf.keras.layers.BatchNormalization):
+                x = layer(x,training)
+            else: 
+                x = layer(x)
+            
+        #print(x.shape)
+        #print("-------------")
+        #exit()
+        return x
+    
diff --git a/helper.py b/helper.py
new file mode 100644
index 0000000..bc71e36
--- /dev/null
+++ b/helper.py
@@ -0,0 +1,281 @@
+import tensorflow as tf
+import cv2 as cv
+import matplotlib.pyplot as plt
+
+# for the color conversion see https://github.com/tensorflow/io/blob/v0.24.0/tensorflow_io/python/experimental/color_ops.py#L398-L459
+# code from tensorflow io we're used, because tfio has no pip package for the jetson nano
+def rgb_to_lab(input, illuminant="D65", observer="2", name=None):
+    """
+    Convert a RGB image to CIE LAB.
+    Args:
+    input: A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    illuminant : {"A", "D50", "D55", "D65", "D75", "E"}, optional
+        The name of the illuminant (the function is NOT case sensitive).
+    observer : {"2", "10"}, optional
+        The aperture angle of the observer.
+    name: A name for the operation (optional).
+    Returns:
+    A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    """
+    input = tf.convert_to_tensor(input)
+    assert input.dtype in (tf.float16, tf.float32, tf.float64)
+
+    illuminants = {
+        "A": {
+            "2": (1.098466069456375, 1, 0.3558228003436005),
+            "10": (1.111420406956693, 1, 0.3519978321919493),
+        },
+        "D50": {
+            "2": (0.9642119944211994, 1, 0.8251882845188288),
+            "10": (0.9672062750333777, 1, 0.8142801513128616),
+        },
+        "D55": {
+            "2": (0.956797052643698, 1, 0.9214805860173273),
+            "10": (0.9579665682254781, 1, 0.9092525159847462),
+        },
+        "D65": {
+            "2": (0.95047, 1.0, 1.08883),
+            "10": (0.94809667673716, 1, 1.0730513595166162),
+        },
+        "D75": {
+            "2": (0.9497220898840717, 1, 1.226393520724154),
+            "10": (0.9441713925645873, 1, 1.2064272211720228),
+        },
+        "E": {"2": (1.0, 1.0, 1.0), "10": (1.0, 1.0, 1.0)},
+    }
+    coords = tf.constant(illuminants[illuminant.upper()][observer], input.dtype)
+
+    xyz = rgb_to_xyz(input)
+
+    xyz = xyz / coords
+
+    xyz = tf.where(
+        tf.math.greater(xyz, 0.008856),
+        tf.math.pow(xyz, 1.0 / 3.0),
+        xyz * 7.787 + 16.0 / 116.0,
+    )
+
+    xyz = tf.unstack(xyz, axis=-1)
+    x, y, z = xyz[0], xyz[1], xyz[2]
+
+    # Vector scaling
+    l = (y * 116.0) - 16.0
+    a = (x - y) * 500.0
+    b = (y - z) * 200.0
+
+    return tf.stack([l, a, b], axis=-1)
+
+
+def lab_to_rgb(input, illuminant="D65", observer="2", name=None):
+    """
+    Convert a CIE LAB image to RGB.
+    Args:
+    input: A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    illuminant : {"A", "D50", "D55", "D65", "D75", "E"}, optional
+        The name of the illuminant (the function is NOT case sensitive).
+    observer : {"2", "10"}, optional
+        The aperture angle of the observer.
+    name: A name for the operation (optional).
+    Returns:
+    A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    """
+    input = tf.convert_to_tensor(input)
+    assert input.dtype in (tf.float16, tf.float32, tf.float64)
+
+    lab = input
+    lab = tf.unstack(lab, axis=-1)
+    l, a, b = lab[0], lab[1], lab[2]
+
+    y = (l + 16.0) / 116.0
+    x = (a / 500.0) + y
+    z = y - (b / 200.0)
+
+    z = tf.math.maximum(z, 0)
+
+    xyz = tf.stack([x, y, z], axis=-1)
+
+    xyz = tf.where(
+        tf.math.greater(xyz, 0.2068966),
+        tf.math.pow(xyz, 3.0),
+        (xyz - 16.0 / 116.0) / 7.787,
+    )
+
+    illuminants = {
+        "A": {
+            "2": (1.098466069456375, 1, 0.3558228003436005),
+            "10": (1.111420406956693, 1, 0.3519978321919493),
+        },
+        "D50": {
+            "2": (0.9642119944211994, 1, 0.8251882845188288),
+            "10": (0.9672062750333777, 1, 0.8142801513128616),
+        },
+        "D55": {
+            "2": (0.956797052643698, 1, 0.9214805860173273),
+            "10": (0.9579665682254781, 1, 0.9092525159847462),
+        },
+        "D65": {
+            "2": (0.95047, 1.0, 1.08883),
+            "10": (0.94809667673716, 1, 1.0730513595166162),
+        },
+        "D75": {
+            "2": (0.9497220898840717, 1, 1.226393520724154),
+            "10": (0.9441713925645873, 1, 1.2064272211720228),
+        },
+        "E": {"2": (1.0, 1.0, 1.0), "10": (1.0, 1.0, 1.0)},
+    }
+    coords = tf.constant(illuminants[illuminant.upper()][observer], input.dtype)
+
+    xyz = xyz * coords
+
+    return xyz_to_rgb(xyz)
+
+
+def rgb_to_xyz(input, name=None):
+    """
+    Convert a RGB image to CIE XYZ.
+    Args:
+    input: A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    name: A name for the operation (optional).
+    Returns:
+    A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    """
+    input = tf.convert_to_tensor(input)
+    assert input.dtype in (tf.float16, tf.float32, tf.float64)
+
+    kernel = tf.constant(
+        [
+            [0.412453, 0.357580, 0.180423],
+            [0.212671, 0.715160, 0.072169],
+            [0.019334, 0.119193, 0.950227],
+        ],
+        input.dtype,
+    )
+    value = tf.where(
+        tf.math.greater(input, 0.04045),
+        tf.math.pow((input + 0.055) / 1.055, 2.4),
+        input / 12.92,
+    )
+    return tf.tensordot(value, tf.transpose(kernel), axes=((-1,), (0,)))
+
+
+def xyz_to_rgb(input, name=None):
+    """
+    Convert a CIE XYZ image to RGB.
+    Args:
+    input: A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    name: A name for the operation (optional).
+    Returns:
+    A 3-D (`[H, W, 3]`) or 4-D (`[N, H, W, 3]`) Tensor.
+    """
+    input = tf.convert_to_tensor(input)
+    assert input.dtype in (tf.float16, tf.float32, tf.float64)
+
+    # inv of:
+    # [[0.412453, 0.35758 , 0.180423],
+    #  [0.212671, 0.71516 , 0.072169],
+    #  [0.019334, 0.119193, 0.950227]]
+    kernel = tf.constant(
+        [
+            [3.24048134, -1.53715152, -0.49853633],
+            [-0.96925495, 1.87599, 0.04155593],
+            [0.05564664, -0.20404134, 1.05731107],
+        ],
+        input.dtype,
+    )
+    value = tf.tensordot(input, tf.transpose(kernel), axes=((-1,), (0,)))
+    value = tf.where(
+        tf.math.greater(value, 0.0031308),
+        tf.math.pow(value, 1.0 / 2.4) * 1.055 - 0.055,
+        value * 12.92,
+    )
+    return tf.clip_by_value(value, 0, 1)
+
+
+# see: https://github.com/JetsonHacksNano/CSI-Camera/blob/master/simple_camera.py
+def gstreamer_pipeline(
+    capture_width,
+    capture_height,
+    display_width,
+    display_height,
+    framerate=10,
+    flip_method=0,
+):
+    return (
+        "nvarguscamerasrc ! "
+        "video/x-raw(memory:NVMM), "
+        "width=(int)%d, height=(int)%d, "
+        "format=(string)NV12, framerate=(fraction)%d/1 ! "
+        "nvvidconv flip-method=%d ! "
+        "video/x-raw, width=(int)%d, height=(int)%d, format=(string)BGRx ! "
+        "videoconvert ! "
+        "video/x-raw, format=(string)BGR ! appsink"
+        % (
+            capture_width,
+            capture_height,
+            framerate,
+            flip_method,
+            display_width,
+            display_height,
+        )
+    )
+
+def get_RGB(L, AB):
+    # Remove normalization
+    L = (L + 1)*50
+    AB = ((AB - 1)*255/2)+128
+
+    LAB = tf.concat([L, AB], 3)
+    #rgb = tfio.experimental.color.lab_to_rgb(LAB)
+    rgb = lab_to_rgb(LAB)
+
+    return rgb
+
+
+def to_LAB(frame):
+    # recast from int to float
+    gray = tf.cast(frame, tf.float32)
+    # normalize
+    gray = gray/255
+
+    gray = rgb_to_lab(gray)
+    # "split" the array
+    L, AB = gray[:,:,0], tf.stack([gray[:,:,1], gray[:,:,2]], axis=2)
+    L = tf.reshape(L, shape=(256,256,1))
+    # normalize
+    L, AB = ((L/50.0) - 1., 1 + (2*(AB - 128)/255))
+    return L, AB
+
+def suitable_rgb(L, ab_img):
+    # CIE LAB color space to RGB
+    recolored_image = get_RGB(L, ab_img)
+    # remove batch dimension
+    recolored_image = tf.squeeze(recolored_image, 0)
+    # remove normalization and interpret as uint8
+    recolored_image = tf.cast(recolored_image*255, tf.uint8)
+    # make it a numpy array
+    recolored_image = recolored_image.numpy()
+    # convert to BGR
+    recolored_image = cv.cvtColor(recolored_image, cv.COLOR_RGB2BGR)
+    return recolored_image
+    
+    
+def create_Plot(loss_array):
+    # set plot configuration
+    plt.grid(True)
+    plt.ylim(0, 0.07)
+    plt.xlabel('Each Fram')
+    plt.ylabel('Loss')
+    
+    plt.plot(loss_array)
+    plt.savefig('loss.png')
+    plt.clf() 
+    
+
+def saturate(image):
+    hsvImg = cv.cvtColor(image,cv.COLOR_BGR2HSV)
+    #multiple by a factor to change the saturation
+    hsvImg[...,1] = hsvImg[...,1]*1.4
+    #multiple by a factor of less than 1 to reduce the brightness 
+    hsvImg[...,2] = hsvImg[...,2]*0.6
+    image=cv.cvtColor(hsvImg,cv.COLOR_HSV2BGR)
+    return image
\ No newline at end of file
diff --git a/live_recolor.py b/live_recolor.py
new file mode 100644
index 0000000..1cf7fe3
--- /dev/null
+++ b/live_recolor.py
@@ -0,0 +1,102 @@
+from calendar import c
+from dis import dis
+import cv2 as cv
+from Autoencoder import Autoencoder
+import helper
+import tensorflow as tf
+
+# Height and width of each frame
+HEIGHT = 256
+WIDTH = 256
+
+# Text on the pictures
+font = cv.FONT_HERSHEY_SIMPLEX
+position = (0,HEIGHT-10)
+fontScale = 1
+fontColor = (255,0,255)
+
+
+def load_model(weight_path):
+    autoencoder = Autoencoder()
+    # need a batch size
+    autoencoder.build((1, 256, 256, 1))
+    # load model weights
+    autoencoder.load_weights(weight_path)
+    return autoencoder
+
+
+def addTextToFrame(frame, text):
+    return (
+        cv.putText(frame,text,
+        position,
+        font,
+        fontScale,
+        fontColor)
+    )
+
+
+def main(pipeline, autoencoder):
+    # capture camera feed
+    if pipeline:
+        cap = cv.VideoCapture(pipeline, cv.CAP_GSTREAMER)
+    else:
+        cap = cv.VideoCapture(0)
+    # loop
+    while True:
+        # get frame
+        ret, frame = cap.read()
+        frame = cv.resize(frame, (256,256))
+
+        # make grayscale image
+        gray = cv.cvtColor(frame, cv.COLOR_BGR2GRAY)
+        # add back the channel, so concatination works
+        three_value_gray = cv.cvtColor(gray, cv.COLOR_GRAY2BGR)
+        
+        L, AB = helper.to_LAB(frame)
+        
+        # add batch dimension
+        L = tf.expand_dims(L, 0)
+        # predict the colored image from gray
+        ab_img = autoencoder(L)
+        
+        recolored_image = helper.suitable_rgb(L, ab_img)
+        
+        
+        # add text to the frame
+        ogFrame =  addTextToFrame(frame, 'Original')
+        three_value_gray =  addTextToFrame(three_value_gray, 'Grayscale')
+        recolored_image =  addTextToFrame(recolored_image, 'Recolored')
+
+        # grab fps counter at to the last frame
+        fps = cap.get(cv.CAP_PROP_FPS)
+        cv.putText(recolored_image, "FPS: {:.2f}".format(fps), (150, 20), font,0.5,fontColor)
+
+        # connect the three frames into one
+        im_h = cv.hconcat([ogFrame, three_value_gray, recolored_image])
+        # display it
+        cv.imshow("Live Recoloration", im_h)
+        # quit on ESC
+        if cv.waitKey(1) == 27 :
+            break
+    # release camera
+    cap.release()
+    cv.destroyAllWindows()
+
+
+if __name__ == '__main__':
+
+    # create the gstreamer pipeline for picking the camera's data
+    pipeline = helper.gstreamer_pipeline(
+        capture_width=WIDTH,
+        capture_height=HEIGHT,
+        display_width=WIDTH,
+        display_height=HEIGHT,
+        flip_method=0
+    )
+
+    autoencoder = load_model(weight_path="./saved_models/trainied_weights_epoch_12")
+    try:
+        main(pipeline, autoencoder)
+    except KeyboardInterrupt:
+        print("[!] Exiting program . . .")
+        exit(1)
\ No newline at end of file
diff --git a/live_recolor_plot.py b/live_recolor_plot.py
new file mode 100644
index 0000000..b364032
--- /dev/null
+++ b/live_recolor_plot.py
@@ -0,0 +1,122 @@
+import cv2 as cv
+from Autoencoder import Autoencoder
+import helper
+import tensorflow as tf
+import numpy as np
+import os
+
+# Height and width of each frame
+HEIGHT = 256
+WIDTH = 256
+
+# Text on the pictures
+font = cv.FONT_HERSHEY_SIMPLEX
+position = (0,HEIGHT-10)
+fontScale = 1
+fontColor = (255,0,255)
+
+
+def load_model(weight_path):
+    autoencoder = Autoencoder()
+    # need a batch size
+    autoencoder.build((1, 256, 256, 1))
+    # load model weights
+    autoencoder.load_weights(weight_path)
+    return autoencoder
+
+
+def addTextToFrame(frame, text):
+    return (
+        cv.putText(frame,text,
+        position,
+        font,
+        fontScale,
+        fontColor)
+    )
+
+
+def main(pipeline, autoencoder):
+    # capture camera feed
+    if pipeline:
+        cap = cv.VideoCapture(pipeline, cv.CAP_GSTREAMER)
+    else:
+        cap = cv.VideoCapture(0)
+    # array with 30 values for loss display
+    loss_array = np.zeros(31)    
+    # loop
+    frame = 1
+    while True:
+        # get frame
+        ret, frame = cap.read()
+        if frame is None:
+            break
+        frame = cv.resize(frame, (256,256))
+
+        # make grayscale image
+        gray = cv.cvtColor(frame, cv.COLOR_BGR2GRAY)
+        # add back the channel, so concatination works
+        three_value_gray = cv.cvtColor(gray, cv.COLOR_GRAY2BGR)
+        
+        L, AB = helper.to_LAB(frame)
+        
+        # add batch dimension
+        L = tf.expand_dims(L, 0)
+        # predict the colored image from gray
+        ab_img = autoencoder(L)
+        
+        recolored_image = helper.suitable_rgb(L, ab_img)
+        
+        
+        # add text to the frame
+        ogFrame =  addTextToFrame(frame, 'Original')
+        three_value_gray =  addTextToFrame(three_value_gray, 'Grayscale')
+        recolored_image =  addTextToFrame(recolored_image, 'Recolored')
+
+        #get loss
+        loss = autoencoder.loss_function(AB, ab_img)
+        # get only the loss value
+        loss = loss.numpy()
+
+        # add loss to array
+        loss_array = loss_array[1:]
+        loss_array = np.append(loss_array, loss)
+        
+        helper.create_Plot(loss_array)
+
+        # grab fps counter at to the last frame
+        fps = cap.get(cv.CAP_PROP_FPS)
+        cv.putText(recolored_image, "FPS: {:.2f}".format(fps), (150, 20), font,0.5,fontColor)
+
+        # load loss graph
+        loss_graph = cv.imread('loss.png')
+        # make 256,256 image
+        loss_graph = cv.resize(loss_graph, (256,256))
+
+        # connect the three frames into one
+        im_h = cv.hconcat([ogFrame, three_value_gray, recolored_image, loss_graph])
+        # display it
+        cv.imshow("Live Recoloration", im_h)
+        # quit on ESC
+        if cv.waitKey(1) == 27 :
+            break
+    # delete loss graph
+    os.remove('loss.png')
+    # release camera
+    cap.release()
+    cv.destroyAllWindows()
+
+
+if __name__ == '__main__':
+
+    # create the gstreamer pipeline for picking the camera's data
+    pipeline = None
+
+    autoencoder = load_model(weight_path="./saved_models/trainied_weights_epoch_12")
+    try:
+        main(pipeline, autoencoder)
+    except KeyboardInterrupt:
+        print("[!] Exiting program . . .")
+        # delete loss graph
+        if os.path.exists('loss.png'):
+            os.remove('loss.png')
+        exit(1)
\ No newline at end of file
diff --git a/saved_models/.DS_Store b/saved_models/.DS_Store
new file mode 100644
index 0000000..7deef98
Binary files /dev/null and b/saved_models/.DS_Store differ
diff --git a/saved_models/checkpoint b/saved_models/checkpoint
new file mode 100644
index 0000000..3b3ae8b
--- /dev/null
+++ b/saved_models/checkpoint
@@ -0,0 +1,2 @@
+model_checkpoint_path: "trainied_weights_epoch_12"
+all_model_checkpoint_paths: "trainied_weights_epoch_12"
diff --git a/saved_models/trainied_weights_epoch_1.data-00000-of-00001 b/saved_models/trainied_weights_epoch_1.data-00000-of-00001
new file mode 100644
index 0000000..82f95cd
Binary files /dev/null and b/saved_models/trainied_weights_epoch_1.data-00000-of-00001 differ
diff --git a/saved_models/trainied_weights_epoch_1.index b/saved_models/trainied_weights_epoch_1.index
new file mode 100644
index 0000000..b541cdb
Binary files /dev/null and b/saved_models/trainied_weights_epoch_1.index differ
diff --git a/saved_models/trainied_weights_epoch_12.data-00000-of-00001 b/saved_models/trainied_weights_epoch_12.data-00000-of-00001
new file mode 100644
index 0000000..e7983f7
Binary files /dev/null and b/saved_models/trainied_weights_epoch_12.data-00000-of-00001 differ
diff --git a/saved_models/trainied_weights_epoch_12.index b/saved_models/trainied_weights_epoch_12.index
new file mode 100644
index 0000000..8aaed32
Binary files /dev/null and b/saved_models/trainied_weights_epoch_12.index differ
diff --git a/saved_models/trainied_weights_epoch_19.data-00000-of-00001 b/saved_models/trainied_weights_epoch_19.data-00000-of-00001
new file mode 100644
index 0000000..a25bc5d
Binary files /dev/null and b/saved_models/trainied_weights_epoch_19.data-00000-of-00001 differ
diff --git a/saved_models/trainied_weights_epoch_19.index b/saved_models/trainied_weights_epoch_19.index
new file mode 100644
index 0000000..a59234c
Binary files /dev/null and b/saved_models/trainied_weights_epoch_19.index differ
diff --git a/saved_models/trainied_weights_epoch_6.data-00000-of-00001 b/saved_models/trainied_weights_epoch_6.data-00000-of-00001
new file mode 100644
index 0000000..8f371ba
Binary files /dev/null and b/saved_models/trainied_weights_epoch_6.data-00000-of-00001 differ
diff --git a/saved_models/trainied_weights_epoch_6.index b/saved_models/trainied_weights_epoch_6.index
new file mode 100644
index 0000000..0fd28f6
Binary files /dev/null and b/saved_models/trainied_weights_epoch_6.index differ
diff --git a/short_presentation.mp4 b/short_presentation.mp4
new file mode 100644
index 0000000..2f50787
Binary files /dev/null and b/short_presentation.mp4 differ
diff --git a/short_presentation_without_text.mp4 b/short_presentation_without_text.mp4
new file mode 100644
index 0000000..d5f762b
Binary files /dev/null and b/short_presentation_without_text.mp4 differ
diff --git a/videoPresentation.gif b/videoPresentation.gif
new file mode 100644
index 0000000..755609d
Binary files /dev/null and b/videoPresentation.gif differ