From 78a50c9e9974501fca9dda57d988031fdd8144cf Mon Sep 17 00:00:00 2001
From: Mirko Birbaumer <mirko.birbaumer@hslu.ch>
Date: Wed, 28 Sep 2022 20:33:39 +0000
Subject: [PATCH] Keras code slightly adapted

---
 ... Notebook Block 2 - Neural Networks .ipynb | 311 +++++++++---------
 1 file changed, 154 insertions(+), 157 deletions(-)

diff --git a/notebooks/Neural_Networks/Jupyter Notebook Block 2 - Neural Networks .ipynb b/notebooks/Neural_Networks/Jupyter Notebook Block 2 - Neural Networks .ipynb
index 2de0fa8..87b673c 100644
--- a/notebooks/Neural_Networks/Jupyter Notebook Block 2 - Neural Networks .ipynb	
+++ b/notebooks/Neural_Networks/Jupyter Notebook Block 2 - Neural Networks .ipynb	
@@ -39,7 +39,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -111,7 +111,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Let us first train a Softmax classifier on this classification dataset. As we saw in the previous sections, the Softmax classifier has a linear score function and uses the cross-entropy loss. The parameters of the linear classifier consist of a weight matrix \n",
+    "Let us first train a Softmax Classifier on this classification dataset. As we saw in the previous sections, the Softmax classifier has a linear score function and uses the cross-entropy loss. The parameters of the linear classifier consist of a weight matrix \n",
     "$ W $ and a  bias vector $ b $ for each class. Let us first initialize these \n",
     "parameters to be random numbers:"
    ]
@@ -213,7 +213,7 @@
     "\n",
     "There are many ways to quantify this intuition, but in this \n",
     "example let us use the _cross-entropy loss_ that is associated with the Softmax \n",
-    "classifier. Recall that if $ z $ is the array of class scores for a single \n",
+    "Classifier. Recall that if $ z $ is the array of class scores for a single \n",
     "example (e.g. array of $ 3 $ numbers here), then the Softmax classifier \n",
     "computes the loss for that example as:\n",
     "\n",
@@ -223,7 +223,7 @@
     "$$\n",
     "\n",
     "where $z_{y_i}$ denotes the element of the score array that represents to the \n",
-    "correct class. We can see that the Softmax classifier interprets every element of $ z $ \n",
+    "correct class. We can see that the Softmax Classifier interprets every element of $ z $ \n",
     "as holding the (unnormalized) log probabilities of the three classes. \n",
     "We exponentiate these to get (unnormalized) probabilities, and then \n",
     "normalize them to get probabilites. Therefore, the expression inside \n",
@@ -278,7 +278,7 @@
    "metadata": {},
    "source": [
     "- Recall that `scores` has shape $[300 \\times 3] $\n",
-    "- `np.exp()` is a _unary_ universal function, or _ufunc_. These functions perform elementwise operations on data in ndarrays. That is,  `exp\\_scores` \n",
+    "- `np.exp()` is a _unary_ universal function, or _ufunc_. These functions perform elementwise operations on data in ndarrays. That is,  `exp_scores` \n",
     "has shape $[300 \\times 3] $\n",
     "- The function `np.sum(exp_scores, axis=1, keepdims=True)` computes the sum over \n",
     "axis $ 1 $. If `keepdims` is set to `True`, the axes which are reduced are \n",
@@ -1062,19 +1062,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 3,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
-      "11493376/11490434 [==============================] - 0s 0us/step\n",
-      "11501568/11490434 [==============================] - 0s 0us/step\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from tensorflow.keras.datasets import mnist\n",
     "(X_train, y_train), (X_test, y_test) = mnist.load_data()"
@@ -1091,7 +1081,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -1100,7 +1090,7 @@
        "(60000, 28, 28)"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1111,7 +1101,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -1120,7 +1110,7 @@
        "60000"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1131,7 +1121,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -1140,7 +1130,7 @@
        "array([5, 0, 4, ..., 5, 6, 8], dtype=uint8)"
       ]
      },
-     "execution_count": 22,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1153,12 +1143,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Let us display an the fourth digit:"
+    "Let us display the fifth digit:"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -1183,7 +1173,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -1192,7 +1182,7 @@
        "9"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1210,7 +1200,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -1219,7 +1209,7 @@
        "(10000, 28, 28)"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1230,7 +1220,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -1239,7 +1229,7 @@
        "10000"
       ]
      },
-     "execution_count": 26,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1250,7 +1240,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -1259,7 +1249,7 @@
        "array([7, 2, 1, ..., 4, 5, 6], dtype=uint8)"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1308,7 +1298,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -1329,7 +1319,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
@@ -1351,7 +1341,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "The workflow will be as follows: First, we’ll feed the neural network the training data, `X_train` and `y_train`. The network will then learn to associate images and labels. Finally, we’ll ask the network to produce predictions for test_images, and we’ll verify whether these predictions match the labels from `test_labels`."
+    "The workflow will be as follows: First, we’ll feed the neural network the training data, `X_train` and `y_train`. The network will then learn to associate images and labels. Finally, we’ll ask the network to produce predictions for `test_images`, and we’ll verify whether these predictions match the labels from `test_labels`."
    ]
   },
   {
@@ -1394,11 +1384,11 @@
     "\n",
     "\n",
     "\n",
-    "* **\"hidden\"** `layers.Dense`— A densely connected layer of 500 neurons. Each neuron (or node) takes input from all 784 nodes in the previous layer - by specifying an `input_shape` to the first layer in the Sequential model - , weighting that input according to hidden parameters which will be learned during training, and outputs a single value to the next layer.  \n",
+    "* **\"hidden\"** `layers.Dense`— A densely connected layer of 500 units. Each unit (or neuron) takes input from all 784 units in the previous layer - by specifying an `input_shape` to the first layer in the Sequential model - , weighting that input according to hidden parameters which will be learned during training, and outputs a single value to the next layer.  \n",
     "\n",
-    "* **\"hidden\"** `layers.Dense`— A densely connected layer of 50 neurons. Each neuron (or node) takes input from all 500 nodes in the previous layer, weighting that input according to hidden parameters which will be learned during training, and outputs a single value to the next layer.\n",
+    "* **\"hidden\"** `layers.Dense`— A densely connected layer of 50 units. Each unit (or neuron) takes input from all 500 units in the previous layer, weighting that input according to hidden parameters which will be learned during training, and outputs a single value to the next layer.\n",
     "\n",
-    "* **output** `layers.Dense` — A 10-node *softmax* layer, with each node representing a class of clothing. As in the previous layer, each node takes input from the 50 nodes in the layer before it. Each node weights the input according to learned parameters, and then outputs a value in the range `[0, 1]`, representing the probability that the image belongs to that class. The sum of all 10 node values is 1.\n"
+    "* **output** `layers.Dense` — A 10-unit *softmax* layer, with each unit representing a class of digit. As in the previous layer, each unit takes input from the 50 units in the layer before it. Each unit weights the input according to learned parameters, and then outputs a value in the range `[0, 1]`, representing the probability that the image belongs to that class. The sum of all 10 unit values is 1.\n"
    ]
   },
   {
@@ -1675,7 +1665,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 25,
    "metadata": {},
    "outputs": [
     {
@@ -1719,16 +1709,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [],
    "source": [
     "model = tf.keras.Sequential()\n",
     "# From Input to first hidden layer\n",
-    "model.add(tf.keras.layers.Dense(100, activation= tf.nn.relu, \n",
+    "model.add(tf.keras.layers.Dense(100, activation= \"relu\", \n",
     "                                input_shape=(2,)))\n",
     "# From first hidden layer to output layer\n",
-    "model.add(tf.keras.layers.Dense(3, activation=tf.nn.softmax))"
+    "model.add(tf.keras.layers.Dense(3, activation=\"softmax\"))"
    ]
   },
   {
@@ -1739,7 +1729,7 @@
     "\n",
     "\n",
     "Once we have our model built, we need to compile it before it can be run. Compiling the Keras \n",
-    "model calls the backend (tensorflow, theano, etc.) and binds the optimizer, loss function, \n",
+    "model calls the backend Tensorflow and binds the optimizer, loss function, \n",
     "and other parameters required before the model can be run on any input data. We'll specify the \n",
     "loss function to be `categorical_crossentropy`, \n",
     "and specify `adam` as the optimizer (which is a reasonable default when speed is a priority). And finally, \n",
@@ -1748,7 +1738,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
@@ -1757,7 +1747,7 @@
        "(300, 2)"
       ]
      },
-     "execution_count": 43,
+     "execution_count": 27,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1778,20 +1768,20 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Model: \"sequential_1\"\n",
+      "Model: \"sequential_3\"\n",
       "_________________________________________________________________\n",
       " Layer (type)                Output Shape              Param #   \n",
       "=================================================================\n",
-      " dense_3 (Dense)             (None, 100)               300       \n",
+      " dense_6 (Dense)             (None, 100)               300       \n",
       "                                                                 \n",
-      " dense_4 (Dense)             (None, 3)                 303       \n",
+      " dense_7 (Dense)             (None, 3)                 303       \n",
       "                                                                 \n",
       "=================================================================\n",
       "Total params: 603\n",
@@ -1820,7 +1810,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 45,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
@@ -1857,7 +1847,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 46,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -1878,7 +1868,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
@@ -1886,214 +1876,214 @@
      "output_type": "stream",
      "text": [
       "Epoch 1/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0648 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0619 - accuracy: 0.9900\n",
       "Epoch 2/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0648 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0618 - accuracy: 0.9900\n",
       "Epoch 3/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0646 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0617 - accuracy: 0.9900\n",
       "Epoch 4/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0646 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0617 - accuracy: 0.9900\n",
       "Epoch 5/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0645 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0616 - accuracy: 0.9900\n",
       "Epoch 6/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0644 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0615 - accuracy: 0.9900\n",
       "Epoch 7/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0643 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0615 - accuracy: 0.9900\n",
       "Epoch 8/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0643 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0613 - accuracy: 0.9900\n",
       "Epoch 9/100\n",
-      "3/3 [==============================] - 0s 8ms/step - loss: 0.0642 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0613 - accuracy: 0.9900\n",
       "Epoch 10/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0641 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0612 - accuracy: 0.9900\n",
       "Epoch 11/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0640 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0613 - accuracy: 0.9867\n",
       "Epoch 12/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0639 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0611 - accuracy: 0.9900\n",
       "Epoch 13/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0639 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0610 - accuracy: 0.9900\n",
       "Epoch 14/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0638 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0609 - accuracy: 0.9900\n",
       "Epoch 15/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0637 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0609 - accuracy: 0.9900\n",
       "Epoch 16/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0636 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0608 - accuracy: 0.9900\n",
       "Epoch 17/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0636 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0607 - accuracy: 0.9900\n",
       "Epoch 18/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0635 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0606 - accuracy: 0.9900\n",
       "Epoch 19/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0634 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0607 - accuracy: 0.9900\n",
       "Epoch 20/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0633 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0605 - accuracy: 0.9900\n",
       "Epoch 21/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0632 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 8ms/step - loss: 0.0604 - accuracy: 0.9900\n",
       "Epoch 22/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0633 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0604 - accuracy: 0.9900\n",
       "Epoch 23/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0630 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0604 - accuracy: 0.9900\n",
       "Epoch 24/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0631 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0602 - accuracy: 0.9900\n",
       "Epoch 25/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0629 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0602 - accuracy: 0.9900\n",
       "Epoch 26/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0628 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0601 - accuracy: 0.9900\n",
       "Epoch 27/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0628 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0601 - accuracy: 0.9900\n",
       "Epoch 28/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0627 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0599 - accuracy: 0.9900\n",
       "Epoch 29/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0626 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0599 - accuracy: 0.9900\n",
       "Epoch 30/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0626 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0598 - accuracy: 0.9900\n",
       "Epoch 31/100\n",
-      "3/3 [==============================] - 0s 8ms/step - loss: 0.0624 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0597 - accuracy: 0.9900\n",
       "Epoch 32/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0624 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0597 - accuracy: 0.9900\n",
       "Epoch 33/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0623 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0596 - accuracy: 0.9900\n",
       "Epoch 34/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0623 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0595 - accuracy: 0.9900\n",
       "Epoch 35/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0621 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0595 - accuracy: 0.9900\n",
       "Epoch 36/100\n",
-      "3/3 [==============================] - 0s 8ms/step - loss: 0.0621 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0594 - accuracy: 0.9900\n",
       "Epoch 37/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0621 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0594 - accuracy: 0.9900\n",
       "Epoch 38/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0619 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0592 - accuracy: 0.9900\n",
       "Epoch 39/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0619 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0592 - accuracy: 0.9900\n",
       "Epoch 40/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0618 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0591 - accuracy: 0.9900\n",
       "Epoch 41/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0617 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0591 - accuracy: 0.9900\n",
       "Epoch 42/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0616 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0590 - accuracy: 0.9900\n",
       "Epoch 43/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0615 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0589 - accuracy: 0.9900\n",
       "Epoch 44/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0616 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0589 - accuracy: 0.9900\n",
       "Epoch 45/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0615 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0588 - accuracy: 0.9900\n",
       "Epoch 46/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0613 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0587 - accuracy: 0.9900\n",
       "Epoch 47/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0613 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0586 - accuracy: 0.9900\n",
       "Epoch 48/100\n",
-      "3/3 [==============================] - 0s 9ms/step - loss: 0.0613 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0586 - accuracy: 0.9900\n",
       "Epoch 49/100\n",
-      "3/3 [==============================] - 0s 11ms/step - loss: 0.0611 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0585 - accuracy: 0.9900\n",
       "Epoch 50/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0611 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0584 - accuracy: 0.9900\n",
       "Epoch 51/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0609 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0585 - accuracy: 0.9900\n",
       "Epoch 52/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0609 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0583 - accuracy: 0.9900\n",
       "Epoch 53/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0608 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0582 - accuracy: 0.9900\n",
       "Epoch 54/100\n",
-      "3/3 [==============================] - 0s 16ms/step - loss: 0.0607 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0582 - accuracy: 0.9900\n",
       "Epoch 55/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0608 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0581 - accuracy: 0.9900\n",
       "Epoch 56/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0606 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0580 - accuracy: 0.9900\n",
       "Epoch 57/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0606 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0580 - accuracy: 0.9900\n",
       "Epoch 58/100\n",
-      "3/3 [==============================] - 0s 20ms/step - loss: 0.0605 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0580 - accuracy: 0.9900\n",
       "Epoch 59/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0604 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0579 - accuracy: 0.9900\n",
       "Epoch 60/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0603 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0580 - accuracy: 0.9900\n",
       "Epoch 61/100\n",
-      "3/3 [==============================] - 0s 21ms/step - loss: 0.0603 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0578 - accuracy: 0.9867\n",
       "Epoch 62/100\n",
-      "3/3 [==============================] - 0s 9ms/step - loss: 0.0602 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0578 - accuracy: 0.9867\n",
       "Epoch 63/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0601 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0577 - accuracy: 0.9900\n",
       "Epoch 64/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0601 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0575 - accuracy: 0.9900\n",
       "Epoch 65/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0600 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0574 - accuracy: 0.9900\n",
       "Epoch 66/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0599 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0574 - accuracy: 0.9900\n",
       "Epoch 67/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0598 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0573 - accuracy: 0.9900\n",
       "Epoch 68/100\n",
-      "3/3 [==============================] - 0s 15ms/step - loss: 0.0598 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0572 - accuracy: 0.9900\n",
       "Epoch 69/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0597 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0572 - accuracy: 0.9900\n",
       "Epoch 70/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0598 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0571 - accuracy: 0.9900\n",
       "Epoch 71/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0596 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0571 - accuracy: 0.9900\n",
       "Epoch 72/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0595 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0570 - accuracy: 0.9867\n",
       "Epoch 73/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0594 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0570 - accuracy: 0.9867\n",
       "Epoch 74/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0594 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0569 - accuracy: 0.9900\n",
       "Epoch 75/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0593 - accuracy: 0.9933\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0569 - accuracy: 0.9900\n",
       "Epoch 76/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0593 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0568 - accuracy: 0.9900\n",
       "Epoch 77/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0591 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0568 - accuracy: 0.9900\n",
       "Epoch 78/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0591 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0566 - accuracy: 0.9900\n",
       "Epoch 79/100\n",
-      "3/3 [==============================] - 0s 9ms/step - loss: 0.0591 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0566 - accuracy: 0.9900\n",
       "Epoch 80/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0589 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0565 - accuracy: 0.9900\n",
       "Epoch 81/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0589 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0564 - accuracy: 0.9900\n",
       "Epoch 82/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0588 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0564 - accuracy: 0.9900\n",
       "Epoch 83/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0588 - accuracy: 0.9833\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0564 - accuracy: 0.9900\n",
       "Epoch 84/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0587 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0563 - accuracy: 0.9900\n",
       "Epoch 85/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0586 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0562 - accuracy: 0.9900\n",
       "Epoch 86/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0585 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0561 - accuracy: 0.9900\n",
       "Epoch 87/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0585 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0561 - accuracy: 0.9900\n",
       "Epoch 88/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0584 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0561 - accuracy: 0.9867\n",
       "Epoch 89/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0583 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0560 - accuracy: 0.9900\n",
       "Epoch 90/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0583 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 3ms/step - loss: 0.0559 - accuracy: 0.9900\n",
       "Epoch 91/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0582 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0558 - accuracy: 0.9900\n",
       "Epoch 92/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0581 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0558 - accuracy: 0.9900\n",
       "Epoch 93/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0582 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 5ms/step - loss: 0.0557 - accuracy: 0.9900\n",
       "Epoch 94/100\n",
-      "3/3 [==============================] - 0s 5ms/step - loss: 0.0580 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0557 - accuracy: 0.9900\n",
       "Epoch 95/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0579 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0556 - accuracy: 0.9900\n",
       "Epoch 96/100\n",
-      "3/3 [==============================] - 0s 8ms/step - loss: 0.0579 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0555 - accuracy: 0.9900\n",
       "Epoch 97/100\n",
-      "3/3 [==============================] - 0s 7ms/step - loss: 0.0578 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0555 - accuracy: 0.9900\n",
       "Epoch 98/100\n",
-      "3/3 [==============================] - 0s 4ms/step - loss: 0.0577 - accuracy: 0.9900\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0554 - accuracy: 0.9900\n",
       "Epoch 99/100\n",
-      "3/3 [==============================] - 0s 3ms/step - loss: 0.0577 - accuracy: 0.9867\n",
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0554 - accuracy: 0.9900\n",
       "Epoch 100/100\n",
-      "3/3 [==============================] - 0s 6ms/step - loss: 0.0576 - accuracy: 0.9900\n"
+      "3/3 [==============================] - 0s 4ms/step - loss: 0.0553 - accuracy: 0.9900\n"
      ]
     },
     {
      "data": {
       "text/plain": [
-       "<keras.callbacks.History at 0x7fe47c15ae90>"
+       "<keras.callbacks.History at 0x7fbbb81ba650>"
       ]
      },
-     "execution_count": 49,
+     "execution_count": 33,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -2116,14 +2106,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10/10 [==============================] - 0s 3ms/step - loss: 0.0575 - accuracy: 0.9900\n",
+      "10/10 [==============================] - 0s 3ms/step - loss: 0.0619 - accuracy: 0.9900\n",
       "Accuracy on test dataset: 0.9900000095367432\n"
      ]
     }
@@ -2132,6 +2122,13 @@
     "test_loss, test_accuracy = model.evaluate(X, y_cat, steps=math.ceil(num_train_examples/32))\n",
     "print('Accuracy on test dataset:', test_accuracy)"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
-- 
GitLab