diff --git a/doc/pub/week42/ipynb/week42.ipynb b/doc/pub/week42/ipynb/week42.ipynb
index 6137ecefa..cac01cddd 100644
--- a/doc/pub/week42/ipynb/week42.ipynb
+++ b/doc/pub/week42/ipynb/week42.ipynb
@@ -327,9 +327,7 @@
   {
    "cell_type": "code",
    "execution_count": 1,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "%matplotlib inline\n",
@@ -386,9 +384,7 @@
   {
    "cell_type": "code",
    "execution_count": 2,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "from tensorflow.keras import datasets, layers, models\n",
@@ -426,9 +422,7 @@
   {
    "cell_type": "code",
    "execution_count": 3,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "def create_convolutional_neural_network_keras(input_shape, receptive_field,\n",
@@ -469,9 +463,7 @@
   {
    "cell_type": "code",
    "execution_count": 4,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "CNN_keras = np.zeros((len(eta_vals), len(lmbd_vals)), dtype=object)\n",
@@ -502,9 +494,7 @@
   {
    "cell_type": "code",
    "execution_count": 5,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "# visual representation of grid search\n",
@@ -554,9 +544,7 @@
   {
    "cell_type": "code",
    "execution_count": 6,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "import tensorflow as tf\n",
@@ -583,9 +571,7 @@
   {
    "cell_type": "code",
    "execution_count": 7,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n",
@@ -618,9 +604,7 @@
   {
    "cell_type": "code",
    "execution_count": 8,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "model = models.Sequential()\n",
@@ -658,9 +642,7 @@
   {
    "cell_type": "code",
    "execution_count": 9,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "model.add(layers.Flatten())\n",
@@ -683,9 +665,7 @@
   {
    "cell_type": "code",
    "execution_count": 10,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "model.compile(optimizer='adam',\n",
@@ -730,9 +710,7 @@
    ],
 =======
    "execution_count": 11,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
 >>>>>>> b95146b3e7172ba164de8bbc2afe0d9cc4dba4f6
    "source": [
@@ -1035,9 +1013,7 @@
    ],
 =======
    "execution_count": 12,
-   "metadata": {
-    "collapsed": false
-   },
+   "metadata": {},
    "outputs": [],
 >>>>>>> b95146b3e7172ba164de8bbc2afe0d9cc4dba4f6
    "source": [
@@ -1125,10 +1101,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {
-    "collapsed": false
-   },
+   "execution_count": 1,
+   "metadata": {},
    "outputs": [],
    "source": [
     "\n",
@@ -1204,10 +1178,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {
-    "collapsed": false
-   },
+   "execution_count": 2,
+   "metadata": {},
    "outputs": [],
    "source": [
     "# FORMAT_DATA\n",
@@ -1293,11 +1265,380 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {
-    "collapsed": false
-   },
-   "outputs": [],
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"functional_1\"\n",
+      "_________________________________________________________________\n",
+      "Layer (type)                 Output Shape              Param #   \n",
+      "=================================================================\n",
+      "input_1 (InputLayer)         [(None, 2, 1)]            0         \n",
+      "_________________________________________________________________\n",
+      "RNN (SimpleRNN)              (None, 200)               40400     \n",
+      "_________________________________________________________________\n",
+      "dense (Dense)                (None, 1)                 201       \n",
+      "=================================================================\n",
+      "Total params: 40,601\n",
+      "Trainable params: 40,601\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n",
+      "Epoch 1/150\n",
+      "1/1 [==============================] - 0s 213ms/step - loss: 0.2314 - val_loss: 0.3638\n",
+      "Epoch 2/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.1197 - val_loss: 0.1298\n",
+      "Epoch 3/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0454 - val_loss: 0.0161\n",
+      "Epoch 4/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0073 - val_loss: 0.0073\n",
+      "Epoch 5/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0013 - val_loss: 0.0667\n",
+      "Epoch 6/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0170 - val_loss: 0.1375\n",
+      "Epoch 7/150\n",
+      "1/1 [==============================] - 0s 50ms/step - loss: 0.0370 - val_loss: 0.1763\n",
+      "Epoch 8/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0477 - val_loss: 0.1734\n",
+      "Epoch 9/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0460 - val_loss: 0.1405\n",
+      "Epoch 10/150\n",
+      "1/1 [==============================] - 0s 44ms/step - loss: 0.0360 - val_loss: 0.0952\n",
+      "Epoch 11/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0229 - val_loss: 0.0521\n",
+      "Epoch 12/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0114 - val_loss: 0.0206\n",
+      "Epoch 13/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0038 - val_loss: 0.0037\n",
+      "Epoch 14/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 8.6322e-04 - val_loss: 1.6649e-04\n",
+      "Epoch 15/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0019 - val_loss: 0.0055\n",
+      "Epoch 16/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0052 - val_loss: 0.0145\n",
+      "Epoch 17/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0092 - val_loss: 0.0225\n",
+      "Epoch 18/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.0124 - val_loss: 0.0267\n",
+      "Epoch 19/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0139 - val_loss: 0.0261\n",
+      "Epoch 20/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0136 - val_loss: 0.0215\n",
+      "Epoch 21/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0117 - val_loss: 0.0147\n",
+      "Epoch 22/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0089 - val_loss: 0.0076\n",
+      "Epoch 23/150\n",
+      "1/1 [==============================] - ETA: 0s - loss: 0.005 - 0s 34ms/step - loss: 0.0058 - val_loss: 0.0023\n",
+      "Epoch 24/150\n",
+      "1/1 [==============================] - 0s 43ms/step - loss: 0.0031 - val_loss: 4.3466e-05\n",
+      "Epoch 25/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0013 - val_loss: 0.0012\n",
+      "Epoch 26/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 5.9378e-04 - val_loss: 0.0052\n",
+      "Epoch 27/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 8.9122e-04 - val_loss: 0.0107\n",
+      "Epoch 28/150\n",
+      "1/1 [==============================] - ETA: 0s - loss: 0.001 - 0s 38ms/step - loss: 0.0018 - val_loss: 0.0162\n",
+      "Epoch 29/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0030 - val_loss: 0.0200\n",
+      "Epoch 30/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0039 - val_loss: 0.0214\n",
+      "Epoch 31/150\n",
+      "1/1 [==============================] - 0s 42ms/step - loss: 0.0043 - val_loss: 0.0200\n",
+      "Epoch 32/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0040 - val_loss: 0.0166\n",
+      "Epoch 33/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0033 - val_loss: 0.0120\n",
+      "Epoch 34/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.0022 - val_loss: 0.0073\n",
+      "Epoch 35/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 0.0013 - val_loss: 0.0035\n",
+      "Epoch 36/150\n",
+      "1/1 [==============================] - 0s 44ms/step - loss: 5.6284e-04 - val_loss: 0.0011\n",
+      "Epoch 37/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 2.2901e-04 - val_loss: 7.0002e-05\n",
+      "Epoch 38/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 2.5718e-04 - val_loss: 1.6331e-04\n",
+      "Epoch 39/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 5.3654e-04 - val_loss: 8.5966e-04\n",
+      "Epoch 40/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 9.0858e-04 - val_loss: 0.0016\n",
+      "Epoch 41/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 0.0012 - val_loss: 0.0021\n",
+      "Epoch 42/150\n",
+      "1/1 [==============================] - ETA: 0s - loss: 0.001 - 0s 38ms/step - loss: 0.0014 - val_loss: 0.0021\n",
+      "Epoch 43/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0013 - val_loss: 0.0017\n",
+      "Epoch 44/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0011 - val_loss: 0.0010\n",
+      "Epoch 45/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 7.3857e-04 - val_loss: 4.0362e-04\n",
+      "Epoch 46/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 4.0645e-04 - val_loss: 4.0384e-05\n",
+      "Epoch 47/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 1.5850e-04 - val_loss: 6.0112e-05\n",
+      "Epoch 48/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 4.2795e-05 - val_loss: 4.3840e-04\n",
+      "Epoch 49/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 6.0607e-05 - val_loss: 0.0010\n",
+      "Epoch 50/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 1.7173e-04 - val_loss: 0.0016\n",
+      "Epoch 51/150\n",
+      "1/1 [==============================] - 0s 42ms/step - loss: 3.1268e-04 - val_loss: 0.0020\n",
+      "Epoch 52/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 4.2084e-04 - val_loss: 0.0021\n",
+      "Epoch 53/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 4.5551e-04 - val_loss: 0.0018\n",
+      "Epoch 54/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 4.0871e-04 - val_loss: 0.0014\n",
+      "Epoch 55/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 3.0284e-04 - val_loss: 8.4764e-04\n",
+      "Epoch 56/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 1.7795e-04 - val_loss: 3.8342e-04\n",
+      "Epoch 57/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 7.4619e-05 - val_loss: 9.2203e-05\n",
+      "Epoch 58/150\n",
+      "1/1 [==============================] - 0s 50ms/step - loss: 1.9558e-05 - val_loss: 1.2961e-08\n",
+      "Epoch 59/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.8495e-05 - val_loss: 6.5151e-05\n",
+      "Epoch 60/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 5.7745e-05 - val_loss: 2.0537e-04\n",
+      "Epoch 61/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 1.1231e-04 - val_loss: 3.3377e-04\n",
+      "Epoch 62/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 1.5653e-04 - val_loss: 3.8977e-04\n",
+      "Epoch 63/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.7324e-04 - val_loss: 3.5514e-04\n",
+      "Epoch 64/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.5830e-04 - val_loss: 2.5241e-04\n",
+      "Epoch 65/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 1.1998e-04 - val_loss: 1.2882e-04\n",
+      "Epoch 66/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 7.3908e-05 - val_loss: 3.3974e-05\n",
+      "Epoch 67/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 3.6225e-05 - val_loss: 8.9173e-10\n",
+      "Epoch 68/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.7529e-05 - val_loss: 3.1312e-05\n",
+      "Epoch 69/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.9752e-05 - val_loss: 1.0643e-04\n",
+      "Epoch 70/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 3.6726e-05 - val_loss: 1.8965e-04\n",
+      "Epoch 71/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 5.7781e-05 - val_loss: 2.4665e-04\n",
+      "Epoch 72/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 7.2560e-05 - val_loss: 2.5749e-04\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 73/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 7.5017e-05 - val_loss: 2.2206e-04\n",
+      "Epoch 74/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 6.5097e-05 - val_loss: 1.5701e-04\n",
+      "Epoch 75/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 4.7762e-05 - val_loss: 8.6576e-05\n",
+      "Epoch 76/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 3.0167e-05 - val_loss: 3.2021e-05\n",
+      "Epoch 77/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.8404e-05 - val_loss: 4.1474e-06\n",
+      "Epoch 78/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.5239e-05 - val_loss: 1.3411e-06\n",
+      "Epoch 79/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.9556e-05 - val_loss: 1.2974e-05\n",
+      "Epoch 80/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.7484e-05 - val_loss: 2.5818e-05\n",
+      "Epoch 81/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 3.4430e-05 - val_loss: 3.0307e-05\n",
+      "Epoch 82/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 3.7071e-05 - val_loss: 2.4076e-05\n",
+      "Epoch 83/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 3.4485e-05 - val_loss: 1.1759e-05\n",
+      "Epoch 84/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 2.8121e-05 - val_loss: 1.6934e-06\n",
+      "Epoch 85/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 2.0800e-05 - val_loss: 1.4238e-06\n",
+      "Epoch 86/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.5306e-05 - val_loss: 1.4082e-05\n",
+      "Epoch 87/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.3221e-05 - val_loss: 3.7102e-05\n",
+      "Epoch 88/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.4453e-05 - val_loss: 6.3555e-05\n",
+      "Epoch 89/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.7548e-05 - val_loss: 8.5208e-05\n",
+      "Epoch 90/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 2.0539e-05 - val_loss: 9.5807e-05\n",
+      "Epoch 91/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 2.1873e-05 - val_loss: 9.3198e-05\n",
+      "Epoch 92/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.1021e-05 - val_loss: 7.9546e-05\n",
+      "Epoch 93/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.8532e-05 - val_loss: 5.9859e-05\n",
+      "Epoch 94/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.5602e-05 - val_loss: 3.9699e-05\n",
+      "Epoch 95/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.3435e-05 - val_loss: 2.3201e-05\n",
+      "Epoch 96/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2703e-05 - val_loss: 1.2098e-05\n",
+      "Epoch 97/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.3337e-05 - val_loss: 5.9680e-06\n",
+      "Epoch 98/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.4687e-05 - val_loss: 3.3030e-06\n",
+      "Epoch 99/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.5906e-05 - val_loss: 2.7153e-06\n",
+      "Epoch 100/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.6372e-05 - val_loss: 3.6866e-06\n",
+      "Epoch 101/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.5922e-05 - val_loss: 6.5827e-06\n",
+      "Epoch 102/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.4848e-05 - val_loss: 1.2076e-05\n",
+      "Epoch 103/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.3687e-05 - val_loss: 2.0380e-05\n",
+      "Epoch 104/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2927e-05 - val_loss: 3.0739e-05\n",
+      "Epoch 105/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2791e-05 - val_loss: 4.1440e-05\n",
+      "Epoch 106/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.3179e-05 - val_loss: 5.0327e-05\n",
+      "Epoch 107/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.3766e-05 - val_loss: 5.5549e-05\n",
+      "Epoch 108/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.4196e-05 - val_loss: 5.6183e-05\n",
+      "Epoch 109/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.4252e-05 - val_loss: 5.2486e-05\n",
+      "Epoch 110/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 1.3938e-05 - val_loss: 4.5694e-05\n",
+      "Epoch 111/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 1.3435e-05 - val_loss: 3.7506e-05\n",
+      "Epoch 112/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.2988e-05 - val_loss: 2.9530e-05\n",
+      "Epoch 113/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2772e-05 - val_loss: 2.2882e-05\n",
+      "Epoch 114/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 1.2823e-05 - val_loss: 1.8086e-05\n",
+      "Epoch 115/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.3040e-05 - val_loss: 1.5205e-05\n",
+      "Epoch 116/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.3266e-05 - val_loss: 1.4073e-05\n",
+      "Epoch 117/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.3365e-05 - val_loss: 1.4495e-05\n",
+      "Epoch 118/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.3291e-05 - val_loss: 1.6304e-05\n",
+      "Epoch 119/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.3095e-05 - val_loss: 1.9297e-05\n",
+      "Epoch 120/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.2878e-05 - val_loss: 2.3134e-05\n",
+      "Epoch 121/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.2738e-05 - val_loss: 2.7279e-05\n",
+      "Epoch 122/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2717e-05 - val_loss: 3.1059e-05\n",
+      "Epoch 123/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2792e-05 - val_loss: 3.3808e-05\n",
+      "Epoch 124/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.2895e-05 - val_loss: 3.5055e-05\n",
+      "Epoch 125/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.2960e-05 - val_loss: 3.4655e-05\n",
+      "Epoch 126/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.2950e-05 - val_loss: 3.2807e-05\n",
+      "Epoch 127/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.2877e-05 - val_loss: 2.9971e-05\n",
+      "Epoch 128/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.2782e-05 - val_loss: 2.6721e-05\n",
+      "Epoch 129/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2712e-05 - val_loss: 2.3596e-05\n",
+      "Epoch 130/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2693e-05 - val_loss: 2.0996e-05\n",
+      "Epoch 131/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 1.2720e-05 - val_loss: 1.9157e-05\n",
+      "Epoch 132/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 1.2767e-05 - val_loss: 1.8173e-05\n",
+      "Epoch 133/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 1.2801e-05 - val_loss: 1.8033e-05\n",
+      "Epoch 134/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2803e-05 - val_loss: 1.8661e-05\n",
+      "Epoch 135/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.2775e-05 - val_loss: 1.9911e-05\n",
+      "Epoch 136/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.2734e-05 - val_loss: 2.1584e-05\n",
+      "Epoch 137/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.2702e-05 - val_loss: 2.3414e-05\n",
+      "Epoch 138/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.2692e-05 - val_loss: 2.5118e-05\n",
+      "Epoch 139/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.2703e-05 - val_loss: 2.6420e-05\n",
+      "Epoch 140/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.2724e-05 - val_loss: 2.7130e-05\n",
+      "Epoch 141/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.2739e-05 - val_loss: 2.7172e-05\n",
+      "Epoch 142/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.2740e-05 - val_loss: 2.6600e-05\n",
+      "Epoch 143/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.2727e-05 - val_loss: 2.5571e-05\n",
+      "Epoch 144/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.2708e-05 - val_loss: 2.4305e-05\n",
+      "Epoch 145/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.2694e-05 - val_loss: 2.3031e-05\n",
+      "Epoch 146/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.2690e-05 - val_loss: 2.1944e-05\n",
+      "Epoch 147/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.2695e-05 - val_loss: 2.1179e-05\n",
+      "Epoch 148/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2703e-05 - val_loss: 2.0816e-05\n",
+      "Epoch 149/150\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2710e-05 - val_loss: 2.0866e-05\n",
+      "Epoch 150/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 1.2709e-05 - val_loss: 2.1290e-05\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MSE:  3.342350373241417e-05\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time:  7.558490419999998\n"
+     ]
+    }
+   ],
    "source": [
     "def test_rnn (x1, y_test, plot_min, plot_max):\n",
     "    \"\"\"\n",
@@ -1413,11 +1754,376 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {
-    "collapsed": false
-   },
-   "outputs": [],
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"functional_3\"\n",
+      "_________________________________________________________________\n",
+      "Layer (type)                 Output Shape              Param #   \n",
+      "=================================================================\n",
+      "input_2 (InputLayer)         [(None, 2, 1)]            0         \n",
+      "_________________________________________________________________\n",
+      "RNN1 (SimpleRNN)             (None, 2, 500)            251000    \n",
+      "_________________________________________________________________\n",
+      "RNN2 (SimpleRNN)             (None, 500)               500500    \n",
+      "_________________________________________________________________\n",
+      "dense (Dense)                (None, 1)                 501       \n",
+      "=================================================================\n",
+      "Total params: 752,001\n",
+      "Trainable params: 752,001\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n",
+      "Epoch 1/150\n",
+      "1/1 [==============================] - 0s 264ms/step - loss: 0.1271 - val_loss: 14.1784\n",
+      "Epoch 2/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 11.2068 - val_loss: 1.7715\n",
+      "Epoch 3/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.8661 - val_loss: 1.2249\n",
+      "Epoch 4/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 2.3802 - val_loss: 3.7312\n",
+      "Epoch 5/150\n",
+      "1/1 [==============================] - 0s 41ms/step - loss: 5.5823 - val_loss: 2.3168\n",
+      "Epoch 6/150\n",
+      "1/1 [==============================] - 0s 47ms/step - loss: 3.8225 - val_loss: 0.3310\n",
+      "Epoch 7/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 1.0375 - val_loss: 0.2287\n",
+      "Epoch 8/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0450 - val_loss: 1.7714\n",
+      "Epoch 9/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.8667 - val_loss: 3.2620\n",
+      "Epoch 10/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 1.9553 - val_loss: 3.5397\n",
+      "Epoch 11/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 2.1693 - val_loss: 2.6810\n",
+      "Epoch 12/150\n",
+      "1/1 [==============================] - 0s 44ms/step - loss: 1.5171 - val_loss: 1.4175\n",
+      "Epoch 13/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.6315 - val_loss: 0.4360\n",
+      "Epoch 14/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 0.0984 - val_loss: 0.0240\n",
+      "Epoch 15/150\n",
+      "1/1 [==============================] - 0s 43ms/step - loss: 0.1136 - val_loss: 0.0574\n",
+      "Epoch 16/150\n",
+      "1/1 [==============================] - 0s 48ms/step - loss: 0.4804 - val_loss: 0.2213\n",
+      "Epoch 17/150\n",
+      "1/1 [==============================] - 0s 41ms/step - loss: 0.8395 - val_loss: 0.2740\n",
+      "Epoch 18/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.9370 - val_loss: 0.1738\n",
+      "Epoch 19/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.7467 - val_loss: 0.0363\n",
+      "Epoch 20/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.4179 - val_loss: 0.0110\n",
+      "Epoch 21/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.1429 - val_loss: 0.1731\n",
+      "Epoch 22/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0419 - val_loss: 0.4825\n",
+      "Epoch 23/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.1159 - val_loss: 0.8140\n",
+      "Epoch 24/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.2720 - val_loss: 1.0321\n",
+      "Epoch 25/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.3940 - val_loss: 1.0599\n",
+      "Epoch 26/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.4103 - val_loss: 0.9069\n",
+      "Epoch 27/150\n",
+      "1/1 [==============================] - 0s 54ms/step - loss: 0.3226 - val_loss: 0.6488\n",
+      "Epoch 28/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.1886 - val_loss: 0.3815\n",
+      "Epoch 29/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0800 - val_loss: 0.1764\n",
+      "Epoch 30/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0419 - val_loss: 0.0587\n",
+      "Epoch 31/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0743 - val_loss: 0.0118\n",
+      "Epoch 32/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.1404 - val_loss: 0.0012\n",
+      "Epoch 33/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.1927 - val_loss: 5.7056e-04\n",
+      "Epoch 34/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.2007 - val_loss: 0.0054\n",
+      "Epoch 35/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.1636 - val_loss: 0.0288\n",
+      "Epoch 36/150\n",
+      "1/1 [==============================] - 0s 45ms/step - loss: 0.1056 - val_loss: 0.0866\n",
+      "Epoch 37/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.0582 - val_loss: 0.1815\n",
+      "Epoch 38/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0419 - val_loss: 0.2965\n",
+      "Epoch 39/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0568 - val_loss: 0.4012\n",
+      "Epoch 40/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0864 - val_loss: 0.4649\n",
+      "Epoch 41/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.1092 - val_loss: 0.4708\n",
+      "Epoch 42/150\n",
+      "1/1 [==============================] - 0s 42ms/step - loss: 0.1115 - val_loss: 0.4221\n",
+      "Epoch 43/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0936 - val_loss: 0.3383\n",
+      "Epoch 44/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0672 - val_loss: 0.2453\n",
+      "Epoch 45/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0472 - val_loss: 0.1644\n",
+      "Epoch 46/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0421 - val_loss: 0.1068\n",
+      "Epoch 47/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0509 - val_loss: 0.0732\n",
+      "Epoch 48/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0647 - val_loss: 0.0594\n",
+      "Epoch 49/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0736 - val_loss: 0.0615\n",
+      "Epoch 50/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0720 - val_loss: 0.0785\n",
+      "Epoch 51/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 0.0619 - val_loss: 0.1104\n",
+      "Epoch 52/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0498 - val_loss: 0.1554\n",
+      "Epoch 53/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0425 - val_loss: 0.2071\n",
+      "Epoch 54/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0429 - val_loss: 0.2546\n",
+      "Epoch 55/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0486 - val_loss: 0.2870\n",
+      "Epoch 56/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0547 - val_loss: 0.2966\n",
+      "Epoch 57/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0568 - val_loss: 0.2828\n",
+      "Epoch 58/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0539 - val_loss: 0.2514\n",
+      "Epoch 59/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0481 - val_loss: 0.2116\n",
+      "Epoch 60/150\n",
+      "1/1 [==============================] - 0s 40ms/step - loss: 0.0432 - val_loss: 0.1729\n",
+      "Epoch 61/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0418 - val_loss: 0.1420\n",
+      "Epoch 62/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.0437 - val_loss: 0.1220\n",
+      "Epoch 63/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0469 - val_loss: 0.1137\n",
+      "Epoch 64/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0488 - val_loss: 0.1166\n",
+      "Epoch 65/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0481 - val_loss: 0.1294\n",
+      "Epoch 66/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0455 - val_loss: 0.1501\n",
+      "Epoch 67/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0428 - val_loss: 0.1754\n",
+      "Epoch 68/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0417 - val_loss: 0.2005\n",
+      "Epoch 69/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0424 - val_loss: 0.2200\n",
+      "Epoch 70/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0439 - val_loss: 0.2299\n",
+      "Epoch 71/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0450 - val_loss: 0.2284\n",
+      "Epoch 72/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0449 - val_loss: 0.2170\n",
+      "Epoch 73/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0436 - val_loss: 0.1993\n",
+      "Epoch 74/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0423 - val_loss: 0.1797\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 75/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0416 - val_loss: 0.1624\n",
+      "Epoch 76/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0419 - val_loss: 0.1503\n",
+      "Epoch 77/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0427 - val_loss: 0.1446\n",
+      "Epoch 78/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0433 - val_loss: 0.1456\n",
+      "Epoch 79/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0432 - val_loss: 0.1525\n",
+      "Epoch 80/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0425 - val_loss: 0.1639\n",
+      "Epoch 81/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0418 - val_loss: 0.1772\n",
+      "Epoch 82/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.0416 - val_loss: 0.1897\n",
+      "Epoch 83/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0418 - val_loss: 0.1987\n",
+      "Epoch 84/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0422 - val_loss: 0.2024\n",
+      "Epoch 85/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0424 - val_loss: 0.2004\n",
+      "Epoch 86/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0423 - val_loss: 0.1937\n",
+      "Epoch 87/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0419 - val_loss: 0.1842\n",
+      "Epoch 88/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0416 - val_loss: 0.1743\n",
+      "Epoch 89/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0415 - val_loss: 0.1661\n",
+      "Epoch 90/150\n",
+      "1/1 [==============================] - 0s 20ms/step - loss: 0.0417 - val_loss: 0.1611\n",
+      "Epoch 91/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0419 - val_loss: 0.1598\n",
+      "Epoch 92/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 0.0419 - val_loss: 0.1623\n",
+      "Epoch 93/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0418 - val_loss: 0.1676\n",
+      "Epoch 94/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0416 - val_loss: 0.1744\n",
+      "Epoch 95/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 0.0415 - val_loss: 0.1813\n",
+      "Epoch 96/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.0415 - val_loss: 0.1865\n",
+      "Epoch 97/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0416 - val_loss: 0.1891\n",
+      "Epoch 98/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0417 - val_loss: 0.1887\n",
+      "Epoch 99/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0416 - val_loss: 0.1855\n",
+      "Epoch 100/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0415 - val_loss: 0.1807\n",
+      "Epoch 101/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.0414 - val_loss: 0.1754\n",
+      "Epoch 102/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0414 - val_loss: 0.1709\n",
+      "Epoch 103/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0414 - val_loss: 0.1682\n",
+      "Epoch 104/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0415 - val_loss: 0.1675\n",
+      "Epoch 105/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0415 - val_loss: 0.1690\n",
+      "Epoch 106/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.0414 - val_loss: 0.1719\n",
+      "Epoch 107/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0414 - val_loss: 0.1756\n",
+      "Epoch 108/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0414 - val_loss: 0.1792\n",
+      "Epoch 109/150\n",
+      "1/1 [==============================] - 0s 41ms/step - loss: 0.0414 - val_loss: 0.1817\n",
+      "Epoch 110/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0414 - val_loss: 0.1827\n",
+      "Epoch 111/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0414 - val_loss: 0.1820\n",
+      "Epoch 112/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0414 - val_loss: 0.1799\n",
+      "Epoch 113/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0413 - val_loss: 0.1772\n",
+      "Epoch 114/150\n",
+      "1/1 [==============================] - 0s 58ms/step - loss: 0.0413 - val_loss: 0.1744\n",
+      "Epoch 115/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.0413 - val_loss: 0.1723\n",
+      "Epoch 116/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0413 - val_loss: 0.1713\n",
+      "Epoch 117/150\n",
+      "1/1 [==============================] - 0s 40ms/step - loss: 0.0413 - val_loss: 0.1714\n",
+      "Epoch 118/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0413 - val_loss: 0.1727\n",
+      "Epoch 119/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0413 - val_loss: 0.1745\n",
+      "Epoch 120/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0412 - val_loss: 0.1765\n",
+      "Epoch 121/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0412 - val_loss: 0.1781\n",
+      "Epoch 122/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0412 - val_loss: 0.1790\n",
+      "Epoch 123/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0412 - val_loss: 0.1789\n",
+      "Epoch 124/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0412 - val_loss: 0.1780\n",
+      "Epoch 125/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0412 - val_loss: 0.1765\n",
+      "Epoch 126/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0412 - val_loss: 0.1750\n",
+      "Epoch 127/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0412 - val_loss: 0.1737\n",
+      "Epoch 128/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0412 - val_loss: 0.1729\n",
+      "Epoch 129/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0412 - val_loss: 0.1729\n",
+      "Epoch 130/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0412 - val_loss: 0.1735\n",
+      "Epoch 131/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.0411 - val_loss: 0.1745\n",
+      "Epoch 132/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0411 - val_loss: 0.1756\n",
+      "Epoch 133/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0411 - val_loss: 0.1764\n",
+      "Epoch 134/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0411 - val_loss: 0.1769\n",
+      "Epoch 135/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0411 - val_loss: 0.1768\n",
+      "Epoch 136/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0411 - val_loss: 0.1763\n",
+      "Epoch 137/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0411 - val_loss: 0.1755\n",
+      "Epoch 138/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0411 - val_loss: 0.1746\n",
+      "Epoch 139/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0410 - val_loss: 0.1739\n",
+      "Epoch 140/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0410 - val_loss: 0.1735\n",
+      "Epoch 141/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0410 - val_loss: 0.1735\n",
+      "Epoch 142/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0410 - val_loss: 0.1739\n",
+      "Epoch 143/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0410 - val_loss: 0.1744\n",
+      "Epoch 144/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0410 - val_loss: 0.1750\n",
+      "Epoch 145/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0410 - val_loss: 0.1754\n",
+      "Epoch 146/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0410 - val_loss: 0.1756\n",
+      "Epoch 147/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0410 - val_loss: 0.1754\n",
+      "Epoch 148/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0410 - val_loss: 0.1750\n",
+      "Epoch 149/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0409 - val_loss: 0.1745\n",
+      "Epoch 150/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0409 - val_loss: 0.1740\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MSE:  0.33070847849998325\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time:  9.590071408\n"
+     ]
+    }
+   ],
    "source": [
     "def rnn_2layers(length_of_sequences, batch_size = None, stateful = False):\n",
     "    \"\"\"\n",
@@ -1531,11 +2237,737 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {
-    "collapsed": false
-   },
-   "outputs": [],
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"functional_5\"\n",
+      "_________________________________________________________________\n",
+      "Layer (type)                 Output Shape              Param #   \n",
+      "=================================================================\n",
+      "input_3 (InputLayer)         [(None, 2, 1)]            0         \n",
+      "_________________________________________________________________\n",
+      "dnn (Dense)                  (None, 2, 125)            250       \n",
+      "_________________________________________________________________\n",
+      "dnn1 (Dense)                 (None, 2, 125)            15750     \n",
+      "_________________________________________________________________\n",
+      "RNN1 (GRU)                   (None, 2, 250)            282750    \n",
+      "_________________________________________________________________\n",
+      "RNN (GRU)                    (None, 250)               376500    \n",
+      "_________________________________________________________________\n",
+      "dense (Dense)                (None, 1)                 251       \n",
+      "=================================================================\n",
+      "Total params: 675,501\n",
+      "Trainable params: 675,501\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n",
+      "Epoch 1/150\n",
+      "1/1 [==============================] - 1s 740ms/step - loss: 0.2439 - val_loss: 0.5786\n",
+      "Epoch 2/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.1770 - val_loss: 0.4062\n",
+      "Epoch 3/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 0.1168 - val_loss: 0.2379\n",
+      "Epoch 4/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0611 - val_loss: 0.0889\n",
+      "Epoch 5/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0183 - val_loss: 0.0040\n",
+      "Epoch 6/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0079 - val_loss: 0.0236\n",
+      "Epoch 7/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0389 - val_loss: 0.0414\n",
+      "Epoch 8/150\n",
+      "1/1 [==============================] - 0s 40ms/step - loss: 0.0509 - val_loss: 0.0189\n",
+      "Epoch 9/150\n",
+      "1/1 [==============================] - 0s 46ms/step - loss: 0.0341 - val_loss: 5.2678e-04\n",
+      "Epoch 10/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 0.0142 - val_loss: 0.0094\n",
+      "Epoch 11/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0054 - val_loss: 0.0393\n",
+      "Epoch 12/150\n",
+      "1/1 [==============================] - 0s 45ms/step - loss: 0.0079 - val_loss: 0.0727\n",
+      "Epoch 13/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0152 - val_loss: 0.0952\n",
+      "Epoch 14/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0212 - val_loss: 0.1012\n",
+      "Epoch 15/150\n",
+      "1/1 [==============================] - 0s 42ms/step - loss: 0.0231 - val_loss: 0.0915\n",
+      "Epoch 16/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 0.0206 - val_loss: 0.0711\n",
+      "Epoch 17/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0153 - val_loss: 0.0462\n",
+      "Epoch 18/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0093 - val_loss: 0.0231\n",
+      "Epoch 19/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0049 - val_loss: 0.0071\n",
+      "Epoch 20/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0035 - val_loss: 4.5431e-04\n",
+      "Epoch 21/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0052 - val_loss: 7.3521e-04\n",
+      "Epoch 22/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0082 - val_loss: 0.0027\n",
+      "Epoch 23/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0099 - val_loss: 0.0027\n",
+      "Epoch 24/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 0.0089 - val_loss: 8.7922e-04\n",
+      "Epoch 25/150\n",
+      "1/1 [==============================] - 0s 44ms/step - loss: 0.0059 - val_loss: 3.7683e-05\n",
+      "Epoch 26/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0029 - val_loss: 0.0022\n",
+      "Epoch 27/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0014 - val_loss: 0.0073\n",
+      "Epoch 28/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 0.0017 - val_loss: 0.0127\n",
+      "Epoch 29/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 0.0029 - val_loss: 0.0159\n",
+      "Epoch 30/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 0.0039 - val_loss: 0.0155\n",
+      "Epoch 31/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0040 - val_loss: 0.0120\n",
+      "Epoch 32/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0032 - val_loss: 0.0068\n",
+      "Epoch 33/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0018 - val_loss: 0.0023\n",
+      "Epoch 34/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 6.0179e-04 - val_loss: 1.0208e-04\n",
+      "Epoch 35/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 1.3386e-04 - val_loss: 6.0544e-04\n",
+      "Epoch 36/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 4.8882e-04 - val_loss: 0.0026\n",
+      "Epoch 37/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0012 - val_loss: 0.0043\n",
+      "Epoch 38/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0017 - val_loss: 0.0042\n",
+      "Epoch 39/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0015 - val_loss: 0.0028\n",
+      "Epoch 40/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 8.1276e-04 - val_loss: 0.0011\n",
+      "Epoch 41/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 2.2473e-04 - val_loss: 9.7403e-05\n",
+      "Epoch 42/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 8.3424e-05 - val_loss: 9.9143e-05\n",
+      "Epoch 43/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 3.7649e-04 - val_loss: 5.0879e-04\n",
+      "Epoch 44/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 7.5683e-04 - val_loss: 6.9709e-04\n",
+      "Epoch 45/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 9.1141e-04 - val_loss: 4.7296e-04\n",
+      "Epoch 46/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 7.5614e-04 - val_loss: 1.0445e-04\n",
+      "Epoch 47/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 4.3201e-04 - val_loss: 2.7720e-05\n",
+      "Epoch 48/150\n",
+      "1/1 [==============================] - 0s 45ms/step - loss: 1.6201e-04 - val_loss: 4.6725e-04\n",
+      "Epoch 49/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 9.5878e-05 - val_loss: 0.0013\n",
+      "Epoch 50/150\n",
+      "1/1 [==============================] - 0s 47ms/step - loss: 2.1512e-04 - val_loss: 0.0019\n",
+      "Epoch 51/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 3.7242e-04 - val_loss: 0.0020\n",
+      "Epoch 52/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 4.1564e-04 - val_loss: 0.0016\n",
+      "Epoch 53/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 3.0959e-04 - val_loss: 8.1396e-04\n",
+      "Epoch 54/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.4397e-04 - val_loss: 2.2728e-04\n",
+      "Epoch 55/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 3.9068e-05 - val_loss: 3.8224e-06\n",
+      "Epoch 56/150\n",
+      "1/1 [==============================] - 0s 49ms/step - loss: 4.7279e-05 - val_loss: 7.0458e-05\n",
+      "Epoch 57/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.2773e-04 - val_loss: 2.0730e-04\n",
+      "Epoch 58/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.9711e-04 - val_loss: 2.4006e-04\n",
+      "Epoch 59/150\n",
+      "1/1 [==============================] - 0s 37ms/step - loss: 1.9807e-04 - val_loss: 1.4621e-04\n",
+      "Epoch 60/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 1.3455e-04 - val_loss: 2.9298e-05\n",
+      "Epoch 61/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 5.7944e-05 - val_loss: 7.6475e-06\n",
+      "Epoch 62/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 2.2173e-05 - val_loss: 1.0943e-04\n",
+      "Epoch 63/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 4.2786e-05 - val_loss: 2.5298e-04\n",
+      "Epoch 64/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 8.9687e-05 - val_loss: 3.2226e-04\n",
+      "Epoch 65/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.1680e-04 - val_loss: 2.6685e-04\n",
+      "Epoch 66/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.0146e-04 - val_loss: 1.3679e-04\n",
+      "Epoch 67/150\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 5.9634e-05 - val_loss: 2.8498e-05\n",
+      "Epoch 68/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 2.6013e-05 - val_loss: 1.3240e-06\n",
+      "Epoch 69/150\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1/1 [==============================] - 0s 33ms/step - loss: 2.2518e-05 - val_loss: 3.9676e-05\n",
+      "Epoch 70/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 4.3177e-05 - val_loss: 8.2166e-05\n",
+      "Epoch 71/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 6.3429e-05 - val_loss: 8.1303e-05\n",
+      "Epoch 72/150\n",
+      "1/1 [==============================] - 0s 69ms/step - loss: 6.3367e-05 - val_loss: 4.0620e-05\n",
+      "Epoch 73/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 4.3226e-05 - val_loss: 3.6459e-06\n",
+      "Epoch 74/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 2.0107e-05 - val_loss: 1.0472e-05\n",
+      "Epoch 75/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.1214e-05 - val_loss: 6.1334e-05\n",
+      "Epoch 76/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.9195e-05 - val_loss: 1.1775e-04\n",
+      "Epoch 77/150\n",
+      "1/1 [==============================] - 0s 34ms/step - loss: 3.1831e-05 - val_loss: 1.3756e-04\n",
+      "Epoch 78/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 3.5069e-05 - val_loss: 1.0979e-04\n",
+      "Epoch 79/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 2.5985e-05 - val_loss: 5.8554e-05\n",
+      "Epoch 80/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.3632e-05 - val_loss: 1.7171e-05\n",
+      "Epoch 81/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 8.6392e-06 - val_loss: 8.4256e-07\n",
+      "Epoch 82/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.2811e-05 - val_loss: 1.5706e-06\n",
+      "Epoch 83/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.8968e-05 - val_loss: 3.8630e-06\n",
+      "Epoch 84/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.9619e-05 - val_loss: 1.5403e-06\n",
+      "Epoch 85/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.3849e-05 - val_loss: 2.7300e-07\n",
+      "Epoch 86/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 7.2519e-06 - val_loss: 7.3492e-06\n",
+      "Epoch 87/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 4.4734e-06 - val_loss: 2.1665e-05\n",
+      "Epoch 88/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 6.4300e-06 - val_loss: 3.3634e-05\n",
+      "Epoch 89/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 9.8174e-06 - val_loss: 3.4244e-05\n",
+      "Epoch 90/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.0702e-05 - val_loss: 2.3429e-05\n",
+      "Epoch 91/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 8.2523e-06 - val_loss: 9.5217e-06\n",
+      "Epoch 92/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 4.9514e-06 - val_loss: 1.1861e-06\n",
+      "Epoch 93/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 3.6405e-06 - val_loss: 3.8227e-07\n",
+      "Epoch 94/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 4.7607e-06 - val_loss: 2.6255e-06\n",
+      "Epoch 95/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 6.3592e-06 - val_loss: 2.9419e-06\n",
+      "Epoch 96/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 6.3621e-06 - val_loss: 9.0631e-07\n",
+      "Epoch 97/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 4.5782e-06 - val_loss: 1.6043e-07\n",
+      "Epoch 98/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 2.5825e-06 - val_loss: 3.7664e-06\n",
+      "Epoch 99/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.9281e-06 - val_loss: 1.0553e-05\n",
+      "Epoch 100/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 2.6603e-06 - val_loss: 1.6018e-05\n",
+      "Epoch 101/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 3.5260e-06 - val_loss: 1.6476e-05\n",
+      "Epoch 102/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 3.4259e-06 - val_loss: 1.2123e-05\n",
+      "Epoch 103/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 2.4713e-06 - val_loss: 6.2562e-06\n",
+      "Epoch 104/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 1.6422e-06 - val_loss: 2.0621e-06\n",
+      "Epoch 105/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.6246e-06 - val_loss: 3.4961e-07\n",
+      "Epoch 106/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 2.1502e-06 - val_loss: 2.9899e-08\n",
+      "Epoch 107/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 2.4330e-06 - val_loss: 5.6929e-08\n",
+      "Epoch 108/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 2.0696e-06 - val_loss: 4.8251e-07\n",
+      "Epoch 109/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.4020e-06 - val_loss: 1.7942e-06\n",
+      "Epoch 110/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.0694e-06 - val_loss: 3.7750e-06\n",
+      "Epoch 111/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.1765e-06 - val_loss: 5.3220e-06\n",
+      "Epoch 112/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.4633e-06 - val_loss: 5.3882e-06\n",
+      "Epoch 113/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.5478e-06 - val_loss: 3.9626e-06\n",
+      "Epoch 114/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.3358e-06 - val_loss: 2.0271e-06\n",
+      "Epoch 115/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.0631e-06 - val_loss: 6.2920e-07\n",
+      "Epoch 116/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 9.8944e-07 - val_loss: 7.7514e-08\n",
+      "Epoch 117/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.1215e-06 - val_loss: 1.6729e-10\n",
+      "Epoch 118/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2486e-06 - val_loss: 2.2454e-09\n",
+      "Epoch 119/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.2010e-06 - val_loss: 1.1726e-07\n",
+      "Epoch 120/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.0210e-06 - val_loss: 6.1884e-07\n",
+      "Epoch 121/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 8.8382e-07 - val_loss: 1.5261e-06\n",
+      "Epoch 122/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 8.9094e-07 - val_loss: 2.4220e-06\n",
+      "Epoch 123/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 9.7694e-07 - val_loss: 2.7847e-06\n",
+      "Epoch 124/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.0091e-06 - val_loss: 2.4430e-06\n",
+      "Epoch 125/150\n",
+      "1/1 [==============================] - 0s 35ms/step - loss: 9.4181e-07 - val_loss: 1.6817e-06\n",
+      "Epoch 126/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 8.4908e-07 - val_loss: 9.4369e-07\n",
+      "Epoch 127/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 8.2222e-07 - val_loss: 4.8508e-07\n",
+      "Epoch 128/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 8.6587e-07 - val_loss: 3.0635e-07\n",
+      "Epoch 129/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 9.0675e-07 - val_loss: 3.3029e-07\n",
+      "Epoch 130/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 8.8686e-07 - val_loss: 5.4134e-07\n",
+      "Epoch 131/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 8.2402e-07 - val_loss: 9.3964e-07\n",
+      "Epoch 132/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 7.7937e-07 - val_loss: 1.4157e-06\n",
+      "Epoch 133/150\n",
+      "1/1 [==============================] - 0s 36ms/step - loss: 7.8307e-07 - val_loss: 1.7498e-06\n",
+      "Epoch 134/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 8.0683e-07 - val_loss: 1.7642e-06\n",
+      "Epoch 135/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 8.0639e-07 - val_loss: 1.4662e-06\n",
+      "Epoch 136/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 7.7428e-07 - val_loss: 1.0310e-06\n",
+      "Epoch 137/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 7.4158e-07 - val_loss: 6.5354e-07\n",
+      "Epoch 138/150\n",
+      "1/1 [==============================] - 0s 38ms/step - loss: 7.3569e-07 - val_loss: 4.3050e-07\n",
+      "Epoch 139/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 7.4955e-07 - val_loss: 3.6220e-07\n",
+      "Epoch 140/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 7.5524e-07 - val_loss: 4.2824e-07\n",
+      "Epoch 141/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 7.3895e-07 - val_loss: 6.2212e-07\n",
+      "Epoch 142/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 7.1419e-07 - val_loss: 9.1359e-07\n",
+      "Epoch 143/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 7.0133e-07 - val_loss: 1.2129e-06\n",
+      "Epoch 144/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 7.0322e-07 - val_loss: 1.4017e-06\n",
+      "Epoch 145/150\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1/1 [==============================] - 0s 27ms/step - loss: 7.0534e-07 - val_loss: 1.4100e-06\n",
+      "Epoch 146/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 6.9584e-07 - val_loss: 1.2624e-06\n",
+      "Epoch 147/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 6.7914e-07 - val_loss: 1.0526e-06\n",
+      "Epoch 148/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 6.6792e-07 - val_loss: 8.7604e-07\n",
+      "Epoch 149/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 6.6668e-07 - val_loss: 7.8853e-07\n",
+      "Epoch 150/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 6.6784e-07 - val_loss: 8.0641e-07\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MSE:  7.069409480740414e-06\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time:  12.516543079999991\n",
+      "Model: \"functional_7\"\n",
+      "_________________________________________________________________\n",
+      "Layer (type)                 Output Shape              Param #   \n",
+      "=================================================================\n",
+      "input_4 (InputLayer)         [(None, 1, 1)]            0         \n",
+      "_________________________________________________________________\n",
+      "RNN (SimpleRNN)              (None, 200)               40400     \n",
+      "_________________________________________________________________\n",
+      "dense (Dense)                (None, 1)                 201       \n",
+      "=================================================================\n",
+      "Total params: 40,601\n",
+      "Trainable params: 40,601\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n",
+      "Epoch 1/150\n",
+      "1/1 [==============================] - 0s 181ms/step - loss: 0.1362 - val_loss: 0.2706\n",
+      "Epoch 2/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0727 - val_loss: 0.0929\n",
+      "Epoch 3/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0761 - val_loss: 0.0485\n",
+      "Epoch 4/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0776 - val_loss: 0.0542\n",
+      "Epoch 5/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 0.0592 - val_loss: 0.0864\n",
+      "Epoch 6/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0385 - val_loss: 0.1339\n",
+      "Epoch 7/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0286 - val_loss: 0.1741\n",
+      "Epoch 8/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0288 - val_loss: 0.1825\n",
+      "Epoch 9/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0293 - val_loss: 0.1550\n",
+      "Epoch 10/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 0.0241 - val_loss: 0.1070\n",
+      "Epoch 11/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0150 - val_loss: 0.0588\n",
+      "Epoch 12/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0077 - val_loss: 0.0244\n",
+      "Epoch 13/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0056 - val_loss: 0.0072\n",
+      "Epoch 14/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0073 - val_loss: 0.0017\n",
+      "Epoch 15/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0090 - val_loss: 6.9909e-04\n",
+      "Epoch 16/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0079 - val_loss: 0.0014\n",
+      "Epoch 17/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 0.0050 - val_loss: 0.0041\n",
+      "Epoch 18/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0030 - val_loss: 0.0089\n",
+      "Epoch 19/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0035 - val_loss: 0.0134\n",
+      "Epoch 20/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0057 - val_loss: 0.0145\n",
+      "Epoch 21/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0075 - val_loss: 0.0112\n",
+      "Epoch 22/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0074 - val_loss: 0.0057\n",
+      "Epoch 23/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0061 - val_loss: 0.0014\n",
+      "Epoch 24/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0052 - val_loss: 2.7612e-07\n",
+      "Epoch 25/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0056 - val_loss: 6.5795e-04\n",
+      "Epoch 26/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0066 - val_loss: 0.0012\n",
+      "Epoch 27/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0070 - val_loss: 7.4705e-04\n",
+      "Epoch 28/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 0.0063 - val_loss: 2.8319e-05\n",
+      "Epoch 29/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 0.0051 - val_loss: 5.6987e-04\n",
+      "Epoch 30/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0042 - val_loss: 0.0027\n",
+      "Epoch 31/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0041 - val_loss: 0.0052\n",
+      "Epoch 32/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0043 - val_loss: 0.0065\n",
+      "Epoch 33/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0040 - val_loss: 0.0059\n",
+      "Epoch 34/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0031 - val_loss: 0.0041\n",
+      "Epoch 35/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0023 - val_loss: 0.0025\n",
+      "Epoch 36/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0019 - val_loss: 0.0015\n",
+      "Epoch 37/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0019 - val_loss: 0.0013\n",
+      "Epoch 38/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0020 - val_loss: 0.0020\n",
+      "Epoch 39/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 0.0017 - val_loss: 0.0038\n",
+      "Epoch 40/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0013 - val_loss: 0.0067\n",
+      "Epoch 41/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0010 - val_loss: 0.0102\n",
+      "Epoch 42/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0011 - val_loss: 0.0128\n",
+      "Epoch 43/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 0.0013 - val_loss: 0.0137\n",
+      "Epoch 44/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 0.0014 - val_loss: 0.0126\n",
+      "Epoch 45/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 0.0012 - val_loss: 0.0103\n",
+      "Epoch 46/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0010 - val_loss: 0.0081\n",
+      "Epoch 47/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 0.0011 - val_loss: 0.0067\n",
+      "Epoch 48/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0012 - val_loss: 0.0063\n",
+      "Epoch 49/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 0.0012 - val_loss: 0.0069\n",
+      "Epoch 50/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 0.0011 - val_loss: 0.0082\n",
+      "Epoch 51/150\n",
+      "1/1 [==============================] - 0s 20ms/step - loss: 9.6670e-04 - val_loss: 0.0098\n",
+      "Epoch 52/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 9.1618e-04 - val_loss: 0.0107\n",
+      "Epoch 53/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 9.3431e-04 - val_loss: 0.0106\n",
+      "Epoch 54/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 9.0218e-04 - val_loss: 0.0093\n",
+      "Epoch 55/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 7.7903e-04 - val_loss: 0.0074\n",
+      "Epoch 56/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 6.4494e-04 - val_loss: 0.0056\n",
+      "Epoch 57/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 5.8288e-04 - val_loss: 0.0043\n",
+      "Epoch 58/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 5.7450e-04 - val_loss: 0.0036\n",
+      "Epoch 59/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 5.4250e-04 - val_loss: 0.0036\n",
+      "Epoch 60/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 4.6284e-04 - val_loss: 0.0039\n",
+      "Epoch 61/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 3.8852e-04 - val_loss: 0.0043\n",
+      "Epoch 62/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 3.6682e-04 - val_loss: 0.0044\n",
+      "Epoch 63/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 3.7627e-04 - val_loss: 0.0041\n",
+      "Epoch 64/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 3.6554e-04 - val_loss: 0.0034\n",
+      "Epoch 65/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 3.2638e-04 - val_loss: 0.0026\n",
+      "Epoch 66/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.9607e-04 - val_loss: 0.0020\n",
+      "Epoch 67/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 2.9687e-04 - val_loss: 0.0016\n",
+      "Epoch 68/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 3.0603e-04 - val_loss: 0.0015\n",
+      "Epoch 69/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.9349e-04 - val_loss: 0.0016\n",
+      "Epoch 70/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 2.6434e-04 - val_loss: 0.0019\n",
+      "Epoch 71/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 2.4509e-04 - val_loss: 0.0021\n",
+      "Epoch 72/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 2.4213e-04 - val_loss: 0.0022\n",
+      "Epoch 73/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 2.3593e-04 - val_loss: 0.0021\n",
+      "Epoch 74/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.1383e-04 - val_loss: 0.0018\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 75/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.8771e-04 - val_loss: 0.0016\n",
+      "Epoch 76/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.7320e-04 - val_loss: 0.0014\n",
+      "Epoch 77/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.6670e-04 - val_loss: 0.0014\n",
+      "Epoch 78/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.5454e-04 - val_loss: 0.0015\n",
+      "Epoch 79/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.3544e-04 - val_loss: 0.0018\n",
+      "Epoch 80/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.2066e-04 - val_loss: 0.0020\n",
+      "Epoch 81/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.1518e-04 - val_loss: 0.0021\n",
+      "Epoch 82/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.1124e-04 - val_loss: 0.0021\n",
+      "Epoch 83/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.0217e-04 - val_loss: 0.0019\n",
+      "Epoch 84/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 9.2181e-05 - val_loss: 0.0018\n",
+      "Epoch 85/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 8.7719e-05 - val_loss: 0.0016\n",
+      "Epoch 86/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 8.6810e-05 - val_loss: 0.0016\n",
+      "Epoch 87/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 8.3207e-05 - val_loss: 0.0017\n",
+      "Epoch 88/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 7.6554e-05 - val_loss: 0.0018\n",
+      "Epoch 89/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 7.1675e-05 - val_loss: 0.0019\n",
+      "Epoch 90/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 6.9901e-05 - val_loss: 0.0019\n",
+      "Epoch 91/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 6.7437e-05 - val_loss: 0.0018\n",
+      "Epoch 92/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 6.2241e-05 - val_loss: 0.0016\n",
+      "Epoch 93/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 5.7004e-05 - val_loss: 0.0015\n",
+      "Epoch 94/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 5.4077e-05 - val_loss: 0.0014\n",
+      "Epoch 95/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 5.1811e-05 - val_loss: 0.0013\n",
+      "Epoch 96/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 4.8036e-05 - val_loss: 0.0013\n",
+      "Epoch 97/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 4.3778e-05 - val_loss: 0.0014\n",
+      "Epoch 98/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 4.1111e-05 - val_loss: 0.0014\n",
+      "Epoch 99/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 3.9607e-05 - val_loss: 0.0013\n",
+      "Epoch 100/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 3.7507e-05 - val_loss: 0.0012\n",
+      "Epoch 101/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 3.4860e-05 - val_loss: 0.0011\n",
+      "Epoch 102/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 3.3098e-05 - val_loss: 0.0010\n",
+      "Epoch 103/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 3.2335e-05 - val_loss: 9.6503e-04\n",
+      "Epoch 104/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 3.1340e-05 - val_loss: 9.6133e-04\n",
+      "Epoch 105/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 2.9776e-05 - val_loss: 9.7994e-04\n",
+      "Epoch 106/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 2.8548e-05 - val_loss: 9.9455e-04\n",
+      "Epoch 107/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 2.7980e-05 - val_loss: 9.8464e-04\n",
+      "Epoch 108/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 2.7312e-05 - val_loss: 9.4668e-04\n",
+      "Epoch 109/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 2.6160e-05 - val_loss: 8.9482e-04\n",
+      "Epoch 110/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 2.5098e-05 - val_loss: 8.5035e-04\n",
+      "Epoch 111/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 2.4483e-05 - val_loss: 8.2914e-04\n",
+      "Epoch 112/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 2.3881e-05 - val_loss: 8.3457e-04\n",
+      "Epoch 113/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 2.2974e-05 - val_loss: 8.5739e-04\n",
+      "Epoch 114/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 2.2113e-05 - val_loss: 8.8064e-04\n",
+      "Epoch 115/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 2.1598e-05 - val_loss: 8.8882e-04\n",
+      "Epoch 116/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 2.1167e-05 - val_loss: 8.7642e-04\n",
+      "Epoch 117/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.0569e-05 - val_loss: 8.5045e-04\n",
+      "Epoch 118/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.0009e-05 - val_loss: 8.2489e-04\n",
+      "Epoch 119/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.9701e-05 - val_loss: 8.1172e-04\n",
+      "Epoch 120/150\n",
+      "1/1 [==============================] - 0s 29ms/step - loss: 1.9465e-05 - val_loss: 8.1485e-04\n",
+      "Epoch 121/150\n",
+      "1/1 [==============================] - 0s 26ms/step - loss: 1.9117e-05 - val_loss: 8.2877e-04\n",
+      "Epoch 122/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.8777e-05 - val_loss: 8.4213e-04\n",
+      "Epoch 123/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.8591e-05 - val_loss: 8.4424e-04\n",
+      "Epoch 124/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.8437e-05 - val_loss: 8.3135e-04\n",
+      "Epoch 125/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.8189e-05 - val_loss: 8.0840e-04\n",
+      "Epoch 126/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.7938e-05 - val_loss: 7.8519e-04\n",
+      "Epoch 127/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.7779e-05 - val_loss: 7.7014e-04\n",
+      "Epoch 128/150\n",
+      "1/1 [==============================] - 0s 28ms/step - loss: 1.7631e-05 - val_loss: 7.6589e-04\n",
+      "Epoch 129/150\n",
+      "1/1 [==============================] - 0s 31ms/step - loss: 1.7418e-05 - val_loss: 7.6843e-04\n",
+      "Epoch 130/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.7210e-05 - val_loss: 7.7016e-04\n",
+      "Epoch 131/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.7070e-05 - val_loss: 7.6447e-04\n",
+      "Epoch 132/150\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 1.6939e-05 - val_loss: 7.4969e-04\n",
+      "Epoch 133/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.6773e-05 - val_loss: 7.2990e-04\n",
+      "Epoch 134/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.6620e-05 - val_loss: 7.1173e-04\n",
+      "Epoch 135/150\n",
+      "1/1 [==============================] - 0s 30ms/step - loss: 1.6519e-05 - val_loss: 7.0041e-04\n",
+      "Epoch 136/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.6421e-05 - val_loss: 6.9684e-04\n",
+      "Epoch 137/150\n",
+      "1/1 [==============================] - 0s 33ms/step - loss: 1.6302e-05 - val_loss: 6.9765e-04\n",
+      "Epoch 138/150\n",
+      "1/1 [==============================] - 0s 27ms/step - loss: 1.6198e-05 - val_loss: 6.9758e-04\n",
+      "Epoch 139/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.6123e-05 - val_loss: 6.9267e-04\n",
+      "Epoch 140/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.6044e-05 - val_loss: 6.8272e-04\n",
+      "Epoch 141/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.5945e-05 - val_loss: 6.7100e-04\n",
+      "Epoch 142/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.5859e-05 - val_loss: 6.6190e-04\n",
+      "Epoch 143/150\n",
+      "1/1 [==============================] - 0s 46ms/step - loss: 1.5787e-05 - val_loss: 6.5813e-04\n",
+      "Epoch 144/150\n",
+      "1/1 [==============================] - 0s 19ms/step - loss: 1.5708e-05 - val_loss: 6.5924e-04\n",
+      "Epoch 145/150\n",
+      "1/1 [==============================] - 0s 32ms/step - loss: 1.5616e-05 - val_loss: 6.6222e-04\n",
+      "Epoch 146/150\n",
+      "1/1 [==============================] - 0s 20ms/step - loss: 1.5536e-05 - val_loss: 6.6341e-04\n",
+      "Epoch 147/150\n",
+      "1/1 [==============================] - 0s 24ms/step - loss: 1.5465e-05 - val_loss: 6.6077e-04\n",
+      "Epoch 148/150\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 1.5385e-05 - val_loss: 6.5512e-04\n",
+      "Epoch 149/150\n",
+      "1/1 [==============================] - 0s 21ms/step - loss: 1.5303e-05 - val_loss: 6.4914e-04\n",
+      "Epoch 150/150\n",
+      "1/1 [==============================] - 0s 23ms/step - loss: 1.5230e-05 - val_loss: 6.4560e-04\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZxcVZn/8c9T1V29d/Y9IQkQCCGshs0lgAubLCoMhE1lUAYZBRlhgB8DIjrjjM7ouKARFRAJm2wGiRAVZBlZkmBICEsIIUsngXT2pffu5/fHuZVUV1d3Vydd6e7K9/161avqnnvvqae2+9Q55y7m7oiIiKSL9XQAIiLSOylBiIhIRkoQIiKSkRKEiIhkpAQhIiIZKUGIiEhGShCdMLNbzOyebqqrxMweN7PNZvY7M7vQzGZ3U93LzOyT3VFXWr1uZvtHj6eb2U3ZLLsLz9Nt70VavSeYWVV31yuyN9jrE4SZbUu5tZhZbcr0hd38dOcAw4BB7v4P7j7D3U/q5ufIGXe/3N2/vbv1mNm4KJkUpNTdp96LbEWv9RkzqzGztzpK4hb8l5mtj27fMzPLti4zu8DMlpvZdjN7zMwGps3/pJm9Gs1faWbnZojhC9Fn86WUsi+aWXPab+WEtPWmmdmbUd3vmtnHUuZ9Ioq3Jop/bMq8E6OyzWa2LEM8z5hZtZltMbPXzOysbF+zmS1Ki7nJzB5PmX+4mc2L4ppnZoe387k8nf59NbOvmtlcM6s3s7syrPMlM1sSPe+TZjYywzKJ6H2pSitv9zWb2Qgzm2lmq6OYxmWKubvs9QnC3cuTN2AFcEZK2YxufrqxwGJ3b+rmeqX3ug/4OzAIuBF4yMyGtLPsZcBngMOAQ4HTgX/Kpi4zOxj4BXAx4U9IDfCz5IpmNgm4N1qvH3A4MC/1yc1sAHADsChDbC+m/lbc/a8p630K+C/gEqACmAosjeYNBh4BbgIGAnOBB1Lq3Q7cAVzbzntyFTDC3Suj9+ceMxuRzWt294NTftsVhN/376J1E8DvgXuAAcBvgN9H5anvyYVAAW2tBr4Txd6KmR0P/AdwVvSa3yN8dumuBdZ25TUDLcCTwNkZ1ut+7q5bdAOWAZ9MK7sFeBC4G9hK+PFMSZk/EngYqCZ8Ea5sp+5vAQ1AI7ANuBT4IvBCyjIOXA68A2wEbgMsmrcf8DSwHlgHzAD6dxR7VH4s8D4QTyn7LLAgenw08CKwCVgD/BRIpMW0f/T4LuA7KfOujdZZDfxj2rKfJmzMtgArgVtS1lsRLbstuh2X4b34MDAH2Bzdfzhl3l+BbwP/F30ms4HB7bzvJwBVKdMHRetvij7LM1PmnQa8EdW5CrgmKh8M/CFaZwPwPBDL4vt0AFAPVKSUPQ9c3s7yfwMuS5m+FHgpm7oIG6R7U+btR/i+VUTT9wLf7iTe6cAV0fvzpZTyVp9NO3Ff2s68y4C/pUyXAbXAxLTlPgks6yS+o4E64OhsXnPausdH37WyaPqk6DO2tO/lKSnT/YDFhN+QAwUZ6v0OcFda2X8Dt6VtIxzYL6VsPPAmcGrq97Oz15xSXhDVOa6z7+Hu3Pb6FkSWzgTuB/oDMwkbUcwsBjwOvAaMAj4BfN3MTk6vwN2/SfhCP+DhX82v23mu04GjCP8izwWSdRnwXcKX7SBgDCF5dcjdXyL8S/t4SvEFhA0GQDNwNWEjeFz0Gq7orF4zOwW4BvgUMIHwA0+1Hfg84T37NPAVM/tMNG9qdN8/ei9eTKt7IPAE8GPCv+UfAE+Y2aC013AJMBRIRLF0FnMh4fOaHa33NWCGmR0YLfJr4J/cvQKYTEjIAN8AqoAhhH+q/4/w48TMfmZmO/61pjkYWOruW1PKXovK21v+tXaW7ayuVuu6+7uEjeUBUdGxUbwLzWyNmd2T1h1zNDCFkCQyOcLM1pnZYjO7KdndYmbxaL0hUZdKlZn91MxK2olrO/BuB+9BG2b2BzOrA14mJK+5Wb7mVF8AHoqeP7nuAo+2tpEFaXH9B/Bzwh+srrDoljoN4TuV9BPC96g2YwXtv+Y9SgkiOy+4+yx3bwZ+S9h4Q9iQD3H3W929wd2XAr8Epu3Gc/2nu29y9xXAM4SuANx9ibv/yd3r3b2asNE8Pss67wPOBzCzCsI/5fuieue5+0vu3uTuywhN9mzqPRe4091fj350t6TOdPe/uvtCd29x9wXR82Ub76eBd9z9t1Fc9wFvAWekLHOnuy9291pCCy9j/3GaY4Fywnvc4O5PE1oG50fzG4FJZlbp7hvd/dWU8hHAWHdvdPfnkxsWd7/C3dtLqOWEFlCqzYTujmyW3wyUm5llUVdn80cTumLOJiT0EsJGKrmR/xnwNXdvyRDXc4SN29Bo/fPZ2SU0DCgkjK99jPA5HAH8W5ZxdcrdT4+WPw14KiXGrOo2s9IovrtSijtc18ymAB8heo+6aBZwrpkdGiXKmwl/KEqjuj9LaI082l4FHbzmPUoJIjup/yBqgOLoH9RYYKSZbUreCP8KhnXjc5UDmNlQM7vfzFaZ2RZC3+ngLOu8F/icmRUBnwNedfflUb0HRP9W3o/q/Y8s6x1J6DpKWp4608yOSRls20zoOss23pHp9UXTo1KmM75P2cSc9mNLrfdswg9yuZk9a2bHReXfB5YAs81sqZldn93LYBtQmVZWSejCymb5SmBblIw6q6uz+bXsTKrbCJ/zadG8Kwj/pl8kA3df6u7vRcl+IXArYYObrBfgJ+6+xt3XEf68JOvu6nuQUZSY/wicbGZndrHuzxG6Bp9NKWt33ahn4GfAVb4L44Xu/hfgm4Su5+WE7t+tQJWZlQHfI7ReO6sn02veo5Qgds9K4D13759yq3D30zpds+u+S/gXcqiHwauLaN2MbZe7v0H4op5K6+4lCE3ot4AJUb3/L8t61xC6uZL2SZt/L6E7boy79yN0XSTr7ewUwqsJyTfVPoQ+492xGhgTbQDa1Ovuc9z9LMI/5ccILRPcfau7f8Pd9yW0Yv7FzD6RxfMtAvaNWm1Jh5F5EDi5/GHtLNtZXa3WNbN9gSJCHzqE7pP23vdPAJ+N/iS8Txj/+R8z+2k7yzvRZ+nuGwndb+3VnR5XGWGsoL33oDMF0fqZ6k5/zUlfAO5O605aBBwatc6SDo3KKwndZg9E78ecaH6Vpeyd1RF3v83dJ7j7UEKiKABeJ7TexgHPR3U/AoyI3vtxWbzmPUoJYve8Amwxs+ssHOMQN7PJZnZUDp6rgvCvZ5OZjaL9vT7acy9wJaH//3dp9W4BtpnZROArWdb3IPBFM5sUNeG/mSHeDe5eF/VvX5Ayr5qwN8a+7dQ9CzjAwi6MBWZ2HjCJ0B20O14mjI38q5kVWthV8wzg/miXwwvNrJ+7NxLek2YAMzvdzPaPNibJ8ubOnszdFwPzgW+aWXHUtXAoYYORyd2E5DPKwm6R3yDqFsmirhnAGWb2sWgjfCvwSMqYxZ3AJWa2b/R5XcfO9/OLhHGtw6PbXMJOFTdGr/9UMxsWPZ5I2CPp9ylx3wl8LWrlDgC+nlL3o8BkMzvbzIoJ3S0L3P2tqL5YVF4YJq3Yoj2JzGxi9Nwl0ed1EeH7m2wJdPaaMbPRwImEvZRS/ZXwGV5pZkVm9tWo/GlCV9PIlPcj+YfvQ4TvENH3shiIA/Eo7uS4THG0HTAz2we4HfhRlExfJ/yxStb9JeCD6PHKLF4z0fMWRZNF0XRudNdodz7caH8vpntSpseRskcD4Yt0H6HLYyPwUnodHdT1RdruxbR/yvRdRHsNEQbP5hGSxHyigdOOYk977n0IG+Un0sqnEloQ2wh7xdzaXky03Yvp+uh1Z9qL6RxCq2UrYWPx07TXfishUWwijA2kvxcfjV7v5uj+oynz/kqWe9nQdi+mgwk/ts2EPZY+G5UnCLsPbiQkgTnJ5yQM4i8jJJcq4KaU+qYD0zt438dF8dYCb6d+RoQ++20p00boftgQ3b5H671s2q0rmn8BYU+c7YQN+MC0+d+K3vNqwljagHZiTn9//5uwEdtO2H31VqAwZX4hoUtmU/R9+DFQnDL/k4TvWG1U97i0z8fTbn+N5h1E2CBvjeqek/y8uvCabwCeb+d1HhF9t2qBV4EjOvgMW+3FRPgtp8d9SzSvP6HFtj16P75Lyl6EnXw/s3nN6c/rXd3WZXtL7kIpIiLSirqYREQkIyUIERHJSAlCREQyUoIQEZGMMp2Eqs8aPHiwjxs3rqfDEBHpM+bNm7fO3TOeQDKvEsS4ceOYO7dHTlkiItInmVn6WQt2UBeTiIhkpAQhIiIZKUGIiEhGeTUGISJ7n8bGRqqqqqirq+vpUHq14uJiRo8eTWFhYdbrKEGISJ9WVVVFRUUF48aNo/XJWSXJ3Vm/fj1VVVWMHz8+6/XUxSQifVpdXR2DBg1ScuiAmTFo0KAut7KUIESkz1Ny6NyuvEdKEADPfg+W/LmnoxAR6VWUIABe+CG8+0xPRyEifVR5eTZXvO17lCAAYoXQ0ukFwkRE9ipKEACxOLR0+drkIiKtuDvXXnstkydP5pBDDuGBBx4AYM2aNUydOpXDDz+cyZMn8/zzz9Pc3MwXv/jFHcv+8Ic/7OHo29JurgCxAiUIkTzwrccX8cbqLd1a56SRlXzzjIOzWvaRRx5h/vz5vPbaa6xbt46jjjqKqVOncu+993LyySdz44030tzcTE1NDfPnz2fVqlW8/vrrAGzatKlb4+4OakGAEoSIdIsXXniB888/n3g8zrBhwzj++OOZM2cORx11FHfeeSe33HILCxcupKKign333ZelS5fyta99jSeffJLKysqeDr8NtSAgShAagxDp67L9p58r7p6xfOrUqTz33HM88cQTXHzxxVx77bV8/vOf57XXXuOpp57itttu48EHH+SOO+7YwxF3TC0I0BiEiHSLqVOn8sADD9Dc3Ex1dTXPPfccRx99NMuXL2fo0KF8+ctf5tJLL+XVV19l3bp1tLS0cPbZZ/Ptb3+bV199tafDb0MtCFAXk4h0i89+9rO8+OKLHHbYYZgZ3/ve9xg+fDi/+c1v+P73v09hYSHl5eXcfffdrFq1iksuuYSWlhYAvvvd7/Zw9G1Ze02ivmjKlCm+SxcMuu1YGDwBzvtt9wclIjn15ptvctBBB/V0GH1CpvfKzOa5+5RMy6uLCTQGISKSgRIEaAxCRCQDJQjQGISISAY5TRBmdoqZvW1mS8zs+gzzLzSzBdHtb2Z2WMq8ZWa20Mzmm9kuDCx0gRKEiEgbOduLycziwG3Ap4AqYI6ZzXT3N1IWew843t03mtmpwO3AMSnzT3T3dbmKcQeNQYiItJHLFsTRwBJ3X+ruDcD9wFmpC7j739x9YzT5EjA6h/G0T2MQIiJt5DJBjAJWpkxXRWXtuRT4Y8q0A7PNbJ6ZXdbeSmZ2mZnNNbO51dXVuxapuphERNrIZYLIdPmijAddmNmJhARxXUrxR9z9SOBU4J/NbGqmdd39dnef4u5ThgwZsmuRKkGIyB7S0bUjli1bxuTJk/dgNB3LZYKoAsakTI8GVqcvZGaHAr8CznL39clyd18d3a8FHiV0WeWGxiBERNrI5ak25gATzGw8sAqYBlyQuoCZ7QM8Alzs7otTysuAmLtvjR6fBNyas0g1BiGSH/54Pby/sHvrHH4InPqf7c6+7rrrGDt2LFdccQUAt9xyC2bGc889x8aNG2lsbOQ73/kOZ511Vrt1ZFJXV8dXvvIV5s6dS0FBAT/4wQ848cQTWbRoEZdccgkNDQ20tLTw8MMPM3LkSM4991yqqqpobm7mpptu4rzzztutlw05TBDu3mRmXwWeAuLAHe6+yMwuj+ZPB24GBgE/iy6o3RQd8j0MeDQqKwDudfcncxWruphEZFdNmzaNr3/96zsSxIMPPsiTTz7J1VdfTWVlJevWrePYY4/lzDPPJNqmZeW2224DYOHChbz11lucdNJJLF68mOnTp3PVVVdx4YUX0tDQQHNzM7NmzWLkyJE88cQTAGzevLlbXltOT9bn7rOAWWll01Mefwn4Uob1lgKHpZfnjBKESH7o4J9+rhxxxBGsXbuW1atXU11dzYABAxgxYgRXX301zz33HLFYjFWrVvHBBx8wfPjwrOt94YUX+NrXvgbAxIkTGTt2LIsXL+a4447j3//936mqquJzn/scEyZM4JBDDuGaa67huuuu4/TTT+djH/tYt7w2HUkNGoMQkd1yzjnn8NBDD/HAAw8wbdo0ZsyYQXV1NfPmzWP+/PkMGzaMurq6LtXZ3olUL7jgAmbOnElJSQknn3wyTz/9NAcccADz5s3jkEMO4YYbbuDWW7unR16n+waNQYjIbpk2bRpf/vKXWbduHc8++ywPPvggQ4cOpbCwkGeeeYbly5d3uc6pU6cyY8YMPv7xj7N48WJWrFjBgQceyNKlS9l333258sorWbp0KQsWLGDixIkMHDiQiy66iPLycu66665ueV1KEKAuJhHZLQcffDBbt25l1KhRjBgxggsvvJAzzjiDKVOmcPjhhzNx4sQu13nFFVdw+eWXc8ghh1BQUMBdd91FUVERDzzwAPfccw+FhYUMHz6cm2++mTlz5nDttdcSi8UoLCzk5z//ebe8Ll0PAuCJa+D1h+G697o/KBHJKV0PInu6HsSu0BiEiEgb6mICjUGIyB61cOFCLr744lZlRUVFvPzyyz0UUWZKEKAxCJE+zt27dIxBTzvkkEOYP3/+Hn3OXRlOUBcTKEGI9GHFxcWsX79+lzaAewt3Z/369RQXF3dpPbUgICQIbwZ36EP/QkQERo8eTVVVFbt8Nue9RHFxMaNHd+2KCkoQEBIEhFZEvLBnYxGRLiksLGT8+PE9HUZeUhcThEFqUDeTiEgKJQho3YIQERFACSJQghARaUMJAlIShA6WExFJUoIAiKsFISKSTgkC1MUkIpKBEgQoQYiIZKAEARqDEBHJQAkCdByEiEgGShCgLiYRkQyUIEAJQkQkAyUIUIIQEclACQJSxiA0SC0ikqQEAWpBiIhkoAQBShAiIhkoQYAShIhIBkoQoAPlREQyUIIAHSgnIpJBThOEmZ1iZm+b2RIzuz7D/AvNbEF0+5uZHZbtut1KXUwiIm3kLEGYWRy4DTgVmAScb2aT0hZ7Dzje3Q8Fvg3c3oV1u48ShIhIG7lsQRwNLHH3pe7eANwPnJW6gLv/zd03RpMvAaOzXbdbaQxCRKSNXCaIUcDKlOmqqKw9lwJ/7Oq6ZnaZmc01s7nV1dW7FqnGIERE2shlgrAMZZ5xQbMTCQniuq6u6+63u/sUd58yZMiQXQpUXUwiIm0V5LDuKmBMyvRoYHX6QmZ2KPAr4FR3X9+VdbuNEoSISBu5bEHMASaY2XgzSwDTgJmpC5jZPsAjwMXuvrgr63YrJQgRkTZy1oJw9yYz+yrwFBAH7nD3RWZ2eTR/OnAzMAj4mZkBNEXdRRnXzVWsGqQWEWkrl11MuPssYFZa2fSUx18CvpTtujmjQWoRkTZ0JDWoi0lEJAMlCFCCEBHJQAkCNAYhIpKBEgSARW+DWhAiIjsoQQCYhVaEEoSIyA5KEEmxAmhu7No6TQ3QWJebeEREepgSRFKsoOtjEI9fBXfn7hyCIiI9KafHQfQpsXjXupgatsOiRyGeyF1MIiI9SC2IpK6OQbwzG5pqoX5zSBYiInlGCSKpqwli0WM7H299v/vjERHpYUoQSV0Zg2ioCS2IgfuFaSUIEclDShBJXRmDeGc2NNbAMf8UpreuyV1cIiI9RAkiKVaYfYJ4exaUDobJZ4dptSBEJA8pQSR1ZQxi00oYMhFKB0FBsVoQIpKXlCCSupIgajdCSf9wBHbFCLUgRCQvKUEkxeLZD1LXbYKSAeGxEoSI5CkliKQutyCSCWK4uphEJC8pQSRlmyAaa6Gprm0Lwj238YmI7GFKEEnZJojajeG+pH+4rxgOjduhfmvuYhMR6QFKEEnZjkHsSBApLQhQN5OI5B0liKQutyBSxiBACUJE8o4SRFLWCWJTuG/TgtCeTCKSX5Qgkna5BTEs3KsFISJ5RgkiaVfHIIoqIFGhFoSI5B0liKSutCBiBZAo31mmYyFEJA8pQSR1JUEUR6fZSKrU0dQikn+UIJK6kiCS3UtJ5WpBiEj+yWmCMLNTzOxtM1tiZtdnmD/RzF40s3ozuyZt3jIzW2hm881sbi7jBLK/YFDqeZiSSgZA3ebcxCUi0kMKclWxmcWB24BPAVXAHDOb6e5vpCy2AbgS+Ew71Zzo7utyFWMr2V4wqHZjaDGkKq4MR1K7t+56EhHpw3LZgjgaWOLuS929AbgfOCt1AXdf6+5zgMYcxpGd3eliKqoAb4GGbbmJTUSkB+QyQYwCVqZMV0Vl2XJgtpnNM7PL2lvIzC4zs7lmNre6unoXQ6VrB8q1SRCV4b5uy64/v4hIL5PLBJGpr6Urpzz9iLsfCZwK/LOZTc20kLvf7u5T3H3KkCFDdiXOIJsxiOZGqN/SNkEURwmiXglCRPJHLhNEFTAmZXo0sDrbld19dXS/FniU0GWVO9mMQSQHotu0IPpF85UgRCR/5DJBzAEmmNl4M0sA04CZ2axoZmVmVpF8DJwEvJ6zSCG7LqYd52Hq37pcLQgRyUM524vJ3ZvM7KvAU0AcuMPdF5nZ5dH86WY2HJgLVAItZvZ1YBIwGHjUwh5BBcC97v5krmIFskwQaafZSCpSghCR/JNVgjCzq4A7ga3Ar4AjgOvdfXZH67n7LGBWWtn0lMfvE7qe0m0BDssmtm4TKwAcWlog1k7Dqr0EUaxBahHJP9l2Mf2ju28hdPUMAS4B/jNnUfWEWDzcd9SKaLcFURHu1YIQkTySbYJI7pF0GnCnu79G5r2U+q5Y1JjalQSRKAeLqQUhInkl2wQxz8xmExLEU9EAckvuwuoBOxJEB8fsJRNEcb/W5WahFaEWhIjkkWwHqS8FDgeWunuNmQ0kdDPlhT8sWM2RWxsZCR0fC1G3KSSHZHdUqqJ+akGISF7JtgVxHPC2u28ys4uAfwPy5ux01/5uAQtWR6fJ6KyLqbh/5nnFlWpBiEheyTZB/ByoMbPDgH8FlgN35yyqPaysKE5dczSk0lmCSB9/SCqKTtgnIpInsk0QTe7uhJPt/cjdfwRU5C6sPaskkW2C2NT2ILmk4kqd8ltE8kq2CWKrmd0AXAw8EZ3KuzB3Ye1ZpYUF1GaTIOq37DwoLl2RuphEJL9kmyDOA+oJx0O8Tzgr6/dzFtUeVpKIU9uUTBAdDVJv2XlQXLqiCg1Si0heySpBRElhBtDPzE4H6tw9r8Ygsm9B9Ms8LzlI7V05Ya2ISO+VVYIws3OBV4B/AM4FXjazc3IZ2J5UUlhATVMnCaKlOVwQqN0WRGVYt7E2N0GKiOxh2R4HcSNwVHTqbcxsCPBn4KFcBbYnlSbi1CTzQnsJIjm+0N4YROoZXROl3RqfiEhPyHYMIpZMDpH1XVi31wsJopMxiOT4QrstCF0TQkTyS7YtiCfN7Cngvmj6PNLO0tqXlSYK2NBk4exSu92C0LEQIpIfskoQ7n6tmZ0NfISwGb3d3R/NaWR7UGkizrZGhwTtJ4hOWxDJBKFjIUQkP2R9wSB3fxh4OIex9JiSRJxG7+R039m2INTFJCJ5osMEYWZbgUz7bRrg7t7O1rJvKUvEaU4OqbTbgohaBulnck3SNSFEJM90mCDcPW9Op9GR0kQBTSRbEJ0MUnd0JHXqciIifVze7Im0O0oScZrprIsp2YLo4EhqUAtCRPKGEgRhkLqp0y6mLRAvgoKizPNjcUjodBsikj+UIEjvYupgkLq91kOSrgkhInlECYLQgtg5SN3BGER74w9JOqOriOQRJQiSXUzd1IJQF5OI5AklCKJB6s6Og1ALQkT2MkoQQFmioPNB6mxaELomhIjkESUI0ndz7WAMor2D5JI0SC0ieUQJAigqiNFiWbQg2rtY0I6KNAYhIvlDCQIwMxKJ6PiGTAmis4sFJRVXQnM9NNV3f5AiIntYThOEmZ1iZm+b2RIzuz7D/Ilm9qKZ1ZvZNV1Zt7sVFRaGB5kSRGcn6ttRia4JISL5I2cJwsziwG3AqcAk4Hwzm5S22AbgSuC/d2HdbpVIJMKDTAmis1N9J6VeVU5EpI/LZQviaGCJuy919wbgfuCs1AXcfa27zwEau7pudyvakSAyDFJn3YJQghCR/JHLBDEKWJkyXRWVdeu6ZnaZmc01s7nV1dW7FChAaVEhLcS6pwWhLiYRyQO5TBCWoSzTtSV2a113v93dp7j7lCFDhmQdXLqS5Ok2dmsMQmd0FZH8kcsEUQWMSZkeDazeA+vuktLksRAdtiCy2M01dXkRkT4slwliDjDBzMabWQKYBszcA+vukh1ndN2dMYhkAlELQkTyQNbXpO4qd28ys68CTwFx4A53X2Rml0fzp5vZcGAuUAm0mNnXgUnuviXTurmKFVKuCdGcPl4O1G0K99mcagPUghCRvJCzBAHg7rOAWWll01Mev0/oPspq3VwqTcSp9SL6N9a2ndnZxYKS4oVQWKoWhIjkBR1JHSlJFLDZS/FkayFVNifqS9IZXUUkTyhBREoTcbZQSktthgSRzam+k3RNCBHJE0oQkbJEnC1eSkvt5rYz1YIQkb2QEkSkJFHAFsqwugwJoistCF0TQkTyhBJEpDRqQVh9hgRRsx5KB2VXka4JISJ5QgkikhyDiDVshZaW1jO3r4OyLI/S1jUhRCRPKEFEShMFbPEyzFvCtR+SmuqhfnP2CaK4n1oQIpIXlCAiyRYEAKnjENvXhfvyLrQgGmuguZ0r04mI9BFKEJGSRJzNXhYmWiWIteE+6xaETvktIvlBCSJSFu3FBEDdJpqaW5i1cA1VVSuiBbrQggAlCBHp83J6qo2+pCTaiwng74uX8S8PNfHeuu3cOGohXwYoG5xdRbomhIjkCbUgIqljEDOeXUgiHmPyqEoaNn8QFigbml1FuiaEiOQJJYhIYTxGbawcgP0qm3nkig/ziYnDKKxbjxeUQKIsu4p0TQgRyRPqYkqVqACHCw7tR1lRAfsOKaPJNtNUPIhCy3SRuwx0TQgRyRNqQaT4ztmH06uMed8AABeGSURBVFxYTj+rAWC/IeUMZgvbCwdkX4laECKSJ5QgUpwyeQTxkv47dnMdP7iMQbaFjdY/+0q0m6uI5AkliHTF/XYkiLKiAobGtrC2pSL79QuKwsWFlCBEpI9TgkiXkiBwZyBbWFmf5QD1jjp0PiYR6fuUINKV9N95Deq6TRTQxNLaEtw9+zp0TQgRyQNKEOmK+0HyokHReZiqGipYv70h+zp0TQgRyQNKEOlSu5i2VwOwnkqWVm/vQh1qQYhI36cEkS55uu6Wlp0JwvuxtHpbJyum0DUhRCQPKEGkK+4HeEgS28KZXLfE+7N0XVdaELomhIj0fUoQ6ZJHQtdtjsYgjH4Dh6sFISJ7HSWIdK0SRDWUDmTM4AqWr6/Jvo6SAdCwFZobcxOjiMgeoASRLj1BlA1hn4GlrNhQk/2urqUDw33NhtzEKCKyByhBpEvvYiobwthBpdQ3tbB2a312dSSvHVGzLjcxiojsATlNEGZ2ipm9bWZLzOz6DPPNzH4czV9gZkemzFtmZgvNbL6Zzc1lnK0UR+dd2rQCPngd+o9lzMBwnYgVG7LsZiqNEsR2JQgR6btyliDMLA7cBpwKTALON7NJaYudCkyIbpcBP0+bf6K7H+7uU3IVZxvJFsTffhL2RDrmMsYOCqfayHocQi0IEckDuWxBHA0scfel7t4A3A+clbbMWcDdHrwE9DezETmMqXNFlYDB1tVw4Gkw4jBG9S8hZrvSglifszBFRHItlwliFLAyZboqKst2GQdmm9k8M7usvScxs8vMbK6Zza2urt79qGOxndd0OP5fAUgUxBjRr4QV67M8FqJ0IGBqQYhIn5bLK8plugRb+m5AHS3zEXdfbWZDgT+Z2Vvu/lybhd1vB24HmDJlShfOqNeBfqNhwEdg5BE7ivYZWMrybFsQsXjY1bVGLQgR6btymSCqgDEp06OB1dku4+7J+7Vm9iihy6pNgsiJz/++zTWoxw4q5c9vfpB9HaWDNEgtIn1aLruY5gATzGy8mSWAacDMtGVmAp+P9mY6Ftjs7mvMrMzMKgDMrAw4CXg9h7G2Vj4EEqWtisYMLGXdtga21TdlV0fZYLUgRKRPy1kLwt2bzOyrwFNAHLjD3ReZ2eXR/OnALOA0YAlQA1wSrT4MeNTMkjHe6+5P5irWbIwdFBLGyg01HDSisvMVSgfB+ndzHJWISO7ksosJd59FSAKpZdNTHjvwzxnWWwoclsvYumqf6FiI5euzTBBlg2HlyzmOSkQkd3QkdZbGDgxjEiu7sqtrzYZw2nARkT5ICSJL/UoLqSwuYPmGLHd1LRsM3rzz8qUiIn2MEkQXjB1UxrJ12bYgBoV7DVSLSB+lBNEFh4/pz6srNlLf1Nz5wskEoV1dRaSPUoLoghMOHEJNQzNzl23sfGGdj0lE+jgliC44dt9BJOIxnl2cxSk9dEZXEenjlCC6oKyogKPGD+Cvb6/NYmG1IESkb1OC6KITDhjK4g+2sXpTbccLFhRBokJXlRORPksJoouOP3AIAM9F3UwtLc6yddszJ4wynY9JRPqunB5JnY8mDC1nRL9ifvCnxfzmxeVUbahha30TA0oLefZfT6SyuHDnwqWD1MUkIn2WWhBdZGZcceL+jB1Uyqj+JXzmiFFcd8pENtY08svnlrZeuHSwWhAi0mepBbELLj52LBcfO7ZV2aLVm/nV8+/x+ePGMaSiKBSWDQ7XtRYR6YPUgugm3zjpQBqaW/jJ0+/sLExeE8K75zpGIiJ7khJENxk/uIxzjhzN/XNWsj15zYjKUdBcD9u74VKoIiJ7mBJENzrriJE0NLXw/DtRQhhyQLivfrvnghIR2UVKEN3o6HED6VdSyOw3okuTDpkY7qvf6rmgRER2kRJENyqIx/j4xKE8/dZamppboGIEFFWqBSEifZISRDc7adIwNtU0MmfZRjCDwQfAOiUIEel7lCC62dQDhpAoiPGn1G4mtSBEpA9SguhmZUUFfHT/wcx+433cHYYcCNs+gNosThGernYjvP4I1G/r/kBFRDqhA+Vy4LRDRvD0W2uZu3wjRw05MBRWL4Z9jsmugobtMOtaWPhQ2E12/0/B+fdDvI98XCtehrceh5WvwNiPwIe+AAPG9XRUItJFakHkwGmHDKcsEefBOStDCwK6tifTs/8F82fAERfBiTfCkj/Bk9f3/gPu3OHp78AdJ8FL06GpHv7vf+FHh8Ofb4GWLK7EJyK9Rh/5S9q3lCYKOOOwkcx8bTXfPOMgygtKsh+HWPsWvHgbHH4RnP6DUFa/Bf72ExhzDBz6D7kLfHe4w+x/gxd/CkdcDKd8F4oqYPMqePY/4YUfwprX4OxfQ+nA7Otc9SosegS2rgn1DT8UDjsfEqW5fT0iohZErvzDlDHUNDTzxML3YfCEVnsybdzewB8XrmHj9obWK7nDrGsgUQ6f+tbO8k/eGjaMz3wHmhv30Cvoohd+GJLD0ZfBGT8OG3OAfqPgzJ+E27IX4PbjYc2Cjutyh7efhJ9/GH71cXjldlj9d3jzD/DEv8D/Tob/+xE0N+X+dYnsxZQgcuTIffqz/9By7ntlJR7tydTS4jwwZwUf/5+/8pUZr3LMf/yFK+/7O5troo3+wodg2fPwiZt3XpEOcDOWTL4KNi7j3T//ktqGXtZVs3g2/OVWmHw2nPo9iGX4Wh35ebjkydDN9OtPwfM/CGMtqdzhnT/DnafCfeeFLqozfwLXvANX/h2uXRLqGHkk/Olm+PUnYe2be+Y1iuyFzHt7v3YXTJkyxefOndvTYezw25eWc9Njr/O9YX/m3M138OXBd/OnqgKOGjeAr5ywH88tXse9L6/goJGV3HPhRCp+dRxUjoQv/QVicQBWbqjh3x57nWcXr+WxxM0Mts18Lv4Tzj9uf847agwj+5cA4O4sWr2Fx/6+igVVm1lSvY1+JYUcM34gJx88nOMPGEIsZq0DbKoPe0ltWh42umOOhpL+XXuRa9+CX58EA/aBf5zdedfPtmqY+VVY/GQ4Hfp+H4cBY2HLaljxEmx4N5zD6qNXw4e+CPHCzPUseiy0Juq3wgnXw4ev6juD+CK9iJnNc/cpGecpQeSOu3PfKyv55cxnmF1wNQ/bJ4md/j+cc+ToHRvrP73xAV+5Zx4/6n8/p9XMxL78NIw6EoCZr63muocWEDO4+lMHcHx8ARNmf4H7B36F61d/DIDDxvSntDDOig01rNpUSyIe49DR/dh/aDnrttXzynsb2FLXxP5Dy7n42LGcdPAwRvQrgb/fE/71b/tgZ8CJCvj4jXDUl7Pb2G5YCnecCjh86c/Qf5/s35yVr8AL/wvvL4DNVWFcYtQUmHQmHHIuFCQ6r2P7OnjiG/DGYzBsckgqkz7Teezu0NwQLgsrspdTguhhC6s2w+NXMnndLOzK+aFfPsVLf3mUo567hN8XnMzwC26jLFHAY/NXcef/LeOocQP40bQjQkvBHWacA8tfZOX5zzBzeZw/v/kBMTNG9i/hmPEDOePQkfQr3fmvu7G5hScWrOH255byxpotgHNT6WNc2vI7FsQm8Qs+xwI7kI+UrOTilsc4uOYVmoZOpuCM/4UxR7X/ota+CTPOhYZtcMksGHrQrr9BzY0QKwhHnu+KRY+GvafWL4GyoWF34qEHQ2Fx6NLa9gFsfb/1fVMdFPeDfvuE5cdPhbEfDZeJFdmL9FiCMLNTgB8BceBX7v6fafMtmn8aUAN80d1fzWbdTHprggBg0wr48RHwoUvg0/+9s/zdp+G+C6gtH83Z9d/kjY07++8vPnYsN50+iURBrHU9Pzsu7NF00cNd2qi+u7qaxplXM/H9x3mx32ncN/RfKCspBpy1W+pZWLWJI2te4JuFdzPCNvDWsE+zeeJ5FO37UZqJ0dTcQmzrKoa/cx+j3/glXlSBXfQINuqIbniDdlNLC7w9K7QmVr4Sus2SivpBxTAoHwYVw8OtqB9sXxuSyoqXoTEaDxk2GcZ9DMZ9NBwF33+fna0Z97BHWc0GqN0QWjDbq8NYSlM9FJZAyYCdt9KBUDIwDNhn+pxamqFuM9RtCgdF1m6E2k3QWAsFxSHBFZZAYWnYcaGoItwKS8FbwJvDfUtzWKepDhprQiwWDy2pWAHEE2GdovJQT9R9KQI9lCDMLA4sBj4FVAFzgPPd/Y2UZU4DvkZIEMcAP3L3Y7JZN5NenSAAHr8KXv0tHHAy7HNc2DPnrSfCXk6f/z1b4/14bP5qhpQnmDSiH/sMaqc//+Xb4Y/XwpR/hOOvCycEXP13qHoFVs0LG5dhB8OoD8Hoo8NG4v0F8PiV8P5COP760G+fttFqaXEWrtrMs68vZeT8H/PpuicosQa2eAmbvJxCa2aEbQDg4eaP8u+NF9FUPJDxQ8oZVlFE/9JCmpqd2sZmahubqWtspraxhfrGZkoSccqLCigvKqCsqIC4GY7T4mG76+440OKOOxTEjOJEnOKCOMWFMQriIUkaIezCeIyighgliTglheFWnIhTFI8RjxkF1kzcmymIxbBEMQWxGDGDWMyImxEzIxaDmBnW0kjig9dIrHyBRNX/kVg1B2uu2/G+eEExHk9gDdsx7/oOAh4rhJKBeEECa2qAlgZoaoDGGoweaMEXlu1MFu2N8VgsSi6FECuM7uMQLwoJM14UvmfJx/ECwFK+U8nH0XSmx22WzfSYjuvocn3p69G6jt2ur73HtFNHyutLj6kr5fEimPBJdkVPJYjjgFvc/eRo+gYAd/9uyjK/AP7q7vdF028DJwDjOls3k16fIGo3wbPfC10iW1eHs73u9wk46dvZHxsA4d/yrGtg3l3hR+st0BLt8jlgfHi8eWWYLigO9011UNwfPvdLOOCkrJ5my5ZNbPz748RWvkhh0zbiODVDD2fTsGNZldiXNZvreG/dNpatq6F6az2bahsojMfCBju5cU/EScRj1Dc1s7WuiW31TWyra8JxDCNm4TrfZtHGOtomNDY79U3N1DW2UNfYTFPLntuQJmhksr3HOHuf0baOUqujiEa2ERLlJi9nE2Ws936so5LtXkwDhRTTQH/bRn+2McC20d+2MYCtDLStDGArhdZMvRfQQCGNFFBDEZu9nE1exiZCvZspo84TFFkjxTRQTAMlVk85dZRRS4XVUmr1ODGaidFCjBaLUecJ6iii3hLUU0icFgpppoAWCq2Jsmj9MqujjDrKqaWMWuLW0uq1Jzc9sR3rN1NAEwU0E6eFBI0kaKSQRhI0kaCBBE0U0BSt61E9zs7NoaeUh+lYTyTGPLYxNoABNy/bpXU7ShC53O1jFLAyZbqK0ErobJlRWa4LgJldBlwGsM8+XRgk7Qkl/eGU/4CTvhP6wSuG71q/eywWDqL78FfhlV+GJDDmaBh91M7dY2s3hq6TZc+Hv+gjDoN9TwhdLVmqrOxP5fEXAxe3Kh8LHNb1qLtF8g9NQ3PLjuRRF7VYahuaqW9qoaXFaWpxmqNb8nGL77w1t4TWSkuL79hUuUNyyv2I8Ch6PgcqHSqBMZ6+zs64UlbB8R2PN7WpP/z4BgED3VPWabucp1TsKfXHU+ovTnteUtbzqP4GoB5YnyG+HctlfC9o9bykrNfRcp4ShGeoP/0Na51I0u69dZJpvVxKMvK2dbRaztuvI7XcdgTYto7kY/OO69i5XOY60mNM1XadpNTynYqLElyVsabdk8sEkWnLl/5utLdMNuuGQvfbgdshtCC6EmCPicWgcsTu1zNw33DEciYlA+DAU8Itj1iUUIsK4hQVxOlX0k4XiYjstlwmiCpgTMr0aGB1lsskslhXRERyKJdHUs8BJpjZeDNLANOAmWnLzAQ+b8GxwGZ3X5PluiIikkM5a0G4e5OZfRV4irCr6h3uvsjMLo/mTwdmEfZgWkLYzfWSjtbNVawiItKWDpQTEdmLdbQXk07WJyIiGSlBiIhIRkoQIiKSkRKEiIhklFeD1GZWDSzvdMHMBgPrujGcXFCMu6+3xweKsbsoxuyMdfchmWbkVYLYHWY2t72R/N5CMe6+3h4fKMbuohh3n7qYREQkIyUIERHJSAlip9t7OoAsKMbd19vjA8XYXRTjbtIYhIiIZKQWhIiIZKQEISIiGe31CcLMTjGzt81siZld39PxAJjZGDN7xszeNLNFZnZVVD7QzP5kZu9E9wN6QaxxM/u7mf2hN8ZoZv3N7CEzeyt6P4/rTTGa2dXRZ/y6md1nZsW9IT4zu8PM1prZ6yll7cZlZjdEv6G3zezkHorv+9HnvMDMHjWz/j0VX3sxpsy7xszczAb3ZIyd2asThJnFgduAU4FJwPlmNqlnowKgCfiGux8EHAv8cxTX9cBf3H0C8JdouqddBbyZMt3bYvwR8KS7TyRcKfVNekmMZjYKuBKY4u6TCae2n9ZL4rsLSL8cYca4ou/mNODgaJ2fRb+tPR3fn4DJ7n4osBi4oQfjay9GzGwM8ClgRUpZT8XYob06QQBHA0vcfam7NwD3A2f1cEy4+xp3fzV6vJWwURtFiO030WK/AT7TMxEGZjYa+DTwq5TiXhOjmVUCU4FfA7h7g7tvohfFSLgmS4mZFQClhCsn9nh87v4csCGtuL24zgLud/d6d3+PcH2Xo/d0fO4+292bosmXCFei7JH42osx8kPgX2l9GeUeibEze3uCGAWsTJmuisp6DTMbBxwBvAwMi664R3Q/tOciA+B/CV/0lpSy3hTjvkA1cGfUDfYrMyvrLTG6+yrgvwn/JNcQrqg4u7fEl0F7cfXG39E/An+MHvea+MzsTGCVu7+WNqvXxJhqb08QlqGs1+z3a2blwMPA1919S0/Hk8rMTgfWuvu8no6lAwXAkcDP3f0IYDs93+W1Q9SHfxYwHhgJlJnZRT0b1S7pVb8jM7uR0E07I1mUYbE9Hp+ZlQI3Ajdnmp2hrMe3RXt7gqgCxqRMjyY08XucmRUSksMMd38kKv7AzEZE80cAa3sqPuAjwJlmtozQNfdxM7uH3hVjFVDl7i9H0w8REkZvifGTwHvuXu3ujcAjwId7UXzp2our1/yOzOwLwOnAhb7zIK/eEt9+hD8Dr0W/m9HAq2Y2nN4TYyt7e4KYA0wws/FmliAMEs3s4ZgwMyP0m7/p7j9ImTUT+EL0+AvA7/d0bEnufoO7j3b3cYT37Wl3v4jeFeP7wEozOzAq+gTwBr0nxhXAsWZWGn3mnyCMN/WW+NK1F9dMYJqZFZnZeGAC8MqeDs7MTgGuA85095qUWb0iPndf6O5D3X1c9LupAo6Mvqe9IsY23H2vvgGnEfZ4eBe4safjiWL6KKF5uQCYH91OAwYR9h55J7of2NOxRvGeAPwhetyrYgQOB+ZG7+VjwIDeFCPwLeAt4HXgt0BRb4gPuI8wLtJI2JBd2lFchK6Td4G3gVN7KL4lhH785G9mek/F116MafOXAYN7MsbObjrVhoiIZLS3dzGJiEg7lCBERCQjJQgREclICUJERDJSghARkYyUIER6ATM7IXlGXJHeQglCREQyUoIQ6QIzu8jMXjGz+Wb2i+h6GNvM7H/M7FUz+4uZDYmWPdzMXkq5PsGAqHx/M/uzmb0WrbNfVH257bx2xYzo6GqRHqMEIZIlMzsIOA/4iLsfDjQDFwJlwKvufiTwLPDNaJW7ges8XJ9gYUr5DOA2dz+McO6lNVH5EcDXCdcm2ZdwviuRHlPQ0wGI9CGfAD4EzIn+3JcQTljXAjwQLXMP8IiZ9QP6u/uzUflvgN+ZWQUwyt0fBXD3OoCovlfcvSqang+MA17I/csSyUwJQiR7BvzG3W9oVWh2U9pyHZ2/pqNuo/qUx83o9yk9TF1MItn7C3COmQ2FHddoHkv4HZ0TLXMB8IK7bwY2mtnHovKLgWc9XNejysw+E9VRFF0nQKTX0T8UkSy5+xtm9m/AbDOLEc7S+c+ECxEdbGbzgM2EcQoIp8SeHiWApcAlUfnFwC/M7Naojn/Ygy9DJGs6m6vIbjKzbe5e3tNxiHQ3dTGJiEhGakGIiEhGakGIiEhGShAiIpKREoSIiGSkBCEiIhkpQYiISEb/Hy74WFz806XbAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Time:  6.072368498000003\n"
+     ]
+    }
+   ],
    "source": [
     "def lstm_2layers(length_of_sequences, batch_size = None, stateful = False):\n",
     "    \"\"\"\n",
@@ -1728,9 +3160,34 @@
     "end = timer()\n",
     "print('Time: ', end-start)"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
- "metadata": {},
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.3"
+  }
+ },
  "nbformat": 4,
  "nbformat_minor": 4
 }