├── Bank_marketing.ipynb
├── Credit_Fraud_detection_with_autoencoders.ipynb
├── Customer_segmentation.ipynb
├── Example of image classification with neural network.ipynb
├── Fraud_detection.ipynb
├── Fraud_detection_ensembles.ipynb
├── Greek_NLP_example_clustering.ipynb
├── Greek_Spacy.ipynb
├── Image_Classification_with_and_without_convlolution.ipynb
├── NLP_example_clustering.ipynb
├── README.md
├── Simple neural network example.ipynb
├── Spacy_example.ipynb
├── Style_transfer.ipynb
├── convolution_example.xlsx
└── data
    ├── README.md
    └── bank-additional-full_test.csv


/Example of image classification with neural network.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "name": "Example of image classification with neural network.ipynb",
  7 |       "provenance": [],
  8 |       "collapsed_sections": [],
  9 |       "include_colab_link": true
 10 |     },
 11 |     "kernelspec": {
 12 |       "name": "python3",
 13 |       "display_name": "Python 3"
 14 |     }
 15 |   },
 16 |   "cells": [
 17 |     {
 18 |       "cell_type": "markdown",
 19 |       "metadata": {
 20 |         "id": "view-in-github",
 21 |         "colab_type": "text"
 22 |       },
 23 |       "source": [
 24 |         "<a href=\"https://colab.research.google.com/github/dpanagop/ML_and_AI_examples/blob/master/Example%20of%20image%20classification%20with%20neural%20network.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
 25 |       ]
 26 |     },
 27 |     {
 28 |       "cell_type": "markdown",
 29 |       "metadata": {
 30 |         "id": "Ww3F3BuDkH_V",
 31 |         "colab_type": "text"
 32 |       },
 33 |       "source": [
 34 |         "#Example of image classification with neural network"
 35 |       ]
 36 |     },
 37 |     {
 38 |       "cell_type": "markdown",
 39 |       "metadata": {
 40 |         "id": "HidJIZ8kkg1Q",
 41 |         "colab_type": "text"
 42 |       },
 43 |       "source": [
 44 |         "###First, we load the necessary libraries"
 45 |       ]
 46 |     },
 47 |     {
 48 |       "cell_type": "code",
 49 |       "metadata": {
 50 |         "id": "QRhENDcTjx1S",
 51 |         "colab_type": "code",
 52 |         "outputId": "77348de5-0b40-45dc-f4dc-bf37ceef9694",
 53 |         "colab": {
 54 |           "base_uri": "https://localhost:8080/",
 55 |           "height": 65
 56 |         }
 57 |       },
 58 |       "source": [
 59 |         "import numpy as np\n",
 60 |         "import tensorflow as tf\n",
 61 |         "from tensorflow import keras\n",
 62 |         "import matplotlib.pyplot as plt"
 63 |       ],
 64 |       "execution_count": 0,
 65 |       "outputs": [
 66 |         {
 67 |           "output_type": "display_data",
 68 |           "data": {
 69 |             "text/html": [
 70 |               "<p style=\"color: red;\">\n",
 71 |               "The default version of TensorFlow in Colab will soon switch to TensorFlow 2.x.<br>\n",
 72 |               "We recommend you <a href=\"https://www.tensorflow.org/guide/migrate\" target=\"_blank\">upgrade</a> now \n",
 73 |               "or ensure your notebook will continue to use TensorFlow 1.x via the <code>%tensorflow_version 1.x</code> magic:\n",
 74 |               "<a href=\"https://colab.research.google.com/notebooks/tensorflow_version.ipynb\" target=\"_blank\">more info</a>.</p>\n"
 75 |             ],
 76 |             "text/plain": [
 77 |               "<IPython.core.display.HTML object>"
 78 |             ]
 79 |           },
 80 |           "metadata": {
 81 |             "tags": []
 82 |           }
 83 |         }
 84 |       ]
 85 |     },
 86 |     {
 87 |       "cell_type": "markdown",
 88 |       "metadata": {
 89 |         "id": "x1dVuDBolRg7",
 90 |         "colab_type": "text"
 91 |       },
 92 |       "source": [
 93 |         "###Then, we load Fashion MINST dataset"
 94 |       ]
 95 |     },
 96 |     {
 97 |       "cell_type": "code",
 98 |       "metadata": {
 99 |         "id": "akPbPBwplY8v",
100 |         "colab_type": "code",
101 |         "outputId": "5d504831-f824-4826-9528-6768e84bc744",
102 |         "colab": {
103 |           "base_uri": "https://localhost:8080/",
104 |           "height": 164
105 |         }
106 |       },
107 |       "source": [
108 |         "fashion_mnist = keras.datasets.fashion_mnist\n",
109 |         "(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()"
110 |       ],
111 |       "execution_count": 0,
112 |       "outputs": [
113 |         {
114 |           "output_type": "stream",
115 |           "text": [
116 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz\n",
117 |             "32768/29515 [=================================] - 0s 0us/step\n",
118 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz\n",
119 |             "26427392/26421880 [==============================] - 1s 0us/step\n",
120 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz\n",
121 |             "8192/5148 [===============================================] - 0s 0us/step\n",
122 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz\n",
123 |             "4423680/4422102 [==============================] - 0s 0us/step\n"
124 |           ],
125 |           "name": "stdout"
126 |         }
127 |       ]
128 |     },
129 |     {
130 |       "cell_type": "markdown",
131 |       "metadata": {
132 |         "id": "Xz_mDVy1ovUL",
133 |         "colab_type": "text"
134 |       },
135 |       "source": [
136 |         "###Label number correspond to:\n",
137 |         "| Label| Class|\n",
138 |         "|---------|----------|\n",
139 |         "| 0  | T-shirt/top|\n",
140 |         "| 1  | Trouser|\n",
141 |         "| 2  | Pullover|\n",
142 |         "| 3  | Dress|\n",
143 |         "| 4  | Coat|\n",
144 |         "| 5  | Sandal|\n",
145 |         "| 6  | Shirt|\n",
146 |         "| 7  | Sneaker|\n",
147 |         "| 8  | bag|\n",
148 |         "| 9  | Ankle boot|\n",
149 |         "\n",
150 |         "###We store correspondence in class_names variable"
151 |       ]
152 |     },
153 |     {
154 |       "cell_type": "code",
155 |       "metadata": {
156 |         "id": "3jTUU9zPo_pY",
157 |         "colab_type": "code",
158 |         "colab": {}
159 |       },
160 |       "source": [
161 |         "class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',\n",
162 |         "               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']"
163 |       ],
164 |       "execution_count": 0,
165 |       "outputs": []
166 |     },
167 |     {
168 |       "cell_type": "markdown",
169 |       "metadata": {
170 |         "id": "Y_7uABuIpBEV",
171 |         "colab_type": "text"
172 |       },
173 |       "source": [
174 |         "###We can see an example"
175 |       ]
176 |     },
177 |     {
178 |       "cell_type": "code",
179 |       "metadata": {
180 |         "id": "9ctTGeODlvt6",
181 |         "colab_type": "code",
182 |         "outputId": "46906eb1-11a1-4c01-f5d0-b7473bf5f08c",
183 |         "colab": {
184 |           "base_uri": "https://localhost:8080/",
185 |           "height": 238
186 |         }
187 |       },
188 |       "source": [
189 |         "f, ax = plt.subplots(1,2)\n",
190 |         "ax[0].imshow(train_images[40])\n",
191 |         "ax[1].imshow(train_images[50])\n",
192 |         "print(train_labels[40],train_labels[50])\n",
193 |         "print(class_names[6],class_names[6])"
194 |       ],
195 |       "execution_count": 0,
196 |       "outputs": [
197 |         {
198 |           "output_type": "stream",
199 |           "text": [
200 |             "6 3\n",
201 |             "Shirt Shirt\n"
202 |           ],
203 |           "name": "stdout"
204 |         },
205 |         {
206 |           "output_type": "display_data",
207 |           "data": {
208 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAC4CAYAAAD61bdSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
209 |             "text/plain": [
210 |               "<Figure size 432x288 with 2 Axes>"
211 |             ]
212 |           },
213 |           "metadata": {
214 |             "tags": []
215 |           }
216 |         }
217 |       ]
218 |     },
219 |     {
220 |       "cell_type": "markdown",
221 |       "metadata": {
222 |         "id": "84j2r_8qle4g",
223 |         "colab_type": "text"
224 |       },
225 |       "source": [
226 |         "###Next, we normalise data"
227 |       ]
228 |     },
229 |     {
230 |       "cell_type": "code",
231 |       "metadata": {
232 |         "id": "ABSLwCLslpkI",
233 |         "colab_type": "code",
234 |         "colab": {}
235 |       },
236 |       "source": [
237 |         "train_images = train_images / 255.0\n",
238 |         "test_images = test_images / 255.0\n"
239 |       ],
240 |       "execution_count": 0,
241 |       "outputs": []
242 |     },
243 |     {
244 |       "cell_type": "markdown",
245 |       "metadata": {
246 |         "id": "1P3gPj5TnWvI",
247 |         "colab_type": "text"
248 |       },
249 |       "source": [
250 |         "###We create a neural net"
251 |       ]
252 |     },
253 |     {
254 |       "cell_type": "code",
255 |       "metadata": {
256 |         "id": "EnO6W_uendrN",
257 |         "colab_type": "code",
258 |         "outputId": "4db42f2d-7f5a-4b1e-a388-2510f62eb8bb",
259 |         "colab": {
260 |           "base_uri": "https://localhost:8080/",
261 |           "height": 92
262 |         }
263 |       },
264 |       "source": [
265 |         "model = keras.Sequential([\n",
266 |         "    keras.layers.Flatten(input_shape=(28, 28)),\n",
267 |         "    keras.layers.Dense(64, activation=tf.nn.relu),\n",
268 |         "    keras.layers.Dense(10, activation=tf.nn.softmax)\n",
269 |         "])"
270 |       ],
271 |       "execution_count": 0,
272 |       "outputs": [
273 |         {
274 |           "output_type": "stream",
275 |           "text": [
276 |             "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
277 |             "Instructions for updating:\n",
278 |             "If using Keras pass *_constraint arguments to layers.\n"
279 |           ],
280 |           "name": "stdout"
281 |         }
282 |       ]
283 |     },
284 |     {
285 |       "cell_type": "markdown",
286 |       "metadata": {
287 |         "id": "f4ZajpTGnkLd",
288 |         "colab_type": "text"
289 |       },
290 |       "source": [
291 |         "###Next, compile the model"
292 |       ]
293 |     },
294 |     {
295 |       "cell_type": "code",
296 |       "metadata": {
297 |         "id": "cs-CinJAnjxy",
298 |         "colab_type": "code",
299 |         "colab": {}
300 |       },
301 |       "source": [
302 |         "model.compile(optimizer='adam', \n",
303 |         "              loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
304 |       ],
305 |       "execution_count": 0,
306 |       "outputs": []
307 |     },
308 |     {
309 |       "cell_type": "markdown",
310 |       "metadata": {
311 |         "id": "F-Cd-ptRns1w",
312 |         "colab_type": "text"
313 |       },
314 |       "source": [
315 |         "###Train model"
316 |       ]
317 |     },
318 |     {
319 |       "cell_type": "code",
320 |       "metadata": {
321 |         "id": "8bfZQh8vnujg",
322 |         "colab_type": "code",
323 |         "outputId": "368fff7b-3aa3-4fe8-efd0-e33cfe3e47d9",
324 |         "colab": {
325 |           "base_uri": "https://localhost:8080/",
326 |           "height": 421
327 |         }
328 |       },
329 |       "source": [
330 |         "model.fit(train_images, train_labels, epochs=10)"
331 |       ],
332 |       "execution_count": 0,
333 |       "outputs": [
334 |         {
335 |           "output_type": "stream",
336 |           "text": [
337 |             "Train on 60000 samples\n",
338 |             "Epoch 1/10\n",
339 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.5175 - acc: 0.8194\n",
340 |             "Epoch 2/10\n",
341 |             "60000/60000 [==============================] - 4s 62us/sample - loss: 0.3921 - acc: 0.8615\n",
342 |             "Epoch 3/10\n",
343 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.3565 - acc: 0.8711\n",
344 |             "Epoch 4/10\n",
345 |             "60000/60000 [==============================] - 4s 60us/sample - loss: 0.3330 - acc: 0.8784\n",
346 |             "Epoch 5/10\n",
347 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.3146 - acc: 0.8836\n",
348 |             "Epoch 6/10\n",
349 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.3026 - acc: 0.8891\n",
350 |             "Epoch 7/10\n",
351 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.2893 - acc: 0.8948\n",
352 |             "Epoch 8/10\n",
353 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.2778 - acc: 0.8982\n",
354 |             "Epoch 9/10\n",
355 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.2695 - acc: 0.9009\n",
356 |             "Epoch 10/10\n",
357 |             "60000/60000 [==============================] - 4s 61us/sample - loss: 0.2632 - acc: 0.9036\n"
358 |           ],
359 |           "name": "stdout"
360 |         },
361 |         {
362 |           "output_type": "execute_result",
363 |           "data": {
364 |             "text/plain": [
365 |               "<tensorflow.python.keras.callbacks.History at 0x7f4ffa687400>"
366 |             ]
367 |           },
368 |           "metadata": {
369 |             "tags": []
370 |           },
371 |           "execution_count": 8
372 |         }
373 |       ]
374 |     },
375 |     {
376 |       "cell_type": "markdown",
377 |       "metadata": {
378 |         "id": "Lb0BXV-6n7aJ",
379 |         "colab_type": "text"
380 |       },
381 |       "source": [
382 |         "###Predict class of second image in test dataset"
383 |       ]
384 |     },
385 |     {
386 |       "cell_type": "code",
387 |       "metadata": {
388 |         "id": "s8jsni0bn_f1",
389 |         "colab_type": "code",
390 |         "outputId": "caca936e-c1f8-45bd-d572-5461e515a52c",
391 |         "colab": {
392 |           "base_uri": "https://localhost:8080/",
393 |           "height": 90
394 |         }
395 |       },
396 |       "source": [
397 |         "prediction=model.predict(test_images[1].reshape(1, 28, 28))\n",
398 |         "print(\"Probabilities of image in each class are\",prediction)\n",
399 |         "print(\"Highest probability in place:\", prediction.argmax())\n",
400 |         "print(\"Image is classified as a: \",class_names[prediction.argmax()])"
401 |       ],
402 |       "execution_count": 0,
403 |       "outputs": [
404 |         {
405 |           "output_type": "stream",
406 |           "text": [
407 |             "Probabilities of image in each class are [[4.7269925e-05 6.7162512e-11 9.9416757e-01 7.9031757e-08 2.6153151e-03\n",
408 |             "  7.7022203e-14 3.1697203e-03 3.2766182e-15 1.2217416e-08 4.0623840e-15]]\n",
409 |             "Highest probability in place: 2\n",
410 |             "Image is classified as a:  Pullover\n"
411 |           ],
412 |           "name": "stdout"
413 |         }
414 |       ]
415 |     },
416 |     {
417 |       "cell_type": "markdown",
418 |       "metadata": {
419 |         "id": "NnRT1sg7rQHi",
420 |         "colab_type": "text"
421 |       },
422 |       "source": [
423 |         ""
424 |       ]
425 |     },
426 |     {
427 |       "cell_type": "code",
428 |       "metadata": {
429 |         "id": "zslaJQ7VoZDu",
430 |         "colab_type": "code",
431 |         "outputId": "e2d9cd85-a5ac-4558-c83f-61b748a005ed",
432 |         "colab": {
433 |           "base_uri": "https://localhost:8080/",
434 |           "height": 320
435 |         }
436 |       },
437 |       "source": [
438 |         "print(\"Actual label number is:\",test_labels[1])\n",
439 |         "print(\"Image is a: \",class_names[test_labels[1]])\n",
440 |         "plt.imshow(test_images[1])\n"
441 |       ],
442 |       "execution_count": 0,
443 |       "outputs": [
444 |         {
445 |           "output_type": "stream",
446 |           "text": [
447 |             "Actual label number is: 2\n",
448 |             "Image is a:  Pullover\n"
449 |           ],
450 |           "name": "stdout"
451 |         },
452 |         {
453 |           "output_type": "execute_result",
454 |           "data": {
455 |             "text/plain": [
456 |               "<matplotlib.image.AxesImage at 0x7f4ff5013cc0>"
457 |             ]
458 |           },
459 |           "metadata": {
460 |             "tags": []
461 |           },
462 |           "execution_count": 10
463 |         },
464 |         {
465 |           "output_type": "display_data",
466 |           "data": {
467 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niZxg2ImcYNiJnGDYiZz4f4SixgHCORlKAAAAAElFTkSuQmCC\n",
468 |             "text/plain": [
469 |               "<Figure size 432x288 with 1 Axes>"
470 |             ]
471 |           },
472 |           "metadata": {
473 |             "tags": []
474 |           }
475 |         }
476 |       ]
477 |     },
478 |     {
479 |       "cell_type": "markdown",
480 |       "metadata": {
481 |         "id": "KaJlvr76sOtB",
482 |         "colab_type": "text"
483 |       },
484 |       "source": [
485 |         "###Evaluate models accuracy"
486 |       ]
487 |     },
488 |     {
489 |       "cell_type": "code",
490 |       "metadata": {
491 |         "id": "iyCJ4rDzsRx8",
492 |         "colab_type": "code",
493 |         "outputId": "32613819-27fa-4aa7-d611-9ce66e2cd5de",
494 |         "colab": {
495 |           "base_uri": "https://localhost:8080/",
496 |           "height": 54
497 |         }
498 |       },
499 |       "source": [
500 |         "test_acc = model.evaluate(test_images, test_labels)\n",
501 |         "print('Test loss, accuracy:', test_acc)\n"
502 |       ],
503 |       "execution_count": 0,
504 |       "outputs": [
505 |         {
506 |           "output_type": "stream",
507 |           "text": [
508 |             "10000/10000 [==============================] - 0s 34us/sample - loss: 0.3430 - acc: 0.8819\n",
509 |             "Test loss, accuracy: [0.34303882870674135, 0.8819]\n"
510 |           ],
511 |           "name": "stdout"
512 |         }
513 |       ]
514 |     },
515 |     {
516 |       "cell_type": "code",
517 |       "metadata": {
518 |         "id": "GQMgOE88CFNZ",
519 |         "colab_type": "code",
520 |         "outputId": "2fbe7902-26de-41de-9c19-6f9e204ffd92",
521 |         "colab": {
522 |           "base_uri": "https://localhost:8080/",
523 |           "height": 394
524 |         }
525 |       },
526 |       "source": [
527 |         "prediction=model.predict(test_images[12].reshape(1, 28, 28))\n",
528 |         "print(\"Probabilities of image in each class are\",prediction)\n",
529 |         "print(\"Highest probability in place:\", prediction.argmax())\n",
530 |         "print(\"Image is classified as a: \",class_names[prediction.argmax()])\n",
531 |         "print(\"Actual label number is:\",test_labels[12])\n",
532 |         "print(\"Image is a: \",class_names[test_labels[12]])\n",
533 |         "plt.imshow(test_images[12])"
534 |       ],
535 |       "execution_count": 0,
536 |       "outputs": [
537 |         {
538 |           "output_type": "stream",
539 |           "text": [
540 |             "Probabilities of image in each class are [[2.9833541e-07 3.4984214e-07 5.2938790e-06 2.2196046e-07 3.0843472e-05\n",
541 |             "  1.4087908e-01 5.4214041e-08 2.2918980e-01 6.2989372e-01 2.9229400e-07]]\n",
542 |             "Highest probability in place: 8\n",
543 |             "Image is classified as a:  Bag\n",
544 |             "Actual label number is: 7\n",
545 |             "Image is a:  Sneaker\n"
546 |           ],
547 |           "name": "stdout"
548 |         },
549 |         {
550 |           "output_type": "execute_result",
551 |           "data": {
552 |             "text/plain": [
553 |               "<matplotlib.image.AxesImage at 0x7f4ffbfe4470>"
554 |             ]
555 |           },
556 |           "metadata": {
557 |             "tags": []
558 |           },
559 |           "execution_count": 12
560 |         },
561 |         {
562 |           "output_type": "display_data",
563 |           "data": {
564 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKF7GlmsGXKMAAAAASUVORK5CYII=\n",
565 |             "text/plain": [
566 |               "<Figure size 432x288 with 1 Axes>"
567 |             ]
568 |           },
569 |           "metadata": {
570 |             "tags": []
571 |           }
572 |         }
573 |       ]
574 |     },
575 |     {
576 |       "cell_type": "code",
577 |       "metadata": {
578 |         "id": "y8Tane_mDHI-",
579 |         "colab_type": "code",
580 |         "outputId": "caa88df2-0a13-4083-dc1b-b88c2b78db19",
581 |         "colab": {
582 |           "base_uri": "https://localhost:8080/",
583 |           "height": 394
584 |         }
585 |       },
586 |       "source": [
587 |         "prediction=model.predict(test_images[17].reshape(1, 28, 28))\n",
588 |         "print(\"Probabilities of image in each class are\",prediction)\n",
589 |         "print(\"Highest probability in place:\", prediction.argmax())\n",
590 |         "print(\"Image is classified as a: \",class_names[prediction.argmax()])\n",
591 |         "print(\"Actual label number is:\",test_labels[17])\n",
592 |         "print(\"Image is a: \",class_names[test_labels[17]])\n",
593 |         "plt.imshow(test_images[17])"
594 |       ],
595 |       "execution_count": 0,
596 |       "outputs": [
597 |         {
598 |           "output_type": "stream",
599 |           "text": [
600 |             "Probabilities of image in each class are [[2.6976159e-03 2.6915019e-09 9.8906553e-01 4.8810068e-07 3.9343471e-03\n",
601 |             "  4.9612314e-10 4.2868727e-03 5.2282022e-11 1.5191695e-05 1.4628523e-10]]\n",
602 |             "Highest probability in place: 2\n",
603 |             "Image is classified as a:  Pullover\n",
604 |             "Actual label number is: 4\n",
605 |             "Image is a:  Coat\n"
606 |           ],
607 |           "name": "stdout"
608 |         },
609 |         {
610 |           "output_type": "execute_result",
611 |           "data": {
612 |             "text/plain": [
613 |               "<matplotlib.image.AxesImage at 0x7f4ff1fb0f28>"
614 |             ]
615 |           },
616 |           "metadata": {
617 |             "tags": []
618 |           },
619 |           "execution_count": 13
620 |         },
621 |         {
622 |           "output_type": "display_data",
623 |           "data": {
624 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
625 |             "text/plain": [
626 |               "<Figure size 432x288 with 1 Axes>"
627 |             ]
628 |           },
629 |           "metadata": {
630 |             "tags": []
631 |           }
632 |         }
633 |       ]
634 |     }
635 |   ]
636 | }


--------------------------------------------------------------------------------
/Greek_Spacy.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "name": "Greek Spacy.ipynb",
  7 |       "provenance": [],
  8 |       "collapsed_sections": [],
  9 |       "include_colab_link": true
 10 |     },
 11 |     "kernelspec": {
 12 |       "name": "python3",
 13 |       "display_name": "Python 3"
 14 |     }
 15 |   },
 16 |   "cells": [
 17 |     {
 18 |       "cell_type": "markdown",
 19 |       "metadata": {
 20 |         "id": "view-in-github",
 21 |         "colab_type": "text"
 22 |       },
 23 |       "source": [
 24 |         "<a href=\"https://colab.research.google.com/github/dpanagop/ML_and_AI_examples/blob/master/Greek_Spacy.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
 25 |       ]
 26 |     },
 27 |     {
 28 |       "cell_type": "code",
 29 |       "metadata": {
 30 |         "id": "yIARIXVzyqZe",
 31 |         "colab_type": "code",
 32 |         "outputId": "0fe8fd77-c0fd-4699-96cb-4ecae7317b14",
 33 |         "colab": {
 34 |           "base_uri": "https://localhost:8080/",
 35 |           "height": 845
 36 |         }
 37 |       },
 38 |       "source": [
 39 |         "!pip install -U spacy"
 40 |       ],
 41 |       "execution_count": 0,
 42 |       "outputs": [
 43 |         {
 44 |           "output_type": "stream",
 45 |           "text": [
 46 |             "Collecting spacy\n",
 47 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/47/13/80ad28ef7a16e2a86d16d73e28588be5f1085afd3e85e4b9b912bd700e8a/spacy-2.2.3-cp36-cp36m-manylinux1_x86_64.whl (10.4MB)\n",
 48 |             "\u001b[K     |████████████████████████████████| 10.4MB 4.2MB/s \n",
 49 |             "\u001b[?25hRequirement already satisfied, skipping upgrade: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy) (2.0.3)\n",
 50 |             "Requirement already satisfied, skipping upgrade: srsly<1.1.0,>=0.1.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (0.2.0)\n",
 51 |             "Requirement already satisfied, skipping upgrade: numpy>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (1.17.4)\n",
 52 |             "Collecting preshed<3.1.0,>=3.0.2\n",
 53 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/db/6b/e07fad36913879757c90ba03d6fb7f406f7279e11dcefc105ee562de63ea/preshed-3.0.2-cp36-cp36m-manylinux1_x86_64.whl (119kB)\n",
 54 |             "\u001b[K     |████████████████████████████████| 122kB 42.2MB/s \n",
 55 |             "\u001b[?25hCollecting thinc<7.4.0,>=7.3.0\n",
 56 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/07/59/6bb553bc9a5f072d3cd479fc939fea0f6f682892f1f5cff98de5c9b615bb/thinc-7.3.1-cp36-cp36m-manylinux1_x86_64.whl (2.2MB)\n",
 57 |             "\u001b[K     |████████████████████████████████| 2.2MB 30.3MB/s \n",
 58 |             "\u001b[?25hRequirement already satisfied, skipping upgrade: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (1.0.2)\n",
 59 |             "Collecting catalogue<1.1.0,>=0.0.7\n",
 60 |             "  Downloading https://files.pythonhosted.org/packages/4f/d5/46ff975f0d7d055cf95557b944fd5d29d9dfb37a4341038e070f212b24fe/catalogue-0.0.8-py2.py3-none-any.whl\n",
 61 |             "Requirement already satisfied, skipping upgrade: setuptools in /usr/local/lib/python3.6/dist-packages (from spacy) (41.6.0)\n",
 62 |             "Requirement already satisfied, skipping upgrade: wasabi<1.1.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (0.4.0)\n",
 63 |             "Requirement already satisfied, skipping upgrade: plac<1.2.0,>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from spacy) (0.9.6)\n",
 64 |             "Requirement already satisfied, skipping upgrade: requests<3.0.0,>=2.13.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (2.21.0)\n",
 65 |             "Collecting blis<0.5.0,>=0.4.0\n",
 66 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/41/19/f95c75562d18eb27219df3a3590b911e78d131b68466ad79fdf5847eaac4/blis-0.4.1-cp36-cp36m-manylinux1_x86_64.whl (3.7MB)\n",
 67 |             "\u001b[K     |████████████████████████████████| 3.7MB 41.9MB/s \n",
 68 |             "\u001b[?25hRequirement already satisfied, skipping upgrade: tqdm<5.0.0,>=4.10.0 in /usr/local/lib/python3.6/dist-packages (from thinc<7.4.0,>=7.3.0->spacy) (4.28.1)\n",
 69 |             "Requirement already satisfied, skipping upgrade: importlib-metadata>=0.20; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from catalogue<1.1.0,>=0.0.7->spacy) (0.23)\n",
 70 |             "Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (2019.9.11)\n",
 71 |             "Requirement already satisfied, skipping upgrade: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (3.0.4)\n",
 72 |             "Requirement already satisfied, skipping upgrade: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (1.24.3)\n",
 73 |             "Requirement already satisfied, skipping upgrade: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (2.8)\n",
 74 |             "Requirement already satisfied, skipping upgrade: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy) (0.6.0)\n",
 75 |             "Requirement already satisfied, skipping upgrade: more-itertools in /usr/local/lib/python3.6/dist-packages (from zipp>=0.5->importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy) (7.2.0)\n",
 76 |             "Installing collected packages: preshed, blis, thinc, catalogue, spacy\n",
 77 |             "  Found existing installation: preshed 2.0.1\n",
 78 |             "    Uninstalling preshed-2.0.1:\n",
 79 |             "      Successfully uninstalled preshed-2.0.1\n",
 80 |             "  Found existing installation: blis 0.2.4\n",
 81 |             "    Uninstalling blis-0.2.4:\n",
 82 |             "      Successfully uninstalled blis-0.2.4\n",
 83 |             "  Found existing installation: thinc 7.0.8\n",
 84 |             "    Uninstalling thinc-7.0.8:\n",
 85 |             "      Successfully uninstalled thinc-7.0.8\n",
 86 |             "  Found existing installation: spacy 2.1.9\n",
 87 |             "    Uninstalling spacy-2.1.9:\n",
 88 |             "      Successfully uninstalled spacy-2.1.9\n",
 89 |             "Successfully installed blis-0.4.1 catalogue-0.0.8 preshed-3.0.2 spacy-2.2.3 thinc-7.3.1\n"
 90 |           ],
 91 |           "name": "stdout"
 92 |         }
 93 |       ]
 94 |     },
 95 |     {
 96 |       "cell_type": "code",
 97 |       "metadata": {
 98 |         "id": "LXpmvaPijt4w",
 99 |         "colab_type": "code",
100 |         "outputId": "8b9850b6-c137-4982-fb82-697212c64d06",
101 |         "colab": {
102 |           "base_uri": "https://localhost:8080/",
103 |           "height": 717
104 |         }
105 |       },
106 |       "source": [
107 |         "!python -m spacy download el"
108 |       ],
109 |       "execution_count": 0,
110 |       "outputs": [
111 |         {
112 |           "output_type": "stream",
113 |           "text": [
114 |             "Collecting el_core_news_sm==2.2.5\n",
115 |             "\u001b[?25l  Downloading https://github.com/explosion/spacy-models/releases/download/el_core_news_sm-2.2.5/el_core_news_sm-2.2.5.tar.gz (11.4MB)\n",
116 |             "\u001b[K     |████████████████████████████████| 11.4MB 790kB/s \n",
117 |             "\u001b[?25hRequirement already satisfied: spacy>=2.2.2 in /usr/local/lib/python3.6/dist-packages (from el_core_news_sm==2.2.5) (2.2.3)\n",
118 |             "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (2.0.3)\n",
119 |             "Requirement already satisfied: numpy>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (1.17.4)\n",
120 |             "Requirement already satisfied: catalogue<1.1.0,>=0.0.7 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (0.0.8)\n",
121 |             "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (41.6.0)\n",
122 |             "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (3.0.2)\n",
123 |             "Requirement already satisfied: thinc<7.4.0,>=7.3.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (7.3.1)\n",
124 |             "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (1.0.2)\n",
125 |             "Requirement already satisfied: wasabi<1.1.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (0.4.0)\n",
126 |             "Requirement already satisfied: plac<1.2.0,>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (0.9.6)\n",
127 |             "Requirement already satisfied: requests<3.0.0,>=2.13.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (2.21.0)\n",
128 |             "Requirement already satisfied: blis<0.5.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (0.4.1)\n",
129 |             "Requirement already satisfied: srsly<1.1.0,>=0.1.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->el_core_news_sm==2.2.5) (0.2.0)\n",
130 |             "Requirement already satisfied: importlib-metadata>=0.20; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->el_core_news_sm==2.2.5) (0.23)\n",
131 |             "Requirement already satisfied: tqdm<5.0.0,>=4.10.0 in /usr/local/lib/python3.6/dist-packages (from thinc<7.4.0,>=7.3.0->spacy>=2.2.2->el_core_news_sm==2.2.5) (4.28.1)\n",
132 |             "Requirement already satisfied: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->el_core_news_sm==2.2.5) (1.24.3)\n",
133 |             "Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->el_core_news_sm==2.2.5) (3.0.4)\n",
134 |             "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->el_core_news_sm==2.2.5) (2019.9.11)\n",
135 |             "Requirement already satisfied: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->el_core_news_sm==2.2.5) (2.8)\n",
136 |             "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->el_core_news_sm==2.2.5) (0.6.0)\n",
137 |             "Requirement already satisfied: more-itertools in /usr/local/lib/python3.6/dist-packages (from zipp>=0.5->importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->el_core_news_sm==2.2.5) (7.2.0)\n",
138 |             "Building wheels for collected packages: el-core-news-sm\n",
139 |             "  Building wheel for el-core-news-sm (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
140 |             "  Created wheel for el-core-news-sm: filename=el_core_news_sm-2.2.5-cp36-none-any.whl size=11422786 sha256=a2e4fd3c86575b7c8ae7a4ec4211ae68a7c27a748573477b9b43a831742e2b44\n",
141 |             "  Stored in directory: /tmp/pip-ephem-wheel-cache-634wxpxz/wheels/70/a1/c5/6690d6b524d87e287a8070cf957f834fb1b1665b9ede11348b\n",
142 |             "Successfully built el-core-news-sm\n",
143 |             "Installing collected packages: el-core-news-sm\n",
144 |             "Successfully installed el-core-news-sm-2.2.5\n",
145 |             "\u001b[38;5;2m✔ Download and installation successful\u001b[0m\n",
146 |             "You can now load the model via spacy.load('el_core_news_sm')\n",
147 |             "\u001b[38;5;2m✔ Linking successful\u001b[0m\n",
148 |             "/usr/local/lib/python3.6/dist-packages/el_core_news_sm -->\n",
149 |             "/usr/local/lib/python3.6/dist-packages/spacy/data/el\n",
150 |             "You can now load the model via spacy.load('el')\n"
151 |           ],
152 |           "name": "stdout"
153 |         }
154 |       ]
155 |     },
156 |     {
157 |       "cell_type": "code",
158 |       "metadata": {
159 |         "id": "Gc4rd8JJyS9Y",
160 |         "colab_type": "code",
161 |         "colab": {}
162 |       },
163 |       "source": [
164 |         "import spacy\n",
165 |         "#nlp = spacy.load(\"el_core_news_sm\")\n",
166 |         "nlp = spacy.load(\"el\")"
167 |       ],
168 |       "execution_count": 0,
169 |       "outputs": []
170 |     },
171 |     {
172 |       "cell_type": "code",
173 |       "metadata": {
174 |         "colab_type": "code",
175 |         "id": "p8dDdyS8kgXO",
176 |         "colab": {}
177 |       },
178 |       "source": [
179 |         "sample_text=\"Αυτό είναι ένα παράδειγμα για την επεξεργασία κειμένου. Δημιουργήθηκε από το Δημήτρη Παναγόπουλο τον Νoέμβριο του 2019 στην Αθήνα. Μπορείτε να το τρέξετε στο Colab της Google\"\n",
180 |         "doc = nlp(sample_text)"
181 |       ],
182 |       "execution_count": 0,
183 |       "outputs": []
184 |     },
185 |     {
186 |       "cell_type": "code",
187 |       "metadata": {
188 |         "id": "ge3FRXSA0pQ0",
189 |         "colab_type": "code",
190 |         "outputId": "bfc246c1-736d-47fb-d4cc-651c333f1654",
191 |         "colab": {
192 |           "base_uri": "https://localhost:8080/",
193 |           "height": 550
194 |         }
195 |       },
196 |       "source": [
197 |         "for token in doc:\n",
198 |         "    print(token.text, token.lemma_, token.pos_)"
199 |       ],
200 |       "execution_count": 0,
201 |       "outputs": [
202 |         {
203 |           "output_type": "stream",
204 |           "text": [
205 |             "Αυτό αυτό PRON\n",
206 |             "είναι είναι AUX\n",
207 |             "ένα ένα DET\n",
208 |             "παράδειγμα παράδειγμα NOUN\n",
209 |             "για για ADP\n",
210 |             "την την DET\n",
211 |             "επεξεργασία επεξεργασίας NOUN\n",
212 |             "κειμένου κειμένο NOUN\n",
213 |             ". . PUNCT\n",
214 |             "Δημιουργήθηκε δημιουργήθηκε VERB\n",
215 |             "από από ADP\n",
216 |             "το το DET\n",
217 |             "Δημήτρη δημήτρη NOUN\n",
218 |             "Παναγόπουλο παναγόπουλο NOUN\n",
219 |             "τον τον DET\n",
220 |             "Νoέμβριο νoέμβριο NOUN\n",
221 |             "του του DET\n",
222 |             "2019 2019 NUM\n",
223 |             "στην στην ADJ\n",
224 |             "Αθήνα Αθήνα PROPN\n",
225 |             ". . PUNCT\n",
226 |             "Μπορείτε μπορείτε VERB\n",
227 |             "να να PART\n",
228 |             "το το PRON\n",
229 |             "τρέξετε τρέξω VERB\n",
230 |             "στο στο ADV\n",
231 |             "Colab colab X\n",
232 |             "της της DET\n",
233 |             "Google google X\n"
234 |           ],
235 |           "name": "stdout"
236 |         }
237 |       ]
238 |     },
239 |     {
240 |       "cell_type": "code",
241 |       "metadata": {
242 |         "id": "uwp80jJtlYUE",
243 |         "colab_type": "code",
244 |         "outputId": "a7270266-1e2d-4680-e732-eaf99c825f8b",
245 |         "colab": {
246 |           "base_uri": "https://localhost:8080/",
247 |           "height": 122
248 |         }
249 |       },
250 |       "source": [
251 |         "from spacy import displacy\n",
252 |         "displacy.render(doc, style=\"ent\", jupyter=True)"
253 |       ],
254 |       "execution_count": 0,
255 |       "outputs": [
256 |         {
257 |           "output_type": "display_data",
258 |           "data": {
259 |             "text/html": [
260 |               "<div class=\"entities\" style=\"line-height: 2.5; direction: ltr\">Αυτό είναι ένα παράδειγμα για την επεξεργασία κειμένου. Δημιουργήθηκε από το \n",
261 |               "<mark class=\"entity\" style=\"background: #7aecec; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
262 |               "    Δημήτρη Παναγόπουλο\n",
263 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">ORG</span>\n",
264 |               "</mark>\n",
265 |               " τον Νoέμβριο του 2019 στην \n",
266 |               "<mark class=\"entity\" style=\"background: #feca74; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
267 |               "    Αθήνα\n",
268 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">GPE</span>\n",
269 |               "</mark>\n",
270 |               ". Μπορείτε να το τρέξετε στο Colab της \n",
271 |               "<mark class=\"entity\" style=\"background: #7aecec; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
272 |               "    Google\n",
273 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">ORG</span>\n",
274 |               "</mark>\n",
275 |               "</div>"
276 |             ],
277 |             "text/plain": [
278 |               "<IPython.core.display.HTML object>"
279 |             ]
280 |           },
281 |           "metadata": {
282 |             "tags": []
283 |           }
284 |         }
285 |       ]
286 |     },
287 |     {
288 |       "cell_type": "code",
289 |       "metadata": {
290 |         "id": "IZXd10Tnmefz",
291 |         "colab_type": "code",
292 |         "colab": {}
293 |       },
294 |       "source": [
295 |         "sample_words=\"σκύλος γάτα βασιλιάς\""
296 |       ],
297 |       "execution_count": 0,
298 |       "outputs": []
299 |     },
300 |     {
301 |       "cell_type": "code",
302 |       "metadata": {
303 |         "id": "Msv3wlgbpk_y",
304 |         "colab_type": "code",
305 |         "colab": {}
306 |       },
307 |       "source": [
308 |         "tokens=nlp(sample_words)"
309 |       ],
310 |       "execution_count": 0,
311 |       "outputs": []
312 |     },
313 |     {
314 |       "cell_type": "code",
315 |       "metadata": {
316 |         "id": "-hacpD7rppfw",
317 |         "colab_type": "code",
318 |         "outputId": "ff897f43-b591-4b97-caa0-ca402550c2d1",
319 |         "colab": {
320 |           "base_uri": "https://localhost:8080/",
321 |           "height": 35
322 |         }
323 |       },
324 |       "source": [
325 |         "print(tokens)"
326 |       ],
327 |       "execution_count": 0,
328 |       "outputs": [
329 |         {
330 |           "output_type": "stream",
331 |           "text": [
332 |             "σκύλος γάτα βασιλιάς\n"
333 |           ],
334 |           "name": "stdout"
335 |         }
336 |       ]
337 |     },
338 |     {
339 |       "cell_type": "code",
340 |       "metadata": {
341 |         "id": "0uNtAMTNpq_i",
342 |         "colab_type": "code",
343 |         "outputId": "14f4ee3e-3bff-40aa-b488-be3cc7f93c3e",
344 |         "colab": {
345 |           "base_uri": "https://localhost:8080/",
346 |           "height": 92
347 |         }
348 |       },
349 |       "source": [
350 |         "print(tokens[0].similarity(tokens[1]))"
351 |       ],
352 |       "execution_count": 0,
353 |       "outputs": [
354 |         {
355 |           "output_type": "stream",
356 |           "text": [
357 |             "/usr/lib/python3.6/runpy.py:193: ModelsWarning: [W007] The model you're using has no word vectors loaded, so the result of the Token.similarity method will be based on the tagger, parser and NER, which may not give useful similarity judgements. This may happen if you're using one of the small models, e.g. `en_core_web_sm`, which don't ship with word vectors and only use context-sensitive tensors. You can always add your own word vectors, or use one of the larger models instead if available.\n",
358 |             "  \"__main__\", mod_spec)\n"
359 |           ],
360 |           "name": "stderr"
361 |         },
362 |         {
363 |           "output_type": "execute_result",
364 |           "data": {
365 |             "text/plain": [
366 |               "0.69062674"
367 |             ]
368 |           },
369 |           "metadata": {
370 |             "tags": []
371 |           },
372 |           "execution_count": 31
373 |         }
374 |       ]
375 |     },
376 |     {
377 |       "cell_type": "code",
378 |       "metadata": {
379 |         "id": "6nWtntI7puNh",
380 |         "colab_type": "code",
381 |         "outputId": "cf44e0eb-8f76-45db-bd80-97a1716c0214",
382 |         "colab": {
383 |           "base_uri": "https://localhost:8080/",
384 |           "height": 92
385 |         }
386 |       },
387 |       "source": [
388 |         "print(tokens[0].similarity(tokens[2]))"
389 |       ],
390 |       "execution_count": 0,
391 |       "outputs": [
392 |         {
393 |           "output_type": "stream",
394 |           "text": [
395 |             "0.4917702\n"
396 |           ],
397 |           "name": "stdout"
398 |         },
399 |         {
400 |           "output_type": "stream",
401 |           "text": [
402 |             "/usr/lib/python3.6/runpy.py:193: ModelsWarning: [W007] The model you're using has no word vectors loaded, so the result of the Token.similarity method will be based on the tagger, parser and NER, which may not give useful similarity judgements. This may happen if you're using one of the small models, e.g. `en_core_web_sm`, which don't ship with word vectors and only use context-sensitive tensors. You can always add your own word vectors, or use one of the larger models instead if available.\n",
403 |             "  \"__main__\", mod_spec)\n"
404 |           ],
405 |           "name": "stderr"
406 |         }
407 |       ]
408 |     },
409 |     {
410 |       "cell_type": "code",
411 |       "metadata": {
412 |         "id": "uflt2YADqmG0",
413 |         "colab_type": "code",
414 |         "colab": {}
415 |       },
416 |       "source": [
417 |         ""
418 |       ],
419 |       "execution_count": 0,
420 |       "outputs": []
421 |     }
422 |   ]
423 | }


--------------------------------------------------------------------------------
/Image_Classification_with_and_without_convlolution.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "name": "Image Classification with and without convlolution.ipynb",
  7 |       "provenance": [],
  8 |       "collapsed_sections": [],
  9 |       "include_colab_link": true
 10 |     },
 11 |     "kernelspec": {
 12 |       "name": "python3",
 13 |       "display_name": "Python 3"
 14 |     },
 15 |     "accelerator": "TPU"
 16 |   },
 17 |   "cells": [
 18 |     {
 19 |       "cell_type": "markdown",
 20 |       "metadata": {
 21 |         "id": "view-in-github",
 22 |         "colab_type": "text"
 23 |       },
 24 |       "source": [
 25 |         "<a href=\"https://colab.research.google.com/github/dpanagop/ML_and_AI_examples/blob/master/Image_Classification_with_and_without_convlolution.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
 26 |       ]
 27 |     },
 28 |     {
 29 |       "cell_type": "markdown",
 30 |       "metadata": {
 31 |         "id": "Ww3F3BuDkH_V",
 32 |         "colab_type": "text"
 33 |       },
 34 |       "source": [
 35 |         "#Image Classification with and without convlolution"
 36 |       ]
 37 |     },
 38 |     {
 39 |       "cell_type": "markdown",
 40 |       "metadata": {
 41 |         "id": "HidJIZ8kkg1Q",
 42 |         "colab_type": "text"
 43 |       },
 44 |       "source": [
 45 |         "###First, we load the necessary libraries"
 46 |       ]
 47 |     },
 48 |     {
 49 |       "cell_type": "code",
 50 |       "metadata": {
 51 |         "id": "QRhENDcTjx1S",
 52 |         "colab_type": "code",
 53 |         "colab": {}
 54 |       },
 55 |       "source": [
 56 |         "import numpy as np\n",
 57 |         "import tensorflow as tf\n",
 58 |         "from tensorflow import keras\n",
 59 |         "import matplotlib.pyplot as plt"
 60 |       ],
 61 |       "execution_count": 0,
 62 |       "outputs": []
 63 |     },
 64 |     {
 65 |       "cell_type": "markdown",
 66 |       "metadata": {
 67 |         "id": "x1dVuDBolRg7",
 68 |         "colab_type": "text"
 69 |       },
 70 |       "source": [
 71 |         "###Then, we load Fashion MINST dataset"
 72 |       ]
 73 |     },
 74 |     {
 75 |       "cell_type": "code",
 76 |       "metadata": {
 77 |         "id": "akPbPBwplY8v",
 78 |         "colab_type": "code",
 79 |         "outputId": "836eafb9-a6e8-4f21-ef89-cbb4ebcd98dc",
 80 |         "colab": {
 81 |           "base_uri": "https://localhost:8080/",
 82 |           "height": 164
 83 |         }
 84 |       },
 85 |       "source": [
 86 |         "fashion_mnist = keras.datasets.fashion_mnist\n",
 87 |         "(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()"
 88 |       ],
 89 |       "execution_count": 0,
 90 |       "outputs": [
 91 |         {
 92 |           "output_type": "stream",
 93 |           "text": [
 94 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz\n",
 95 |             "32768/29515 [=================================] - 0s 0us/step\n",
 96 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz\n",
 97 |             "26427392/26421880 [==============================] - 1s 0us/step\n",
 98 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz\n",
 99 |             "8192/5148 [===============================================] - 0s 0us/step\n",
100 |             "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz\n",
101 |             "4423680/4422102 [==============================] - 0s 0us/step\n"
102 |           ],
103 |           "name": "stdout"
104 |         }
105 |       ]
106 |     },
107 |     {
108 |       "cell_type": "markdown",
109 |       "metadata": {
110 |         "id": "Xz_mDVy1ovUL",
111 |         "colab_type": "text"
112 |       },
113 |       "source": [
114 |         "###Label number correspond to:\n",
115 |         "| Label| Class|\n",
116 |         "|---------|----------|\n",
117 |         "| 0  | T-shirt/top|\n",
118 |         "| 1  | Trouser|\n",
119 |         "| 2  | Pullover|\n",
120 |         "| 3  | Dress|\n",
121 |         "| 4  | Coat|\n",
122 |         "| 5  | Sandal|\n",
123 |         "| 6  | Shirt|\n",
124 |         "| 7  | Sneaker|\n",
125 |         "| 8  | bag|\n",
126 |         "| 9  | Ankle boot|\n",
127 |         "\n",
128 |         "###We store correspondence in class_names variable"
129 |       ]
130 |     },
131 |     {
132 |       "cell_type": "code",
133 |       "metadata": {
134 |         "id": "3jTUU9zPo_pY",
135 |         "colab_type": "code",
136 |         "colab": {}
137 |       },
138 |       "source": [
139 |         "class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',\n",
140 |         "               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']"
141 |       ],
142 |       "execution_count": 0,
143 |       "outputs": []
144 |     },
145 |     {
146 |       "cell_type": "markdown",
147 |       "metadata": {
148 |         "id": "Y_7uABuIpBEV",
149 |         "colab_type": "text"
150 |       },
151 |       "source": [
152 |         "###We can see an example"
153 |       ]
154 |     },
155 |     {
156 |       "cell_type": "code",
157 |       "metadata": {
158 |         "id": "9ctTGeODlvt6",
159 |         "colab_type": "code",
160 |         "outputId": "49b94b1b-a1d9-4886-f132-f5abcf71a0a3",
161 |         "colab": {
162 |           "base_uri": "https://localhost:8080/",
163 |           "height": 238
164 |         }
165 |       },
166 |       "source": [
167 |         "f, ax = plt.subplots(1,2)\n",
168 |         "ax[0].imshow(train_images[40])\n",
169 |         "ax[1].imshow(train_images[50])\n",
170 |         "print(train_labels[40],train_labels[50])\n",
171 |         "print(class_names[6],class_names[6])"
172 |       ],
173 |       "execution_count": 0,
174 |       "outputs": [
175 |         {
176 |           "output_type": "stream",
177 |           "text": [
178 |             "6 3\n",
179 |             "Shirt Shirt\n"
180 |           ],
181 |           "name": "stdout"
182 |         },
183 |         {
184 |           "output_type": "display_data",
185 |           "data": {
186 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAC4CAYAAAD61bdSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
187 |             "text/plain": [
188 |               "<Figure size 432x288 with 2 Axes>"
189 |             ]
190 |           },
191 |           "metadata": {
192 |             "tags": []
193 |           }
194 |         }
195 |       ]
196 |     },
197 |     {
198 |       "cell_type": "markdown",
199 |       "metadata": {
200 |         "id": "84j2r_8qle4g",
201 |         "colab_type": "text"
202 |       },
203 |       "source": [
204 |         "###Next, we normalise data"
205 |       ]
206 |     },
207 |     {
208 |       "cell_type": "code",
209 |       "metadata": {
210 |         "id": "ABSLwCLslpkI",
211 |         "colab_type": "code",
212 |         "colab": {}
213 |       },
214 |       "source": [
215 |         "train_images = train_images / 255.0\n",
216 |         "test_images = test_images / 255.0\n"
217 |       ],
218 |       "execution_count": 0,
219 |       "outputs": []
220 |     },
221 |     {
222 |       "cell_type": "markdown",
223 |       "metadata": {
224 |         "id": "1P3gPj5TnWvI",
225 |         "colab_type": "text"
226 |       },
227 |       "source": [
228 |         "#Simple neural net"
229 |       ]
230 |     },
231 |     {
232 |       "cell_type": "code",
233 |       "metadata": {
234 |         "id": "EnO6W_uendrN",
235 |         "colab_type": "code",
236 |         "colab": {}
237 |       },
238 |       "source": [
239 |         "model = keras.Sequential([\n",
240 |         "    keras.layers.Flatten(),\n",
241 |         "    keras.layers.Dense(128, activation=tf.nn.relu),\n",
242 |         "    keras.layers.Dense(10, activation=tf.nn.softmax)\n",
243 |         "])"
244 |       ],
245 |       "execution_count": 0,
246 |       "outputs": []
247 |     },
248 |     {
249 |       "cell_type": "markdown",
250 |       "metadata": {
251 |         "id": "f4ZajpTGnkLd",
252 |         "colab_type": "text"
253 |       },
254 |       "source": [
255 |         "###Next, compile the model"
256 |       ]
257 |     },
258 |     {
259 |       "cell_type": "code",
260 |       "metadata": {
261 |         "id": "cs-CinJAnjxy",
262 |         "colab_type": "code",
263 |         "colab": {}
264 |       },
265 |       "source": [
266 |         "model.compile(optimizer='adam', \n",
267 |         "              loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
268 |       ],
269 |       "execution_count": 0,
270 |       "outputs": []
271 |     },
272 |     {
273 |       "cell_type": "markdown",
274 |       "metadata": {
275 |         "id": "F-Cd-ptRns1w",
276 |         "colab_type": "text"
277 |       },
278 |       "source": [
279 |         "###Train model"
280 |       ]
281 |     },
282 |     {
283 |       "cell_type": "code",
284 |       "metadata": {
285 |         "id": "8bfZQh8vnujg",
286 |         "colab_type": "code",
287 |         "outputId": "cea87abe-2111-4981-ee8d-69baf82ca6be",
288 |         "colab": {
289 |           "base_uri": "https://localhost:8080/",
290 |           "height": 421
291 |         }
292 |       },
293 |       "source": [
294 |         "model.fit(train_images, train_labels, epochs=10)"
295 |       ],
296 |       "execution_count": 0,
297 |       "outputs": [
298 |         {
299 |           "output_type": "stream",
300 |           "text": [
301 |             "Train on 60000 samples\n",
302 |             "Epoch 1/10\n",
303 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1739 - acc: 0.9341\n",
304 |             "Epoch 2/10\n",
305 |             "60000/60000 [==============================] - 4s 67us/sample - loss: 0.1711 - acc: 0.9362\n",
306 |             "Epoch 3/10\n",
307 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1671 - acc: 0.9366\n",
308 |             "Epoch 4/10\n",
309 |             "60000/60000 [==============================] - 4s 69us/sample - loss: 0.1649 - acc: 0.9383\n",
310 |             "Epoch 5/10\n",
311 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1578 - acc: 0.9410\n",
312 |             "Epoch 6/10\n",
313 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1566 - acc: 0.9410\n",
314 |             "Epoch 7/10\n",
315 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1504 - acc: 0.9447\n",
316 |             "Epoch 8/10\n",
317 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1476 - acc: 0.9449\n",
318 |             "Epoch 9/10\n",
319 |             "60000/60000 [==============================] - 4s 68us/sample - loss: 0.1454 - acc: 0.9447\n",
320 |             "Epoch 10/10\n",
321 |             "60000/60000 [==============================] - 4s 67us/sample - loss: 0.1442 - acc: 0.9456\n"
322 |           ],
323 |           "name": "stdout"
324 |         },
325 |         {
326 |           "output_type": "execute_result",
327 |           "data": {
328 |             "text/plain": [
329 |               "<tensorflow.python.keras.callbacks.History at 0x7f5725d56278>"
330 |             ]
331 |           },
332 |           "metadata": {
333 |             "tags": []
334 |           },
335 |           "execution_count": 32
336 |         }
337 |       ]
338 |     },
339 |     {
340 |       "cell_type": "markdown",
341 |       "metadata": {
342 |         "id": "Lb0BXV-6n7aJ",
343 |         "colab_type": "text"
344 |       },
345 |       "source": [
346 |         "###Predict class of second image in test dataset"
347 |       ]
348 |     },
349 |     {
350 |       "cell_type": "code",
351 |       "metadata": {
352 |         "id": "s8jsni0bn_f1",
353 |         "colab_type": "code",
354 |         "outputId": "12838771-6576-4398-d046-5305204fd42e",
355 |         "colab": {
356 |           "base_uri": "https://localhost:8080/",
357 |           "height": 90
358 |         }
359 |       },
360 |       "source": [
361 |         "prediction=model.predict(test_images[1].reshape(1, 28, 28))\n",
362 |         "print(\"Probabilities of image in each class are\",prediction)\n",
363 |         "print(\"Highest probability in place:\", prediction.argmax())\n",
364 |         "print(\"Image is classified as a: \",class_names[prediction.argmax()])"
365 |       ],
366 |       "execution_count": 0,
367 |       "outputs": [
368 |         {
369 |           "output_type": "stream",
370 |           "text": [
371 |             "Probabilities of image in each class are [[1.4792593e-05 7.4572272e-15 9.9984133e-01 1.0109936e-13 9.9008597e-05\n",
372 |             "  8.3782468e-15 4.4799071e-05 7.3859103e-23 1.0118387e-09 6.3930431e-16]]\n",
373 |             "Highest probability in place: 2\n",
374 |             "Image is classified as a:  Pullover\n"
375 |           ],
376 |           "name": "stdout"
377 |         }
378 |       ]
379 |     },
380 |     {
381 |       "cell_type": "code",
382 |       "metadata": {
383 |         "id": "zslaJQ7VoZDu",
384 |         "colab_type": "code",
385 |         "outputId": "d103fa5e-cecf-430c-f94d-42962e2a8c61",
386 |         "colab": {
387 |           "base_uri": "https://localhost:8080/",
388 |           "height": 320
389 |         }
390 |       },
391 |       "source": [
392 |         "print(\"Actual label number is:\",test_labels[1])\n",
393 |         "print(\"Image is a: \",class_names[test_labels[1]])\n",
394 |         "plt.imshow(test_images[1])\n"
395 |       ],
396 |       "execution_count": 0,
397 |       "outputs": [
398 |         {
399 |           "output_type": "stream",
400 |           "text": [
401 |             "Actual label number is: 2\n",
402 |             "Image is a:  Pullover\n"
403 |           ],
404 |           "name": "stdout"
405 |         },
406 |         {
407 |           "output_type": "execute_result",
408 |           "data": {
409 |             "text/plain": [
410 |               "<matplotlib.image.AxesImage at 0x7f5724a69278>"
411 |             ]
412 |           },
413 |           "metadata": {
414 |             "tags": []
415 |           },
416 |           "execution_count": 10
417 |         },
418 |         {
419 |           "output_type": "display_data",
420 |           "data": {
421 |             "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niZxg2ImcYNiJnGDYiZz4f4SixgHCORlKAAAAAElFTkSuQmCC\n",
422 |             "text/plain": [
423 |               "<Figure size 432x288 with 1 Axes>"
424 |             ]
425 |           },
426 |           "metadata": {
427 |             "tags": []
428 |           }
429 |         }
430 |       ]
431 |     },
432 |     {
433 |       "cell_type": "markdown",
434 |       "metadata": {
435 |         "id": "NnRT1sg7rQHi",
436 |         "colab_type": "text"
437 |       },
438 |       "source": [
439 |         "Second image is a pullover and it is classified correctly."
440 |       ]
441 |     },
442 |     {
443 |       "cell_type": "markdown",
444 |       "metadata": {
445 |         "id": "KaJlvr76sOtB",
446 |         "colab_type": "text"
447 |       },
448 |       "source": [
449 |         "###Evaluate models accuracy"
450 |       ]
451 |     },
452 |     {
453 |       "cell_type": "code",
454 |       "metadata": {
455 |         "id": "iyCJ4rDzsRx8",
456 |         "colab_type": "code",
457 |         "outputId": "210ea44e-d8f0-44a1-d3dc-2875c7f923fa",
458 |         "colab": {
459 |           "base_uri": "https://localhost:8080/",
460 |           "height": 54
461 |         }
462 |       },
463 |       "source": [
464 |         "test_acc = model.evaluate(test_images, test_labels)\n",
465 |         "print('Test loss, accuracy:', test_acc)\n"
466 |       ],
467 |       "execution_count": 0,
468 |       "outputs": [
469 |         {
470 |           "output_type": "stream",
471 |           "text": [
472 |             "10000/10000 [==============================] - 0s 32us/sample - loss: 0.4091 - acc: 0.8865\n",
473 |             "Test accuracy: [0.4091187917113304, 0.8865]\n"
474 |           ],
475 |           "name": "stdout"
476 |         }
477 |       ]
478 |     },
479 |     {
480 |       "cell_type": "markdown",
481 |       "metadata": {
482 |         "id": "Xnki0uOpxoKz",
483 |         "colab_type": "text"
484 |       },
485 |       "source": [
486 |         "The overal model's accuracy is very low. This in part because of the low number of training epochs. Next, we will create a convolution neural network and compare the performance of the two models."
487 |       ]
488 |     },
489 |     {
490 |       "cell_type": "markdown",
491 |       "metadata": {
492 |         "id": "Z_Eli9ZX0Yio",
493 |         "colab_type": "text"
494 |       },
495 |       "source": [
496 |         "## Convolution Neural Network"
497 |       ]
498 |     },
499 |     {
500 |       "cell_type": "code",
501 |       "metadata": {
502 |         "id": "15ZvqCtO1R39",
503 |         "colab_type": "code",
504 |         "colab": {}
505 |       },
506 |       "source": [
507 |         "cnn_train_images=train_images.reshape(60000, 28, 28, 1)\n",
508 |         "cnn_test_images = test_images.reshape(10000, 28, 28, 1)"
509 |       ],
510 |       "execution_count": 0,
511 |       "outputs": []
512 |     },
513 |     {
514 |       "cell_type": "code",
515 |       "metadata": {
516 |         "id": "tNfabnuzxTmz",
517 |         "colab_type": "code",
518 |         "colab": {}
519 |       },
520 |       "source": [
521 |         "cnn_model = tf.keras.models.Sequential([\n",
522 |         "  tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(28, 28,1)),\n",
523 |         "  tf.keras.layers.MaxPooling2D(2, 2),\n",
524 |         "  tf.keras.layers.Flatten(),\n",
525 |         "  keras.layers.Dense(128, activation=tf.nn.relu),\n",
526 |         "    keras.layers.Dense(10, activation=tf.nn.softmax)\n",
527 |         "])"
528 |       ],
529 |       "execution_count": 0,
530 |       "outputs": []
531 |     },
532 |     {
533 |       "cell_type": "code",
534 |       "metadata": {
535 |         "id": "3SfRYTO40tzN",
536 |         "colab_type": "code",
537 |         "colab": {}
538 |       },
539 |       "source": [
540 |         "cnn_model.compile(optimizer='adam', \n",
541 |         "              loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
542 |       ],
543 |       "execution_count": 0,
544 |       "outputs": []
545 |     },
546 |     {
547 |       "cell_type": "code",
548 |       "metadata": {
549 |         "id": "l7ASxW-70zIN",
550 |         "colab_type": "code",
551 |         "outputId": "c37abb7d-5419-40db-a6d0-82fdd4e0ab17",
552 |         "colab": {
553 |           "base_uri": "https://localhost:8080/",
554 |           "height": 421
555 |         }
556 |       },
557 |       "source": [
558 |         "cnn_model.fit(cnn_train_images, train_labels, epochs=10)"
559 |       ],
560 |       "execution_count": 0,
561 |       "outputs": [
562 |         {
563 |           "output_type": "stream",
564 |           "text": [
565 |             "Train on 60000 samples\n",
566 |             "Epoch 1/10\n",
567 |             "60000/60000 [==============================] - 36s 604us/sample - loss: 0.3938 - acc: 0.8609\n",
568 |             "Epoch 2/10\n",
569 |             "60000/60000 [==============================] - 36s 596us/sample - loss: 0.2643 - acc: 0.9043\n",
570 |             "Epoch 3/10\n",
571 |             "60000/60000 [==============================] - 36s 595us/sample - loss: 0.2214 - acc: 0.9185\n",
572 |             "Epoch 4/10\n",
573 |             "60000/60000 [==============================] - 36s 596us/sample - loss: 0.1910 - acc: 0.9294\n",
574 |             "Epoch 5/10\n",
575 |             "60000/60000 [==============================] - 36s 592us/sample - loss: 0.1613 - acc: 0.9407\n",
576 |             "Epoch 6/10\n",
577 |             "60000/60000 [==============================] - 35s 591us/sample - loss: 0.1386 - acc: 0.9488\n",
578 |             "Epoch 7/10\n",
579 |             "60000/60000 [==============================] - 35s 590us/sample - loss: 0.1174 - acc: 0.9567\n",
580 |             "Epoch 8/10\n",
581 |             "60000/60000 [==============================] - 36s 594us/sample - loss: 0.1005 - acc: 0.9628\n",
582 |             "Epoch 9/10\n",
583 |             "60000/60000 [==============================] - 36s 597us/sample - loss: 0.0848 - acc: 0.9688\n",
584 |             "Epoch 10/10\n",
585 |             "60000/60000 [==============================] - 36s 607us/sample - loss: 0.0727 - acc: 0.9743\n"
586 |           ],
587 |           "name": "stdout"
588 |         },
589 |         {
590 |           "output_type": "execute_result",
591 |           "data": {
592 |             "text/plain": [
593 |               "<tensorflow.python.keras.callbacks.History at 0x7f571c0796d8>"
594 |             ]
595 |           },
596 |           "metadata": {
597 |             "tags": []
598 |           },
599 |           "execution_count": 36
600 |         }
601 |       ]
602 |     },
603 |     {
604 |       "cell_type": "code",
605 |       "metadata": {
606 |         "id": "88D3VU7O672K",
607 |         "colab_type": "code",
608 |         "outputId": "db316b9c-dcfd-4285-d8b5-e824b0bc8c06",
609 |         "colab": {
610 |           "base_uri": "https://localhost:8080/",
611 |           "height": 54
612 |         }
613 |       },
614 |       "source": [
615 |         "test_cnn_acc = cnn_model.evaluate(cnn_test_images, test_labels)\n",
616 |         "print('Test loss, accuracy:', test_cnn_acc)\n"
617 |       ],
618 |       "execution_count": 0,
619 |       "outputs": [
620 |         {
621 |           "output_type": "stream",
622 |           "text": [
623 |             "10000/10000 [==============================] - 2s 200us/sample - loss: 0.3218 - acc: 0.9175\n",
624 |             "Test loss, accuracy: [0.32177444730997085, 0.9175]\n"
625 |           ],
626 |           "name": "stdout"
627 |         }
628 |       ]
629 |     }
630 |   ]
631 | }


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # Examples of Machine Learning and Artificial Inteligence in Python
 2 | 
 3 | This repository contains examples of Machine learining and AI in Python jupyter notebooks. In most cases, a link at the top of the top notebook opens the notebook in [Google's Colab](https://colab.research.google.com/notebooks/intro.ipynb). In that way it is easy for everyone to experiment with the code without  having to go through the pain of setting up a Python envoronment. If for some reason github fails to render the notebook (it happens from time to time), you can use [nbviewer](https://nbviewer.jupyter.org/).  
 4 | 
 5 | Below is a short description of the available notebooks.
 6 | 
 7 | * **[Customer_segmentation](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Customer_segmentation.ipynb)** Example of customer segmentation. A demonstration of hierarchical clustering. 
 8 | * **[Bank_marketing](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Bank_marketing.ipynb)** Example of classification in unbalanced datasets using logistic regression. SMOTE - Synthetic Minority Over-sampling Technique is used as an example of how to deal with the imbalance.
 9 | * **[Fraud_detection](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Fraud_detection.ipynb)** Example of Fraud detection. Detection of fraudulent credit card transactions. This set is also (highly) unbalanced. Logistic regresion with SMOTE and decission trees are demonstrated.
10 | * **[Fraud_detection_ensembles](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Fraud_detection_ensembles.ipynb)** A continuation of Fraud_detection above. Random forests, xgboost and ensebles are used.
11 | * **[Simple neural network example](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Simple%20neural%20network%20example.ipynb)** A simple neural network with tensorflow. Nothing special other than introducing basic Python/Tensorflow syntax.
12 | * **[Example of image classification with neural network](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Example%20of%20image%20classification%20with%20neural%20network.ipynb)** Example of classification of images taken from Fashion MNIST using neural nets.
13 | * **[Image_Classification_with_and_without_convlolution](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Image_Classification_with_and_without_convlolution.ipynb)** Image classification of Fashion MNIST images with neural nets and convolutional layers. There is also the excel file [convolution_example.xlsx](https://github.com/dpanagop/ML_and_AI_examples/blob/master/convolution_example.xlsx) which you can use to play with the notion of convolutions.
14 | * **[Credit_Fraud_detection_with_autoencoders](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Credit_Fraud_detection_with_autoencoders.ipynb)** Here we revisit Credit- Fraud Detection with the help of autoencoders, a special kind of neural networks. Two different methods are shown:
15 |   1. using the autoenconder's reconstruction error,
16 |   2. using the autenconder for mapping data in a vector space and using the vector distance.
17 |  * **[Style_transfer](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Style_transfer.ipynb)** demonstrates how one can use neural nets to transfer image style from one picture to an other. The nice thing with this one is that if you run it in Colab (better use Chrome as a browser), then you can transfer style to an image that you upload. It is Raymond Yuan's code. You can read his post in [medium](https://medium.com/tensorflow/neural-style-transfer-creating-art-with-deep-learning-using-tf-keras-and-eager-execution-7d541ac31398) for more info.
18 |  * **[NLP_example_clustering](https://github.com/dpanagop/ML_and_AI_examples/blob/master/NLP_example_clustering.ipynb)** Text classification using term frequency/inverse term frequency (TF-IDF) and k-means. Text classified is articles downloaded from wikipedia. A wordcloud is created for each cluster. (Note: It is best to view this one in nbviewer).
19 |   * **[Greek_NLP_example_clustering](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Greek_NLP_example_clustering.ipynb)** Like but for greek language. (Note: Is is best to view this one in nbviewer).
20 |  * **[Spacy_example](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Spacy_example.ipynb)** Simple SpaCy example.  (Note: It is best to view this one in nbviewer).
21 |  * **[Greek_Spacy](https://github.com/dpanagop/ML_and_AI_examples/blob/master/Greek_Spacy.ipynb)** Simple SpaCy example in Greek.  (Note: It is best to view this one in nbviewer). 
22 | 


--------------------------------------------------------------------------------
/Simple neural network example.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |   "nbformat": 4,
   3 |   "nbformat_minor": 0,
   4 |   "metadata": {
   5 |     "colab": {
   6 |       "name": "Untitled6.ipynb",
   7 |       "version": "0.3.2",
   8 |       "provenance": [],
   9 |       "include_colab_link": true
  10 |     },
  11 |     "kernelspec": {
  12 |       "name": "python3",
  13 |       "display_name": "Python 3"
  14 |     }
  15 |   },
  16 |   "cells": [
  17 |     {
  18 |       "cell_type": "markdown",
  19 |       "metadata": {
  20 |         "id": "view-in-github",
  21 |         "colab_type": "text"
  22 |       },
  23 |       "source": [
  24 |         "<a href=\"https://colab.research.google.com/github/dpanagop/ML_and_AI_examples/blob/master/Simple%20neural%20network%20example.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
  25 |       ]
  26 |     },
  27 |     {
  28 |       "cell_type": "markdown",
  29 |       "metadata": {
  30 |         "id": "cowX9NMVcf0V",
  31 |         "colab_type": "text"
  32 |       },
  33 |       "source": [
  34 |         "#Simple neural network example "
  35 |       ]
  36 |     },
  37 |     {
  38 |       "cell_type": "markdown",
  39 |       "metadata": {
  40 |         "id": "Oh-dLc9TdApN",
  41 |         "colab_type": "text"
  42 |       },
  43 |       "source": [
  44 |         "###First, we load the necessary libraries"
  45 |       ]
  46 |     },
  47 |     {
  48 |       "cell_type": "code",
  49 |       "metadata": {
  50 |         "id": "wVBEHrxTc0EQ",
  51 |         "colab_type": "code",
  52 |         "colab": {}
  53 |       },
  54 |       "source": [
  55 |         "import numpy as np\n",
  56 |         "import tensorflow as tf\n",
  57 |         "from tensorflow import keras"
  58 |       ],
  59 |       "execution_count": 0,
  60 |       "outputs": []
  61 |     },
  62 |     {
  63 |       "cell_type": "markdown",
  64 |       "metadata": {
  65 |         "id": "eSgP4smidKoK",
  66 |         "colab_type": "text"
  67 |       },
  68 |       "source": [
  69 |         "###Then, we create a simple neural network"
  70 |       ]
  71 |     },
  72 |     {
  73 |       "cell_type": "code",
  74 |       "metadata": {
  75 |         "id": "u1gTL06IegTe",
  76 |         "colab_type": "code",
  77 |         "outputId": "9b7af300-e2fa-43ca-b99f-ae6bfd932e34",
  78 |         "colab": {
  79 |           "base_uri": "https://localhost:8080/",
  80 |           "height": 110
  81 |         }
  82 |       },
  83 |       "source": [
  84 |         "model=keras.Sequential([keras.layers.Dense(units=1, input_shape=[1])])\n",
  85 |         "model.add(keras.layers.Dense(1))"
  86 |       ],
  87 |       "execution_count": 0,
  88 |       "outputs": [
  89 |         {
  90 |           "output_type": "stream",
  91 |           "text": [
  92 |             "WARNING: Logging before flag parsing goes to stderr.\n",
  93 |             "W0716 11:11:20.355340 140689434380160 deprecation.py:506] From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/init_ops.py:1251: calling VarianceScaling.__init__ (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.\n",
  94 |             "Instructions for updating:\n",
  95 |             "Call initializer instance with the dtype argument instead of passing it to the constructor\n"
  96 |           ],
  97 |           "name": "stderr"
  98 |         }
  99 |       ]
 100 |     },
 101 |     {
 102 |       "cell_type": "markdown",
 103 |       "metadata": {
 104 |         "id": "aUHRqW64ewY3",
 105 |         "colab_type": "text"
 106 |       },
 107 |       "source": [
 108 |         "###Next, compile the model"
 109 |       ]
 110 |     },
 111 |     {
 112 |       "cell_type": "code",
 113 |       "metadata": {
 114 |         "id": "gltf5zKeeswG",
 115 |         "colab_type": "code",
 116 |         "colab": {}
 117 |       },
 118 |       "source": [
 119 |         "model.compile(optimizer='sgd', loss='mean_squared_error')"
 120 |       ],
 121 |       "execution_count": 0,
 122 |       "outputs": []
 123 |     },
 124 |     {
 125 |       "cell_type": "markdown",
 126 |       "metadata": {
 127 |         "id": "bMW3iNWqe5dw",
 128 |         "colab_type": "text"
 129 |       },
 130 |       "source": [
 131 |         "###Create \n",
 132 |         "1. input data, variable x\n",
 133 |         "2. target data, variable y\n",
 134 |         "\n",
 135 |         "Note that: $y=2x-1$"
 136 |       ]
 137 |     },
 138 |     {
 139 |       "cell_type": "code",
 140 |       "metadata": {
 141 |         "id": "UtH9Nwj3e1_8",
 142 |         "colab_type": "code",
 143 |         "colab": {}
 144 |       },
 145 |       "source": [
 146 |         "x=np.array([-1, 0,1,2,3,4],dtype=float)\n",
 147 |         "y=np.array([-3,-1,1,3,5,7],dtype=float)"
 148 |       ],
 149 |       "execution_count": 0,
 150 |       "outputs": []
 151 |     },
 152 |     {
 153 |       "cell_type": "markdown",
 154 |       "metadata": {
 155 |         "id": "Lnw1qLqWfVFb",
 156 |         "colab_type": "text"
 157 |       },
 158 |       "source": [
 159 |         "###Train model"
 160 |       ]
 161 |     },
 162 |     {
 163 |       "cell_type": "code",
 164 |       "metadata": {
 165 |         "id": "TD1rWVBBfYY0",
 166 |         "colab_type": "code",
 167 |         "outputId": "8802dd8f-3bf0-4f9f-c787-b3206f2d66ee",
 168 |         "colab": {
 169 |           "base_uri": "https://localhost:8080/",
 170 |           "height": 1000
 171 |         }
 172 |       },
 173 |       "source": [
 174 |         "model.fit(x,y,epochs=500)"
 175 |       ],
 176 |       "execution_count": 0,
 177 |       "outputs": [
 178 |         {
 179 |           "output_type": "stream",
 180 |           "text": [
 181 |             "Epoch 1/500\n",
 182 |             "6/6 [==============================] - 1s 83ms/sample - loss: 26.0347\n",
 183 |             "Epoch 2/500\n",
 184 |             "6/6 [==============================] - 0s 774us/sample - loss: 19.9218\n",
 185 |             "Epoch 3/500\n",
 186 |             "6/6 [==============================] - 0s 615us/sample - loss: 16.5734\n",
 187 |             "Epoch 4/500\n",
 188 |             "6/6 [==============================] - 0s 581us/sample - loss: 14.2282\n",
 189 |             "Epoch 5/500\n",
 190 |             "6/6 [==============================] - 0s 558us/sample - loss: 12.2851\n",
 191 |             "Epoch 6/500\n",
 192 |             "6/6 [==============================] - 0s 503us/sample - loss: 10.4984\n",
 193 |             "Epoch 7/500\n",
 194 |             "6/6 [==============================] - 0s 600us/sample - loss: 8.7823\n",
 195 |             "Epoch 8/500\n",
 196 |             "6/6 [==============================] - 0s 632us/sample - loss: 7.1440\n",
 197 |             "Epoch 9/500\n",
 198 |             "6/6 [==============================] - 0s 207us/sample - loss: 5.6455\n",
 199 |             "Epoch 10/500\n",
 200 |             "6/6 [==============================] - 0s 209us/sample - loss: 4.3646\n",
 201 |             "Epoch 11/500\n",
 202 |             "6/6 [==============================] - 0s 210us/sample - loss: 3.3553\n",
 203 |             "Epoch 12/500\n",
 204 |             "6/6 [==============================] - 0s 364us/sample - loss: 2.6236\n",
 205 |             "Epoch 13/500\n",
 206 |             "6/6 [==============================] - 0s 324us/sample - loss: 2.1296\n",
 207 |             "Epoch 14/500\n",
 208 |             "6/6 [==============================] - 0s 338us/sample - loss: 1.8097\n",
 209 |             "Epoch 15/500\n",
 210 |             "6/6 [==============================] - 0s 365us/sample - loss: 1.6019\n",
 211 |             "Epoch 16/500\n",
 212 |             "6/6 [==============================] - 0s 189us/sample - loss: 1.4595\n",
 213 |             "Epoch 17/500\n",
 214 |             "6/6 [==============================] - 0s 416us/sample - loss: 1.3528\n",
 215 |             "Epoch 18/500\n",
 216 |             "6/6 [==============================] - 0s 191us/sample - loss: 1.2655\n",
 217 |             "Epoch 19/500\n",
 218 |             "6/6 [==============================] - 0s 251us/sample - loss: 1.1890\n",
 219 |             "Epoch 20/500\n",
 220 |             "6/6 [==============================] - 0s 243us/sample - loss: 1.1193\n",
 221 |             "Epoch 21/500\n",
 222 |             "6/6 [==============================] - 0s 599us/sample - loss: 1.0543\n",
 223 |             "Epoch 22/500\n",
 224 |             "6/6 [==============================] - 0s 234us/sample - loss: 0.9932\n",
 225 |             "Epoch 23/500\n",
 226 |             "6/6 [==============================] - 0s 441us/sample - loss: 0.9353\n",
 227 |             "Epoch 24/500\n",
 228 |             "6/6 [==============================] - 0s 200us/sample - loss: 0.8806\n",
 229 |             "Epoch 25/500\n",
 230 |             "6/6 [==============================] - 0s 215us/sample - loss: 0.8287\n",
 231 |             "Epoch 26/500\n",
 232 |             "6/6 [==============================] - 0s 308us/sample - loss: 0.7795\n",
 233 |             "Epoch 27/500\n",
 234 |             "6/6 [==============================] - 0s 277us/sample - loss: 0.7329\n",
 235 |             "Epoch 28/500\n",
 236 |             "6/6 [==============================] - 0s 277us/sample - loss: 0.6888\n",
 237 |             "Epoch 29/500\n",
 238 |             "6/6 [==============================] - 0s 256us/sample - loss: 0.6470\n",
 239 |             "Epoch 30/500\n",
 240 |             "6/6 [==============================] - 0s 281us/sample - loss: 0.6075\n",
 241 |             "Epoch 31/500\n",
 242 |             "6/6 [==============================] - 0s 365us/sample - loss: 0.5701\n",
 243 |             "Epoch 32/500\n",
 244 |             "6/6 [==============================] - 0s 387us/sample - loss: 0.5348\n",
 245 |             "Epoch 33/500\n",
 246 |             "6/6 [==============================] - 0s 506us/sample - loss: 0.5014\n",
 247 |             "Epoch 34/500\n",
 248 |             "6/6 [==============================] - 0s 536us/sample - loss: 0.4699\n",
 249 |             "Epoch 35/500\n",
 250 |             "6/6 [==============================] - 0s 315us/sample - loss: 0.4402\n",
 251 |             "Epoch 36/500\n",
 252 |             "6/6 [==============================] - 0s 384us/sample - loss: 0.4122\n",
 253 |             "Epoch 37/500\n",
 254 |             "6/6 [==============================] - 0s 544us/sample - loss: 0.3857\n",
 255 |             "Epoch 38/500\n",
 256 |             "6/6 [==============================] - 0s 442us/sample - loss: 0.3608\n",
 257 |             "Epoch 39/500\n",
 258 |             "6/6 [==============================] - 0s 311us/sample - loss: 0.3374\n",
 259 |             "Epoch 40/500\n",
 260 |             "6/6 [==============================] - 0s 291us/sample - loss: 0.3153\n",
 261 |             "Epoch 41/500\n",
 262 |             "6/6 [==============================] - 0s 480us/sample - loss: 0.2946\n",
 263 |             "Epoch 42/500\n",
 264 |             "6/6 [==============================] - 0s 665us/sample - loss: 0.2751\n",
 265 |             "Epoch 43/500\n",
 266 |             "6/6 [==============================] - 0s 347us/sample - loss: 0.2568\n",
 267 |             "Epoch 44/500\n",
 268 |             "6/6 [==============================] - 0s 218us/sample - loss: 0.2396\n",
 269 |             "Epoch 45/500\n",
 270 |             "6/6 [==============================] - 0s 213us/sample - loss: 0.2234\n",
 271 |             "Epoch 46/500\n",
 272 |             "6/6 [==============================] - 0s 283us/sample - loss: 0.2083\n",
 273 |             "Epoch 47/500\n",
 274 |             "6/6 [==============================] - 0s 309us/sample - loss: 0.1941\n",
 275 |             "Epoch 48/500\n",
 276 |             "6/6 [==============================] - 0s 300us/sample - loss: 0.1808\n",
 277 |             "Epoch 49/500\n",
 278 |             "6/6 [==============================] - 0s 252us/sample - loss: 0.1684\n",
 279 |             "Epoch 50/500\n",
 280 |             "6/6 [==============================] - 0s 284us/sample - loss: 0.1567\n",
 281 |             "Epoch 51/500\n",
 282 |             "6/6 [==============================] - 0s 385us/sample - loss: 0.1458\n",
 283 |             "Epoch 52/500\n",
 284 |             "6/6 [==============================] - 0s 279us/sample - loss: 0.1356\n",
 285 |             "Epoch 53/500\n",
 286 |             "6/6 [==============================] - 0s 291us/sample - loss: 0.1261\n",
 287 |             "Epoch 54/500\n",
 288 |             "6/6 [==============================] - 0s 292us/sample - loss: 0.1172\n",
 289 |             "Epoch 55/500\n",
 290 |             "6/6 [==============================] - 0s 257us/sample - loss: 0.1089\n",
 291 |             "Epoch 56/500\n",
 292 |             "6/6 [==============================] - 0s 315us/sample - loss: 0.1012\n",
 293 |             "Epoch 57/500\n",
 294 |             "6/6 [==============================] - 0s 281us/sample - loss: 0.0940\n",
 295 |             "Epoch 58/500\n",
 296 |             "6/6 [==============================] - 0s 228us/sample - loss: 0.0872\n",
 297 |             "Epoch 59/500\n",
 298 |             "6/6 [==============================] - 0s 510us/sample - loss: 0.0810\n",
 299 |             "Epoch 60/500\n",
 300 |             "6/6 [==============================] - 0s 284us/sample - loss: 0.0751\n",
 301 |             "Epoch 61/500\n",
 302 |             "6/6 [==============================] - 0s 470us/sample - loss: 0.0697\n",
 303 |             "Epoch 62/500\n",
 304 |             "6/6 [==============================] - 0s 328us/sample - loss: 0.0646\n",
 305 |             "Epoch 63/500\n",
 306 |             "6/6 [==============================] - 0s 1ms/sample - loss: 0.0599\n",
 307 |             "Epoch 64/500\n",
 308 |             "6/6 [==============================] - 0s 355us/sample - loss: 0.0555\n",
 309 |             "Epoch 65/500\n",
 310 |             "6/6 [==============================] - 0s 314us/sample - loss: 0.0514\n",
 311 |             "Epoch 66/500\n",
 312 |             "6/6 [==============================] - 0s 302us/sample - loss: 0.0476\n",
 313 |             "Epoch 67/500\n",
 314 |             "6/6 [==============================] - 0s 361us/sample - loss: 0.0441\n",
 315 |             "Epoch 68/500\n",
 316 |             "6/6 [==============================] - 0s 276us/sample - loss: 0.0408\n",
 317 |             "Epoch 69/500\n",
 318 |             "6/6 [==============================] - 0s 352us/sample - loss: 0.0378\n",
 319 |             "Epoch 70/500\n",
 320 |             "6/6 [==============================] - 0s 270us/sample - loss: 0.0350\n",
 321 |             "Epoch 71/500\n",
 322 |             "6/6 [==============================] - 0s 272us/sample - loss: 0.0324\n",
 323 |             "Epoch 72/500\n",
 324 |             "6/6 [==============================] - 0s 246us/sample - loss: 0.0299\n",
 325 |             "Epoch 73/500\n",
 326 |             "6/6 [==============================] - 0s 379us/sample - loss: 0.0277\n",
 327 |             "Epoch 74/500\n",
 328 |             "6/6 [==============================] - 0s 267us/sample - loss: 0.0256\n",
 329 |             "Epoch 75/500\n",
 330 |             "6/6 [==============================] - 0s 299us/sample - loss: 0.0237\n",
 331 |             "Epoch 76/500\n",
 332 |             "6/6 [==============================] - 0s 369us/sample - loss: 0.0219\n",
 333 |             "Epoch 77/500\n",
 334 |             "6/6 [==============================] - 0s 458us/sample - loss: 0.0202\n",
 335 |             "Epoch 78/500\n",
 336 |             "6/6 [==============================] - 0s 288us/sample - loss: 0.0187\n",
 337 |             "Epoch 79/500\n",
 338 |             "6/6 [==============================] - 0s 400us/sample - loss: 0.0173\n",
 339 |             "Epoch 80/500\n",
 340 |             "6/6 [==============================] - 0s 440us/sample - loss: 0.0159\n",
 341 |             "Epoch 81/500\n",
 342 |             "6/6 [==============================] - 0s 337us/sample - loss: 0.0147\n",
 343 |             "Epoch 82/500\n",
 344 |             "6/6 [==============================] - 0s 630us/sample - loss: 0.0136\n",
 345 |             "Epoch 83/500\n",
 346 |             "6/6 [==============================] - 0s 425us/sample - loss: 0.0126\n",
 347 |             "Epoch 84/500\n",
 348 |             "6/6 [==============================] - 0s 452us/sample - loss: 0.0116\n",
 349 |             "Epoch 85/500\n",
 350 |             "6/6 [==============================] - 0s 364us/sample - loss: 0.0107\n",
 351 |             "Epoch 86/500\n",
 352 |             "6/6 [==============================] - 0s 418us/sample - loss: 0.0099\n",
 353 |             "Epoch 87/500\n",
 354 |             "6/6 [==============================] - 0s 411us/sample - loss: 0.0091\n",
 355 |             "Epoch 88/500\n",
 356 |             "6/6 [==============================] - 0s 286us/sample - loss: 0.0084\n",
 357 |             "Epoch 89/500\n",
 358 |             "6/6 [==============================] - 0s 260us/sample - loss: 0.0078\n",
 359 |             "Epoch 90/500\n",
 360 |             "6/6 [==============================] - 0s 242us/sample - loss: 0.0072\n",
 361 |             "Epoch 91/500\n",
 362 |             "6/6 [==============================] - 0s 264us/sample - loss: 0.0066\n",
 363 |             "Epoch 92/500\n",
 364 |             "6/6 [==============================] - 0s 288us/sample - loss: 0.0061\n",
 365 |             "Epoch 93/500\n",
 366 |             "6/6 [==============================] - 0s 285us/sample - loss: 0.0056\n",
 367 |             "Epoch 94/500\n",
 368 |             "6/6 [==============================] - 0s 271us/sample - loss: 0.0052\n",
 369 |             "Epoch 95/500\n",
 370 |             "6/6 [==============================] - 0s 276us/sample - loss: 0.0048\n",
 371 |             "Epoch 96/500\n",
 372 |             "6/6 [==============================] - 0s 265us/sample - loss: 0.0044\n",
 373 |             "Epoch 97/500\n",
 374 |             "6/6 [==============================] - 0s 262us/sample - loss: 0.0041\n",
 375 |             "Epoch 98/500\n",
 376 |             "6/6 [==============================] - 0s 259us/sample - loss: 0.0037\n",
 377 |             "Epoch 99/500\n",
 378 |             "6/6 [==============================] - 0s 298us/sample - loss: 0.0034\n",
 379 |             "Epoch 100/500\n",
 380 |             "6/6 [==============================] - 0s 298us/sample - loss: 0.0032\n",
 381 |             "Epoch 101/500\n",
 382 |             "6/6 [==============================] - 0s 263us/sample - loss: 0.0029\n",
 383 |             "Epoch 102/500\n",
 384 |             "6/6 [==============================] - 0s 269us/sample - loss: 0.0027\n",
 385 |             "Epoch 103/500\n",
 386 |             "6/6 [==============================] - 0s 272us/sample - loss: 0.0025\n",
 387 |             "Epoch 104/500\n",
 388 |             "6/6 [==============================] - 0s 378us/sample - loss: 0.0023\n",
 389 |             "Epoch 105/500\n",
 390 |             "6/6 [==============================] - 0s 310us/sample - loss: 0.0021\n",
 391 |             "Epoch 106/500\n",
 392 |             "6/6 [==============================] - 0s 270us/sample - loss: 0.0019\n",
 393 |             "Epoch 107/500\n",
 394 |             "6/6 [==============================] - 0s 428us/sample - loss: 0.0018\n",
 395 |             "Epoch 108/500\n",
 396 |             "6/6 [==============================] - 0s 240us/sample - loss: 0.0017\n",
 397 |             "Epoch 109/500\n",
 398 |             "6/6 [==============================] - 0s 233us/sample - loss: 0.0015\n",
 399 |             "Epoch 110/500\n",
 400 |             "6/6 [==============================] - 0s 234us/sample - loss: 0.0014\n",
 401 |             "Epoch 111/500\n",
 402 |             "6/6 [==============================] - 0s 344us/sample - loss: 0.0013\n",
 403 |             "Epoch 112/500\n",
 404 |             "6/6 [==============================] - 0s 277us/sample - loss: 0.0012\n",
 405 |             "Epoch 113/500\n",
 406 |             "6/6 [==============================] - 0s 279us/sample - loss: 0.0011\n",
 407 |             "Epoch 114/500\n",
 408 |             "6/6 [==============================] - 0s 235us/sample - loss: 0.0010\n",
 409 |             "Epoch 115/500\n",
 410 |             "6/6 [==============================] - 0s 263us/sample - loss: 9.2830e-04\n",
 411 |             "Epoch 116/500\n",
 412 |             "6/6 [==============================] - 0s 225us/sample - loss: 8.5486e-04\n",
 413 |             "Epoch 117/500\n",
 414 |             "6/6 [==============================] - 0s 222us/sample - loss: 7.8720e-04\n",
 415 |             "Epoch 118/500\n",
 416 |             "6/6 [==============================] - 0s 303us/sample - loss: 7.2487e-04\n",
 417 |             "Epoch 119/500\n",
 418 |             "6/6 [==============================] - 0s 276us/sample - loss: 6.6744e-04\n",
 419 |             "Epoch 120/500\n",
 420 |             "6/6 [==============================] - 0s 222us/sample - loss: 6.1455e-04\n",
 421 |             "Epoch 121/500\n",
 422 |             "6/6 [==============================] - 0s 274us/sample - loss: 5.6583e-04\n",
 423 |             "Epoch 122/500\n",
 424 |             "6/6 [==============================] - 0s 427us/sample - loss: 5.2095e-04\n",
 425 |             "Epoch 123/500\n",
 426 |             "6/6 [==============================] - 0s 307us/sample - loss: 4.7963e-04\n",
 427 |             "Epoch 124/500\n",
 428 |             "6/6 [==============================] - 0s 225us/sample - loss: 4.4156e-04\n",
 429 |             "Epoch 125/500\n",
 430 |             "6/6 [==============================] - 0s 309us/sample - loss: 4.0651e-04\n",
 431 |             "Epoch 126/500\n",
 432 |             "6/6 [==============================] - 0s 234us/sample - loss: 3.7423e-04\n",
 433 |             "Epoch 127/500\n",
 434 |             "6/6 [==============================] - 0s 204us/sample - loss: 3.4450e-04\n",
 435 |             "Epoch 128/500\n",
 436 |             "6/6 [==============================] - 0s 288us/sample - loss: 3.1713e-04\n",
 437 |             "Epoch 129/500\n",
 438 |             "6/6 [==============================] - 0s 1ms/sample - loss: 2.9193e-04\n",
 439 |             "Epoch 130/500\n",
 440 |             "6/6 [==============================] - 0s 239us/sample - loss: 2.6872e-04\n",
 441 |             "Epoch 131/500\n",
 442 |             "6/6 [==============================] - 0s 238us/sample - loss: 2.4735e-04\n",
 443 |             "Epoch 132/500\n",
 444 |             "6/6 [==============================] - 0s 226us/sample - loss: 2.2768e-04\n",
 445 |             "Epoch 133/500\n",
 446 |             "6/6 [==============================] - 0s 213us/sample - loss: 2.0956e-04\n",
 447 |             "Epoch 134/500\n",
 448 |             "6/6 [==============================] - 0s 222us/sample - loss: 1.9289e-04\n",
 449 |             "Epoch 135/500\n",
 450 |             "6/6 [==============================] - 0s 277us/sample - loss: 1.7754e-04\n",
 451 |             "Epoch 136/500\n",
 452 |             "6/6 [==============================] - 0s 223us/sample - loss: 1.6340e-04\n",
 453 |             "Epoch 137/500\n",
 454 |             "6/6 [==============================] - 0s 224us/sample - loss: 1.5039e-04\n",
 455 |             "Epoch 138/500\n",
 456 |             "6/6 [==============================] - 0s 205us/sample - loss: 1.3842e-04\n",
 457 |             "Epoch 139/500\n",
 458 |             "6/6 [==============================] - 0s 250us/sample - loss: 1.2739e-04\n",
 459 |             "Epoch 140/500\n",
 460 |             "6/6 [==============================] - 0s 275us/sample - loss: 1.1724e-04\n",
 461 |             "Epoch 141/500\n",
 462 |             "6/6 [==============================] - 0s 191us/sample - loss: 1.0790e-04\n",
 463 |             "Epoch 142/500\n",
 464 |             "6/6 [==============================] - 0s 197us/sample - loss: 9.9301e-05\n",
 465 |             "Epoch 143/500\n",
 466 |             "6/6 [==============================] - 0s 191us/sample - loss: 9.1387e-05\n",
 467 |             "Epoch 144/500\n",
 468 |             "6/6 [==============================] - 0s 194us/sample - loss: 8.4100e-05\n",
 469 |             "Epoch 145/500\n",
 470 |             "6/6 [==============================] - 0s 193us/sample - loss: 7.7397e-05\n",
 471 |             "Epoch 146/500\n",
 472 |             "6/6 [==============================] - 0s 234us/sample - loss: 7.1224e-05\n",
 473 |             "Epoch 147/500\n",
 474 |             "6/6 [==============================] - 0s 194us/sample - loss: 6.5544e-05\n",
 475 |             "Epoch 148/500\n",
 476 |             "6/6 [==============================] - 0s 182us/sample - loss: 6.0316e-05\n",
 477 |             "Epoch 149/500\n",
 478 |             "6/6 [==============================] - 0s 194us/sample - loss: 5.5503e-05\n",
 479 |             "Epoch 150/500\n",
 480 |             "6/6 [==============================] - 0s 196us/sample - loss: 5.1077e-05\n",
 481 |             "Epoch 151/500\n",
 482 |             "6/6 [==============================] - 0s 199us/sample - loss: 4.7001e-05\n",
 483 |             "Epoch 152/500\n",
 484 |             "6/6 [==============================] - 0s 196us/sample - loss: 4.3251e-05\n",
 485 |             "Epoch 153/500\n",
 486 |             "6/6 [==============================] - 0s 195us/sample - loss: 3.9799e-05\n",
 487 |             "Epoch 154/500\n",
 488 |             "6/6 [==============================] - 0s 195us/sample - loss: 3.6622e-05\n",
 489 |             "Epoch 155/500\n",
 490 |             "6/6 [==============================] - 0s 204us/sample - loss: 3.3699e-05\n",
 491 |             "Epoch 156/500\n",
 492 |             "6/6 [==============================] - 0s 192us/sample - loss: 3.1010e-05\n",
 493 |             "Epoch 157/500\n",
 494 |             "6/6 [==============================] - 0s 200us/sample - loss: 2.8533e-05\n",
 495 |             "Epoch 158/500\n",
 496 |             "6/6 [==============================] - 0s 160us/sample - loss: 2.6256e-05\n",
 497 |             "Epoch 159/500\n",
 498 |             "6/6 [==============================] - 0s 254us/sample - loss: 2.4159e-05\n",
 499 |             "Epoch 160/500\n",
 500 |             "6/6 [==============================] - 0s 193us/sample - loss: 2.2231e-05\n",
 501 |             "Epoch 161/500\n",
 502 |             "6/6 [==============================] - 0s 188us/sample - loss: 2.0455e-05\n",
 503 |             "Epoch 162/500\n",
 504 |             "6/6 [==============================] - 0s 267us/sample - loss: 1.8822e-05\n",
 505 |             "Epoch 163/500\n",
 506 |             "6/6 [==============================] - 0s 199us/sample - loss: 1.7319e-05\n",
 507 |             "Epoch 164/500\n",
 508 |             "6/6 [==============================] - 0s 163us/sample - loss: 1.5935e-05\n",
 509 |             "Epoch 165/500\n",
 510 |             "6/6 [==============================] - 0s 315us/sample - loss: 1.4662e-05\n",
 511 |             "Epoch 166/500\n",
 512 |             "6/6 [==============================] - 0s 256us/sample - loss: 1.3491e-05\n",
 513 |             "Epoch 167/500\n",
 514 |             "6/6 [==============================] - 0s 202us/sample - loss: 1.2413e-05\n",
 515 |             "Epoch 168/500\n",
 516 |             "6/6 [==============================] - 0s 205us/sample - loss: 1.1422e-05\n",
 517 |             "Epoch 169/500\n",
 518 |             "6/6 [==============================] - 0s 169us/sample - loss: 1.0509e-05\n",
 519 |             "Epoch 170/500\n",
 520 |             "6/6 [==============================] - 0s 168us/sample - loss: 9.6697e-06\n",
 521 |             "Epoch 171/500\n",
 522 |             "6/6 [==============================] - 0s 201us/sample - loss: 8.8971e-06\n",
 523 |             "Epoch 172/500\n",
 524 |             "6/6 [==============================] - 0s 196us/sample - loss: 8.1860e-06\n",
 525 |             "Epoch 173/500\n",
 526 |             "6/6 [==============================] - 0s 294us/sample - loss: 7.5324e-06\n",
 527 |             "Epoch 174/500\n",
 528 |             "6/6 [==============================] - 0s 196us/sample - loss: 6.9304e-06\n",
 529 |             "Epoch 175/500\n",
 530 |             "6/6 [==============================] - 0s 227us/sample - loss: 6.3762e-06\n",
 531 |             "Epoch 176/500\n",
 532 |             "6/6 [==============================] - 0s 197us/sample - loss: 5.8664e-06\n",
 533 |             "Epoch 177/500\n",
 534 |             "6/6 [==============================] - 0s 203us/sample - loss: 5.3972e-06\n",
 535 |             "Epoch 178/500\n",
 536 |             "6/6 [==============================] - 0s 233us/sample - loss: 4.9662e-06\n",
 537 |             "Epoch 179/500\n",
 538 |             "6/6 [==============================] - 0s 207us/sample - loss: 4.5692e-06\n",
 539 |             "Epoch 180/500\n",
 540 |             "6/6 [==============================] - 0s 195us/sample - loss: 4.2041e-06\n",
 541 |             "Epoch 181/500\n",
 542 |             "6/6 [==============================] - 0s 208us/sample - loss: 3.8676e-06\n",
 543 |             "Epoch 182/500\n",
 544 |             "6/6 [==============================] - 0s 204us/sample - loss: 3.5589e-06\n",
 545 |             "Epoch 183/500\n",
 546 |             "6/6 [==============================] - 0s 272us/sample - loss: 3.2744e-06\n",
 547 |             "Epoch 184/500\n",
 548 |             "6/6 [==============================] - 0s 270us/sample - loss: 3.0123e-06\n",
 549 |             "Epoch 185/500\n",
 550 |             "6/6 [==============================] - 0s 173us/sample - loss: 2.7716e-06\n",
 551 |             "Epoch 186/500\n",
 552 |             "6/6 [==============================] - 0s 200us/sample - loss: 2.5499e-06\n",
 553 |             "Epoch 187/500\n",
 554 |             "6/6 [==============================] - 0s 205us/sample - loss: 2.3458e-06\n",
 555 |             "Epoch 188/500\n",
 556 |             "6/6 [==============================] - 0s 198us/sample - loss: 2.1582e-06\n",
 557 |             "Epoch 189/500\n",
 558 |             "6/6 [==============================] - 0s 194us/sample - loss: 1.9860e-06\n",
 559 |             "Epoch 190/500\n",
 560 |             "6/6 [==============================] - 0s 166us/sample - loss: 1.8270e-06\n",
 561 |             "Epoch 191/500\n",
 562 |             "6/6 [==============================] - 0s 165us/sample - loss: 1.6809e-06\n",
 563 |             "Epoch 192/500\n",
 564 |             "6/6 [==============================] - 0s 173us/sample - loss: 1.5465e-06\n",
 565 |             "Epoch 193/500\n",
 566 |             "6/6 [==============================] - 0s 290us/sample - loss: 1.4232e-06\n",
 567 |             "Epoch 194/500\n",
 568 |             "6/6 [==============================] - 0s 205us/sample - loss: 1.3093e-06\n",
 569 |             "Epoch 195/500\n",
 570 |             "6/6 [==============================] - 0s 198us/sample - loss: 1.2047e-06\n",
 571 |             "Epoch 196/500\n",
 572 |             "6/6 [==============================] - 0s 203us/sample - loss: 1.1081e-06\n",
 573 |             "Epoch 197/500\n",
 574 |             "6/6 [==============================] - 0s 206us/sample - loss: 1.0197e-06\n",
 575 |             "Epoch 198/500\n",
 576 |             "6/6 [==============================] - 0s 260us/sample - loss: 9.3804e-07\n",
 577 |             "Epoch 199/500\n",
 578 |             "6/6 [==============================] - 0s 196us/sample - loss: 8.6302e-07\n",
 579 |             "Epoch 200/500\n",
 580 |             "6/6 [==============================] - 0s 277us/sample - loss: 7.9411e-07\n",
 581 |             "Epoch 201/500\n",
 582 |             "6/6 [==============================] - 0s 297us/sample - loss: 7.3066e-07\n",
 583 |             "Epoch 202/500\n",
 584 |             "6/6 [==============================] - 0s 261us/sample - loss: 6.7218e-07\n",
 585 |             "Epoch 203/500\n",
 586 |             "6/6 [==============================] - 0s 270us/sample - loss: 6.1840e-07\n",
 587 |             "Epoch 204/500\n",
 588 |             "6/6 [==============================] - 0s 191us/sample - loss: 5.6893e-07\n",
 589 |             "Epoch 205/500\n",
 590 |             "6/6 [==============================] - 0s 202us/sample - loss: 5.2350e-07\n",
 591 |             "Epoch 206/500\n",
 592 |             "6/6 [==============================] - 0s 185us/sample - loss: 4.8157e-07\n",
 593 |             "Epoch 207/500\n",
 594 |             "6/6 [==============================] - 0s 197us/sample - loss: 4.4306e-07\n",
 595 |             "Epoch 208/500\n",
 596 |             "6/6 [==============================] - 0s 168us/sample - loss: 4.0759e-07\n",
 597 |             "Epoch 209/500\n",
 598 |             "6/6 [==============================] - 0s 265us/sample - loss: 3.7508e-07\n",
 599 |             "Epoch 210/500\n",
 600 |             "6/6 [==============================] - 0s 267us/sample - loss: 3.4518e-07\n",
 601 |             "Epoch 211/500\n",
 602 |             "6/6 [==============================] - 0s 222us/sample - loss: 3.1746e-07\n",
 603 |             "Epoch 212/500\n",
 604 |             "6/6 [==============================] - 0s 262us/sample - loss: 2.9209e-07\n",
 605 |             "Epoch 213/500\n",
 606 |             "6/6 [==============================] - 0s 195us/sample - loss: 2.6875e-07\n",
 607 |             "Epoch 214/500\n",
 608 |             "6/6 [==============================] - 0s 271us/sample - loss: 2.4723e-07\n",
 609 |             "Epoch 215/500\n",
 610 |             "6/6 [==============================] - 0s 252us/sample - loss: 2.2746e-07\n",
 611 |             "Epoch 216/500\n",
 612 |             "6/6 [==============================] - 0s 258us/sample - loss: 2.0927e-07\n",
 613 |             "Epoch 217/500\n",
 614 |             "6/6 [==============================] - 0s 269us/sample - loss: 1.9251e-07\n",
 615 |             "Epoch 218/500\n",
 616 |             "6/6 [==============================] - 0s 166us/sample - loss: 1.7710e-07\n",
 617 |             "Epoch 219/500\n",
 618 |             "6/6 [==============================] - 0s 317us/sample - loss: 1.6291e-07\n",
 619 |             "Epoch 220/500\n",
 620 |             "6/6 [==============================] - 0s 201us/sample - loss: 1.4988e-07\n",
 621 |             "Epoch 221/500\n",
 622 |             "6/6 [==============================] - 0s 168us/sample - loss: 1.3789e-07\n",
 623 |             "Epoch 222/500\n",
 624 |             "6/6 [==============================] - 0s 167us/sample - loss: 1.2688e-07\n",
 625 |             "Epoch 223/500\n",
 626 |             "6/6 [==============================] - 0s 250us/sample - loss: 1.1677e-07\n",
 627 |             "Epoch 224/500\n",
 628 |             "6/6 [==============================] - 0s 172us/sample - loss: 1.0739e-07\n",
 629 |             "Epoch 225/500\n",
 630 |             "6/6 [==============================] - 0s 196us/sample - loss: 9.8760e-08\n",
 631 |             "Epoch 226/500\n",
 632 |             "6/6 [==============================] - 0s 301us/sample - loss: 9.0924e-08\n",
 633 |             "Epoch 227/500\n",
 634 |             "6/6 [==============================] - 0s 202us/sample - loss: 8.3661e-08\n",
 635 |             "Epoch 228/500\n",
 636 |             "6/6 [==============================] - 0s 294us/sample - loss: 7.6935e-08\n",
 637 |             "Epoch 229/500\n",
 638 |             "6/6 [==============================] - 0s 190us/sample - loss: 7.0760e-08\n",
 639 |             "Epoch 230/500\n",
 640 |             "6/6 [==============================] - 0s 257us/sample - loss: 6.5102e-08\n",
 641 |             "Epoch 231/500\n",
 642 |             "6/6 [==============================] - 0s 248us/sample - loss: 5.9931e-08\n",
 643 |             "Epoch 232/500\n",
 644 |             "6/6 [==============================] - 0s 272us/sample - loss: 5.5135e-08\n",
 645 |             "Epoch 233/500\n",
 646 |             "6/6 [==============================] - 0s 197us/sample - loss: 5.0742e-08\n",
 647 |             "Epoch 234/500\n",
 648 |             "6/6 [==============================] - 0s 201us/sample - loss: 4.6624e-08\n",
 649 |             "Epoch 235/500\n",
 650 |             "6/6 [==============================] - 0s 200us/sample - loss: 4.2910e-08\n",
 651 |             "Epoch 236/500\n",
 652 |             "6/6 [==============================] - 0s 201us/sample - loss: 3.9502e-08\n",
 653 |             "Epoch 237/500\n",
 654 |             "6/6 [==============================] - 0s 205us/sample - loss: 3.6308e-08\n",
 655 |             "Epoch 238/500\n",
 656 |             "6/6 [==============================] - 0s 262us/sample - loss: 3.3432e-08\n",
 657 |             "Epoch 239/500\n",
 658 |             "6/6 [==============================] - 0s 197us/sample - loss: 3.0731e-08\n",
 659 |             "Epoch 240/500\n",
 660 |             "6/6 [==============================] - 0s 280us/sample - loss: 2.8310e-08\n",
 661 |             "Epoch 241/500\n",
 662 |             "6/6 [==============================] - 0s 262us/sample - loss: 2.6046e-08\n",
 663 |             "Epoch 242/500\n",
 664 |             "6/6 [==============================] - 0s 201us/sample - loss: 2.3968e-08\n",
 665 |             "Epoch 243/500\n",
 666 |             "6/6 [==============================] - 0s 315us/sample - loss: 2.2020e-08\n",
 667 |             "Epoch 244/500\n",
 668 |             "6/6 [==============================] - 0s 206us/sample - loss: 2.0253e-08\n",
 669 |             "Epoch 245/500\n",
 670 |             "6/6 [==============================] - 0s 259us/sample - loss: 1.8659e-08\n",
 671 |             "Epoch 246/500\n",
 672 |             "6/6 [==============================] - 0s 269us/sample - loss: 1.7149e-08\n",
 673 |             "Epoch 247/500\n",
 674 |             "6/6 [==============================] - 0s 223us/sample - loss: 1.5799e-08\n",
 675 |             "Epoch 248/500\n",
 676 |             "6/6 [==============================] - 0s 202us/sample - loss: 1.4530e-08\n",
 677 |             "Epoch 249/500\n",
 678 |             "6/6 [==============================] - 0s 208us/sample - loss: 1.3376e-08\n",
 679 |             "Epoch 250/500\n",
 680 |             "6/6 [==============================] - 0s 203us/sample - loss: 1.2297e-08\n",
 681 |             "Epoch 251/500\n",
 682 |             "6/6 [==============================] - 0s 208us/sample - loss: 1.1315e-08\n",
 683 |             "Epoch 252/500\n",
 684 |             "6/6 [==============================] - 0s 195us/sample - loss: 1.0409e-08\n",
 685 |             "Epoch 253/500\n",
 686 |             "6/6 [==============================] - 0s 192us/sample - loss: 9.5717e-09\n",
 687 |             "Epoch 254/500\n",
 688 |             "6/6 [==============================] - 0s 202us/sample - loss: 8.8057e-09\n",
 689 |             "Epoch 255/500\n",
 690 |             "6/6 [==============================] - 0s 189us/sample - loss: 8.1160e-09\n",
 691 |             "Epoch 256/500\n",
 692 |             "6/6 [==============================] - 0s 252us/sample - loss: 7.4598e-09\n",
 693 |             "Epoch 257/500\n",
 694 |             "6/6 [==============================] - 0s 326us/sample - loss: 6.8711e-09\n",
 695 |             "Epoch 258/500\n",
 696 |             "6/6 [==============================] - 0s 267us/sample - loss: 6.3167e-09\n",
 697 |             "Epoch 259/500\n",
 698 |             "6/6 [==============================] - 0s 1ms/sample - loss: 5.8123e-09\n",
 699 |             "Epoch 260/500\n",
 700 |             "6/6 [==============================] - 0s 218us/sample - loss: 5.3382e-09\n",
 701 |             "Epoch 261/500\n",
 702 |             "6/6 [==============================] - 0s 209us/sample - loss: 4.9115e-09\n",
 703 |             "Epoch 262/500\n",
 704 |             "6/6 [==============================] - 0s 305us/sample - loss: 4.5189e-09\n",
 705 |             "Epoch 263/500\n",
 706 |             "6/6 [==============================] - 0s 207us/sample - loss: 4.1606e-09\n",
 707 |             "Epoch 264/500\n",
 708 |             "6/6 [==============================] - 0s 264us/sample - loss: 3.8249e-09\n",
 709 |             "Epoch 265/500\n",
 710 |             "6/6 [==============================] - 0s 220us/sample - loss: 3.5184e-09\n",
 711 |             "Epoch 266/500\n",
 712 |             "6/6 [==============================] - 0s 214us/sample - loss: 3.2300e-09\n",
 713 |             "Epoch 267/500\n",
 714 |             "6/6 [==============================] - 0s 264us/sample - loss: 2.9778e-09\n",
 715 |             "Epoch 268/500\n",
 716 |             "6/6 [==============================] - 0s 297us/sample - loss: 2.7494e-09\n",
 717 |             "Epoch 269/500\n",
 718 |             "6/6 [==============================] - 0s 188us/sample - loss: 2.5152e-09\n",
 719 |             "Epoch 270/500\n",
 720 |             "6/6 [==============================] - 0s 202us/sample - loss: 2.3199e-09\n",
 721 |             "Epoch 271/500\n",
 722 |             "6/6 [==============================] - 0s 196us/sample - loss: 2.1375e-09\n",
 723 |             "Epoch 272/500\n",
 724 |             "6/6 [==============================] - 0s 202us/sample - loss: 1.9731e-09\n",
 725 |             "Epoch 273/500\n",
 726 |             "6/6 [==============================] - 0s 318us/sample - loss: 1.8056e-09\n",
 727 |             "Epoch 274/500\n",
 728 |             "6/6 [==============================] - 0s 238us/sample - loss: 1.6582e-09\n",
 729 |             "Epoch 275/500\n",
 730 |             "6/6 [==============================] - 0s 205us/sample - loss: 1.5355e-09\n",
 731 |             "Epoch 276/500\n",
 732 |             "6/6 [==============================] - 0s 238us/sample - loss: 1.4176e-09\n",
 733 |             "Epoch 277/500\n",
 734 |             "6/6 [==============================] - 0s 204us/sample - loss: 1.2955e-09\n",
 735 |             "Epoch 278/500\n",
 736 |             "6/6 [==============================] - 0s 273us/sample - loss: 1.1932e-09\n",
 737 |             "Epoch 279/500\n",
 738 |             "6/6 [==============================] - 0s 287us/sample - loss: 1.0982e-09\n",
 739 |             "Epoch 280/500\n",
 740 |             "6/6 [==============================] - 0s 289us/sample - loss: 1.0128e-09\n",
 741 |             "Epoch 281/500\n",
 742 |             "6/6 [==============================] - 0s 207us/sample - loss: 9.3499e-10\n",
 743 |             "Epoch 282/500\n",
 744 |             "6/6 [==============================] - 0s 266us/sample - loss: 8.5961e-10\n",
 745 |             "Epoch 283/500\n",
 746 |             "6/6 [==============================] - 0s 288us/sample - loss: 7.9097e-10\n",
 747 |             "Epoch 284/500\n",
 748 |             "6/6 [==============================] - 0s 300us/sample - loss: 7.2965e-10\n",
 749 |             "Epoch 285/500\n",
 750 |             "6/6 [==============================] - 0s 272us/sample - loss: 6.6897e-10\n",
 751 |             "Epoch 286/500\n",
 752 |             "6/6 [==============================] - 0s 282us/sample - loss: 6.1333e-10\n",
 753 |             "Epoch 287/500\n",
 754 |             "6/6 [==============================] - 0s 242us/sample - loss: 5.6922e-10\n",
 755 |             "Epoch 288/500\n",
 756 |             "6/6 [==============================] - 0s 306us/sample - loss: 5.1862e-10\n",
 757 |             "Epoch 289/500\n",
 758 |             "6/6 [==============================] - 0s 212us/sample - loss: 4.8086e-10\n",
 759 |             "Epoch 290/500\n",
 760 |             "6/6 [==============================] - 0s 267us/sample - loss: 4.4203e-10\n",
 761 |             "Epoch 291/500\n",
 762 |             "6/6 [==============================] - 0s 192us/sample - loss: 4.0309e-10\n",
 763 |             "Epoch 292/500\n",
 764 |             "6/6 [==============================] - 0s 201us/sample - loss: 3.7168e-10\n",
 765 |             "Epoch 293/500\n",
 766 |             "6/6 [==============================] - 0s 305us/sample - loss: 3.4139e-10\n",
 767 |             "Epoch 294/500\n",
 768 |             "6/6 [==============================] - 0s 169us/sample - loss: 3.1737e-10\n",
 769 |             "Epoch 295/500\n",
 770 |             "6/6 [==============================] - 0s 190us/sample - loss: 2.8630e-10\n",
 771 |             "Epoch 296/500\n",
 772 |             "6/6 [==============================] - 0s 257us/sample - loss: 2.6683e-10\n",
 773 |             "Epoch 297/500\n",
 774 |             "6/6 [==============================] - 0s 260us/sample - loss: 2.4443e-10\n",
 775 |             "Epoch 298/500\n",
 776 |             "6/6 [==============================] - 0s 296us/sample - loss: 2.2801e-10\n",
 777 |             "Epoch 299/500\n",
 778 |             "6/6 [==============================] - 0s 209us/sample - loss: 2.0742e-10\n",
 779 |             "Epoch 300/500\n",
 780 |             "6/6 [==============================] - 0s 274us/sample - loss: 1.9276e-10\n",
 781 |             "Epoch 301/500\n",
 782 |             "6/6 [==============================] - 0s 267us/sample - loss: 1.7687e-10\n",
 783 |             "Epoch 302/500\n",
 784 |             "6/6 [==============================] - 0s 275us/sample - loss: 1.6151e-10\n",
 785 |             "Epoch 303/500\n",
 786 |             "6/6 [==============================] - 0s 304us/sample - loss: 1.5130e-10\n",
 787 |             "Epoch 304/500\n",
 788 |             "6/6 [==============================] - 0s 169us/sample - loss: 1.3725e-10\n",
 789 |             "Epoch 305/500\n",
 790 |             "6/6 [==============================] - 0s 207us/sample - loss: 1.2639e-10\n",
 791 |             "Epoch 306/500\n",
 792 |             "6/6 [==============================] - 0s 262us/sample - loss: 1.1742e-10\n",
 793 |             "Epoch 307/500\n",
 794 |             "6/6 [==============================] - 0s 212us/sample - loss: 1.0780e-10\n",
 795 |             "Epoch 308/500\n",
 796 |             "6/6 [==============================] - 0s 203us/sample - loss: 9.9273e-11\n",
 797 |             "Epoch 309/500\n",
 798 |             "6/6 [==============================] - 0s 211us/sample - loss: 9.1340e-11\n",
 799 |             "Epoch 310/500\n",
 800 |             "6/6 [==============================] - 0s 265us/sample - loss: 8.3849e-11\n",
 801 |             "Epoch 311/500\n",
 802 |             "6/6 [==============================] - 0s 201us/sample - loss: 7.6223e-11\n",
 803 |             "Epoch 312/500\n",
 804 |             "6/6 [==============================] - 0s 170us/sample - loss: 7.1384e-11\n",
 805 |             "Epoch 313/500\n",
 806 |             "6/6 [==============================] - 0s 257us/sample - loss: 6.4340e-11\n",
 807 |             "Epoch 314/500\n",
 808 |             "6/6 [==============================] - 0s 261us/sample - loss: 6.0235e-11\n",
 809 |             "Epoch 315/500\n",
 810 |             "6/6 [==============================] - 0s 300us/sample - loss: 5.4162e-11\n",
 811 |             "Epoch 316/500\n",
 812 |             "6/6 [==============================] - 0s 172us/sample - loss: 4.8945e-11\n",
 813 |             "Epoch 317/500\n",
 814 |             "6/6 [==============================] - 0s 265us/sample - loss: 4.6366e-11\n",
 815 |             "Epoch 318/500\n",
 816 |             "6/6 [==============================] - 0s 223us/sample - loss: 4.0482e-11\n",
 817 |             "Epoch 319/500\n",
 818 |             "6/6 [==============================] - 0s 198us/sample - loss: 3.9130e-11\n",
 819 |             "Epoch 320/500\n",
 820 |             "6/6 [==============================] - 0s 321us/sample - loss: 3.5792e-11\n",
 821 |             "Epoch 321/500\n",
 822 |             "6/6 [==============================] - 0s 301us/sample - loss: 3.2583e-11\n",
 823 |             "Epoch 322/500\n",
 824 |             "6/6 [==============================] - 0s 250us/sample - loss: 3.0700e-11\n",
 825 |             "Epoch 323/500\n",
 826 |             "6/6 [==============================] - 0s 220us/sample - loss: 2.7359e-11\n",
 827 |             "Epoch 324/500\n",
 828 |             "6/6 [==============================] - 0s 281us/sample - loss: 2.5542e-11\n",
 829 |             "Epoch 325/500\n",
 830 |             "6/6 [==============================] - 0s 308us/sample - loss: 2.3801e-11\n",
 831 |             "Epoch 326/500\n",
 832 |             "6/6 [==============================] - 0s 190us/sample - loss: 2.2238e-11\n",
 833 |             "Epoch 327/500\n",
 834 |             "6/6 [==============================] - 0s 182us/sample - loss: 1.9808e-11\n",
 835 |             "Epoch 328/500\n",
 836 |             "6/6 [==============================] - 0s 168us/sample - loss: 1.8505e-11\n",
 837 |             "Epoch 329/500\n",
 838 |             "6/6 [==============================] - 0s 190us/sample - loss: 1.7497e-11\n",
 839 |             "Epoch 330/500\n",
 840 |             "6/6 [==============================] - 0s 267us/sample - loss: 1.6212e-11\n",
 841 |             "Epoch 331/500\n",
 842 |             "6/6 [==============================] - 0s 212us/sample - loss: 1.4533e-11\n",
 843 |             "Epoch 332/500\n",
 844 |             "6/6 [==============================] - 0s 199us/sample - loss: 1.3463e-11\n",
 845 |             "Epoch 333/500\n",
 846 |             "6/6 [==============================] - 0s 216us/sample - loss: 1.3084e-11\n",
 847 |             "Epoch 334/500\n",
 848 |             "6/6 [==============================] - 0s 211us/sample - loss: 1.1414e-11\n",
 849 |             "Epoch 335/500\n",
 850 |             "6/6 [==============================] - 0s 258us/sample - loss: 1.1210e-11\n",
 851 |             "Epoch 336/500\n",
 852 |             "6/6 [==============================] - 0s 302us/sample - loss: 9.7208e-12\n",
 853 |             "Epoch 337/500\n",
 854 |             "6/6 [==============================] - 0s 274us/sample - loss: 9.5077e-12\n",
 855 |             "Epoch 338/500\n",
 856 |             "6/6 [==============================] - 0s 259us/sample - loss: 9.0689e-12\n",
 857 |             "Epoch 339/500\n",
 858 |             "6/6 [==============================] - 0s 199us/sample - loss: 1.0128e-11\n",
 859 |             "Epoch 340/500\n",
 860 |             "6/6 [==============================] - 0s 216us/sample - loss: 7.7195e-12\n",
 861 |             "Epoch 341/500\n",
 862 |             "6/6 [==============================] - 0s 266us/sample - loss: 7.4139e-12\n",
 863 |             "Epoch 342/500\n",
 864 |             "6/6 [==============================] - 0s 172us/sample - loss: 6.7271e-12\n",
 865 |             "Epoch 343/500\n",
 866 |             "6/6 [==============================] - 0s 203us/sample - loss: 6.1586e-12\n",
 867 |             "Epoch 344/500\n",
 868 |             "6/6 [==============================] - 0s 459us/sample - loss: 5.9478e-12\n",
 869 |             "Epoch 345/500\n",
 870 |             "6/6 [==============================] - 0s 170us/sample - loss: 5.6938e-12\n",
 871 |             "Epoch 346/500\n",
 872 |             "6/6 [==============================] - 0s 262us/sample - loss: 5.6867e-12\n",
 873 |             "Epoch 347/500\n",
 874 |             "6/6 [==============================] - 0s 263us/sample - loss: 5.5547e-12\n",
 875 |             "Epoch 348/500\n",
 876 |             "6/6 [==============================] - 0s 302us/sample - loss: 5.0970e-12\n",
 877 |             "Epoch 349/500\n",
 878 |             "6/6 [==============================] - 0s 273us/sample - loss: 5.0407e-12\n",
 879 |             "Epoch 350/500\n",
 880 |             "6/6 [==============================] - 0s 262us/sample - loss: 5.5280e-12\n",
 881 |             "Epoch 351/500\n",
 882 |             "6/6 [==============================] - 0s 272us/sample - loss: 4.2372e-12\n",
 883 |             "Epoch 352/500\n",
 884 |             "6/6 [==============================] - 0s 313us/sample - loss: 3.5551e-12\n",
 885 |             "Epoch 353/500\n",
 886 |             "6/6 [==============================] - 0s 197us/sample - loss: 3.2022e-12\n",
 887 |             "Epoch 354/500\n",
 888 |             "6/6 [==============================] - 0s 260us/sample - loss: 3.0915e-12\n",
 889 |             "Epoch 355/500\n",
 890 |             "6/6 [==============================] - 0s 264us/sample - loss: 2.9866e-12\n",
 891 |             "Epoch 356/500\n",
 892 |             "6/6 [==============================] - 0s 298us/sample - loss: 2.9322e-12\n",
 893 |             "Epoch 357/500\n",
 894 |             "6/6 [==============================] - 0s 275us/sample - loss: 2.8445e-12\n",
 895 |             "Epoch 358/500\n",
 896 |             "6/6 [==============================] - 0s 285us/sample - loss: 2.8025e-12\n",
 897 |             "Epoch 359/500\n",
 898 |             "6/6 [==============================] - 0s 163us/sample - loss: 2.4751e-12\n",
 899 |             "Epoch 360/500\n",
 900 |             "6/6 [==============================] - 0s 206us/sample - loss: 2.7024e-12\n",
 901 |             "Epoch 361/500\n",
 902 |             "6/6 [==============================] - 0s 192us/sample - loss: 2.7095e-12\n",
 903 |             "Epoch 362/500\n",
 904 |             "6/6 [==============================] - 0s 268us/sample - loss: 2.4093e-12\n",
 905 |             "Epoch 363/500\n",
 906 |             "6/6 [==============================] - 0s 290us/sample - loss: 2.3329e-12\n",
 907 |             "Epoch 364/500\n",
 908 |             "6/6 [==============================] - 0s 208us/sample - loss: 2.2625e-12\n",
 909 |             "Epoch 365/500\n",
 910 |             "6/6 [==============================] - 0s 184us/sample - loss: 2.8351e-12\n",
 911 |             "Epoch 366/500\n",
 912 |             "6/6 [==============================] - 0s 308us/sample - loss: 1.7053e-12\n",
 913 |             "Epoch 367/500\n",
 914 |             "6/6 [==============================] - 0s 354us/sample - loss: 1.5964e-12\n",
 915 |             "Epoch 368/500\n",
 916 |             "6/6 [==============================] - 0s 356us/sample - loss: 1.3175e-12\n",
 917 |             "Epoch 369/500\n",
 918 |             "6/6 [==============================] - 0s 351us/sample - loss: 1.2885e-12\n",
 919 |             "Epoch 370/500\n",
 920 |             "6/6 [==============================] - 0s 341us/sample - loss: 1.2600e-12\n",
 921 |             "Epoch 371/500\n",
 922 |             "6/6 [==============================] - 0s 235us/sample - loss: 1.1700e-12\n",
 923 |             "Epoch 372/500\n",
 924 |             "6/6 [==============================] - 0s 350us/sample - loss: 9.7167e-13\n",
 925 |             "Epoch 373/500\n",
 926 |             "6/6 [==============================] - 0s 348us/sample - loss: 9.2134e-13\n",
 927 |             "Epoch 374/500\n",
 928 |             "6/6 [==============================] - 0s 287us/sample - loss: 9.2134e-13\n",
 929 |             "Epoch 375/500\n",
 930 |             "6/6 [==============================] - 0s 215us/sample - loss: 8.9824e-13\n",
 931 |             "Epoch 376/500\n",
 932 |             "6/6 [==============================] - 0s 355us/sample - loss: 8.9824e-13\n",
 933 |             "Epoch 377/500\n",
 934 |             "6/6 [==============================] - 0s 295us/sample - loss: 8.9824e-13\n",
 935 |             "Epoch 378/500\n",
 936 |             "6/6 [==============================] - 0s 285us/sample - loss: 8.7634e-13\n",
 937 |             "Epoch 379/500\n",
 938 |             "6/6 [==============================] - 0s 286us/sample - loss: 8.7634e-13\n",
 939 |             "Epoch 380/500\n",
 940 |             "6/6 [==============================] - 0s 487us/sample - loss: 8.5502e-13\n",
 941 |             "Epoch 381/500\n",
 942 |             "6/6 [==============================] - 0s 350us/sample - loss: 8.5502e-13\n",
 943 |             "Epoch 382/500\n",
 944 |             "6/6 [==============================] - 0s 462us/sample - loss: 8.5502e-13\n",
 945 |             "Epoch 383/500\n",
 946 |             "6/6 [==============================] - 0s 367us/sample - loss: 8.3430e-13\n",
 947 |             "Epoch 384/500\n",
 948 |             "6/6 [==============================] - 0s 355us/sample - loss: 7.0166e-13\n",
 949 |             "Epoch 385/500\n",
 950 |             "6/6 [==============================] - 0s 255us/sample - loss: 7.0166e-13\n",
 951 |             "Epoch 386/500\n",
 952 |             "6/6 [==============================] - 0s 347us/sample - loss: 7.0166e-13\n",
 953 |             "Epoch 387/500\n",
 954 |             "6/6 [==============================] - 0s 345us/sample - loss: 7.0166e-13\n",
 955 |             "Epoch 388/500\n",
 956 |             "6/6 [==============================] - 0s 258us/sample - loss: 7.0166e-13\n",
 957 |             "Epoch 389/500\n",
 958 |             "6/6 [==============================] - 0s 348us/sample - loss: 7.0166e-13\n",
 959 |             "Epoch 390/500\n",
 960 |             "6/6 [==============================] - 0s 329us/sample - loss: 7.0166e-13\n",
 961 |             "Epoch 391/500\n",
 962 |             "6/6 [==============================] - 0s 307us/sample - loss: 7.0166e-13\n",
 963 |             "Epoch 392/500\n",
 964 |             "6/6 [==============================] - 0s 331us/sample - loss: 7.0166e-13\n",
 965 |             "Epoch 393/500\n",
 966 |             "6/6 [==============================] - 0s 338us/sample - loss: 7.0166e-13\n",
 967 |             "Epoch 394/500\n",
 968 |             "6/6 [==============================] - 0s 308us/sample - loss: 7.0166e-13\n",
 969 |             "Epoch 395/500\n",
 970 |             "6/6 [==============================] - 0s 347us/sample - loss: 7.0166e-13\n",
 971 |             "Epoch 396/500\n",
 972 |             "6/6 [==============================] - 0s 289us/sample - loss: 7.0166e-13\n",
 973 |             "Epoch 397/500\n",
 974 |             "6/6 [==============================] - 0s 219us/sample - loss: 7.0166e-13\n",
 975 |             "Epoch 398/500\n",
 976 |             "6/6 [==============================] - 0s 662us/sample - loss: 7.0166e-13\n",
 977 |             "Epoch 399/500\n",
 978 |             "6/6 [==============================] - 0s 188us/sample - loss: 7.0166e-13\n",
 979 |             "Epoch 400/500\n",
 980 |             "6/6 [==============================] - 0s 254us/sample - loss: 7.0166e-13\n",
 981 |             "Epoch 401/500\n",
 982 |             "6/6 [==============================] - 0s 319us/sample - loss: 7.0166e-13\n",
 983 |             "Epoch 402/500\n",
 984 |             "6/6 [==============================] - 0s 228us/sample - loss: 7.0166e-13\n",
 985 |             "Epoch 403/500\n",
 986 |             "6/6 [==============================] - 0s 348us/sample - loss: 7.0166e-13\n",
 987 |             "Epoch 404/500\n",
 988 |             "6/6 [==============================] - 0s 347us/sample - loss: 7.0166e-13\n",
 989 |             "Epoch 405/500\n",
 990 |             "6/6 [==============================] - 0s 346us/sample - loss: 7.0166e-13\n",
 991 |             "Epoch 406/500\n",
 992 |             "6/6 [==============================] - 0s 289us/sample - loss: 7.0166e-13\n",
 993 |             "Epoch 407/500\n",
 994 |             "6/6 [==============================] - 0s 293us/sample - loss: 7.0166e-13\n",
 995 |             "Epoch 408/500\n",
 996 |             "6/6 [==============================] - 0s 338us/sample - loss: 7.0166e-13\n",
 997 |             "Epoch 409/500\n",
 998 |             "6/6 [==============================] - 0s 349us/sample - loss: 7.0166e-13\n",
 999 |             "Epoch 410/500\n",
1000 |             "6/6 [==============================] - 0s 227us/sample - loss: 7.0166e-13\n",
1001 |             "Epoch 411/500\n",
1002 |             "6/6 [==============================] - 0s 774us/sample - loss: 7.0166e-13\n",
1003 |             "Epoch 412/500\n",
1004 |             "6/6 [==============================] - 0s 414us/sample - loss: 7.0166e-13\n",
1005 |             "Epoch 413/500\n",
1006 |             "6/6 [==============================] - 0s 365us/sample - loss: 7.0166e-13\n",
1007 |             "Epoch 414/500\n",
1008 |             "6/6 [==============================] - 0s 354us/sample - loss: 7.0166e-13\n",
1009 |             "Epoch 415/500\n",
1010 |             "6/6 [==============================] - 0s 360us/sample - loss: 7.0166e-13\n",
1011 |             "Epoch 416/500\n",
1012 |             "6/6 [==============================] - 0s 250us/sample - loss: 7.0166e-13\n",
1013 |             "Epoch 417/500\n",
1014 |             "6/6 [==============================] - 0s 219us/sample - loss: 7.0166e-13\n",
1015 |             "Epoch 418/500\n",
1016 |             "6/6 [==============================] - 0s 250us/sample - loss: 7.0166e-13\n",
1017 |             "Epoch 419/500\n",
1018 |             "6/6 [==============================] - 0s 311us/sample - loss: 7.0166e-13\n",
1019 |             "Epoch 420/500\n",
1020 |             "6/6 [==============================] - 0s 274us/sample - loss: 7.0166e-13\n",
1021 |             "Epoch 421/500\n",
1022 |             "6/6 [==============================] - 0s 200us/sample - loss: 7.0166e-13\n",
1023 |             "Epoch 422/500\n",
1024 |             "6/6 [==============================] - 0s 205us/sample - loss: 7.0166e-13\n",
1025 |             "Epoch 423/500\n",
1026 |             "6/6 [==============================] - 0s 305us/sample - loss: 7.0166e-13\n",
1027 |             "Epoch 424/500\n",
1028 |             "6/6 [==============================] - 0s 201us/sample - loss: 7.0166e-13\n",
1029 |             "Epoch 425/500\n",
1030 |             "6/6 [==============================] - 0s 207us/sample - loss: 7.0166e-13\n",
1031 |             "Epoch 426/500\n",
1032 |             "6/6 [==============================] - 0s 216us/sample - loss: 7.0166e-13\n",
1033 |             "Epoch 427/500\n",
1034 |             "6/6 [==============================] - 0s 288us/sample - loss: 7.0166e-13\n",
1035 |             "Epoch 428/500\n",
1036 |             "6/6 [==============================] - 0s 294us/sample - loss: 7.0166e-13\n",
1037 |             "Epoch 429/500\n",
1038 |             "6/6 [==============================] - 0s 236us/sample - loss: 7.0166e-13\n",
1039 |             "Epoch 430/500\n",
1040 |             "6/6 [==============================] - 0s 216us/sample - loss: 7.0166e-13\n",
1041 |             "Epoch 431/500\n",
1042 |             "6/6 [==============================] - 0s 276us/sample - loss: 7.0166e-13\n",
1043 |             "Epoch 432/500\n",
1044 |             "6/6 [==============================] - 0s 297us/sample - loss: 7.0166e-13\n",
1045 |             "Epoch 433/500\n",
1046 |             "6/6 [==============================] - 0s 205us/sample - loss: 7.0166e-13\n",
1047 |             "Epoch 434/500\n",
1048 |             "6/6 [==============================] - 0s 203us/sample - loss: 7.0166e-13\n",
1049 |             "Epoch 435/500\n",
1050 |             "6/6 [==============================] - 0s 264us/sample - loss: 7.0166e-13\n",
1051 |             "Epoch 436/500\n",
1052 |             "6/6 [==============================] - 0s 198us/sample - loss: 7.0166e-13\n",
1053 |             "Epoch 437/500\n",
1054 |             "6/6 [==============================] - 0s 220us/sample - loss: 7.0166e-13\n",
1055 |             "Epoch 438/500\n",
1056 |             "6/6 [==============================] - 0s 242us/sample - loss: 7.0166e-13\n",
1057 |             "Epoch 439/500\n",
1058 |             "6/6 [==============================] - 0s 235us/sample - loss: 7.0166e-13\n",
1059 |             "Epoch 440/500\n",
1060 |             "6/6 [==============================] - 0s 301us/sample - loss: 7.0166e-13\n",
1061 |             "Epoch 441/500\n",
1062 |             "6/6 [==============================] - 0s 249us/sample - loss: 7.0166e-13\n",
1063 |             "Epoch 442/500\n",
1064 |             "6/6 [==============================] - 0s 204us/sample - loss: 7.0166e-13\n",
1065 |             "Epoch 443/500\n",
1066 |             "6/6 [==============================] - 0s 259us/sample - loss: 7.0166e-13\n",
1067 |             "Epoch 444/500\n",
1068 |             "6/6 [==============================] - 0s 209us/sample - loss: 7.0166e-13\n",
1069 |             "Epoch 445/500\n",
1070 |             "6/6 [==============================] - 0s 206us/sample - loss: 7.0166e-13\n",
1071 |             "Epoch 446/500\n",
1072 |             "6/6 [==============================] - 0s 190us/sample - loss: 7.0166e-13\n",
1073 |             "Epoch 447/500\n",
1074 |             "6/6 [==============================] - 0s 356us/sample - loss: 7.0166e-13\n",
1075 |             "Epoch 448/500\n",
1076 |             "6/6 [==============================] - 0s 294us/sample - loss: 7.0166e-13\n",
1077 |             "Epoch 449/500\n",
1078 |             "6/6 [==============================] - 0s 291us/sample - loss: 7.0166e-13\n",
1079 |             "Epoch 450/500\n",
1080 |             "6/6 [==============================] - 0s 415us/sample - loss: 7.0166e-13\n",
1081 |             "Epoch 451/500\n",
1082 |             "6/6 [==============================] - 0s 331us/sample - loss: 7.0166e-13\n",
1083 |             "Epoch 452/500\n",
1084 |             "6/6 [==============================] - 0s 247us/sample - loss: 7.0166e-13\n",
1085 |             "Epoch 453/500\n",
1086 |             "6/6 [==============================] - 0s 312us/sample - loss: 7.0166e-13\n",
1087 |             "Epoch 454/500\n",
1088 |             "6/6 [==============================] - 0s 363us/sample - loss: 7.0166e-13\n",
1089 |             "Epoch 455/500\n",
1090 |             "6/6 [==============================] - 0s 357us/sample - loss: 7.0166e-13\n",
1091 |             "Epoch 456/500\n",
1092 |             "6/6 [==============================] - 0s 371us/sample - loss: 7.0166e-13\n",
1093 |             "Epoch 457/500\n",
1094 |             "6/6 [==============================] - 0s 411us/sample - loss: 7.0166e-13\n",
1095 |             "Epoch 458/500\n",
1096 |             "6/6 [==============================] - 0s 311us/sample - loss: 7.0166e-13\n",
1097 |             "Epoch 459/500\n",
1098 |             "6/6 [==============================] - 0s 410us/sample - loss: 7.0166e-13\n",
1099 |             "Epoch 460/500\n",
1100 |             "6/6 [==============================] - 0s 369us/sample - loss: 7.0166e-13\n",
1101 |             "Epoch 461/500\n",
1102 |             "6/6 [==============================] - 0s 404us/sample - loss: 7.0166e-13\n",
1103 |             "Epoch 462/500\n",
1104 |             "6/6 [==============================] - 0s 305us/sample - loss: 7.0166e-13\n",
1105 |             "Epoch 463/500\n",
1106 |             "6/6 [==============================] - 0s 369us/sample - loss: 7.0166e-13\n",
1107 |             "Epoch 464/500\n",
1108 |             "6/6 [==============================] - 0s 391us/sample - loss: 7.0166e-13\n",
1109 |             "Epoch 465/500\n",
1110 |             "6/6 [==============================] - 0s 356us/sample - loss: 7.0166e-13\n",
1111 |             "Epoch 466/500\n",
1112 |             "6/6 [==============================] - 0s 351us/sample - loss: 7.0166e-13\n",
1113 |             "Epoch 467/500\n",
1114 |             "6/6 [==============================] - 0s 311us/sample - loss: 7.0166e-13\n",
1115 |             "Epoch 468/500\n",
1116 |             "6/6 [==============================] - 0s 359us/sample - loss: 7.0166e-13\n",
1117 |             "Epoch 469/500\n",
1118 |             "6/6 [==============================] - 0s 460us/sample - loss: 7.0166e-13\n",
1119 |             "Epoch 470/500\n",
1120 |             "6/6 [==============================] - 0s 433us/sample - loss: 7.0166e-13\n",
1121 |             "Epoch 471/500\n",
1122 |             "6/6 [==============================] - 0s 497us/sample - loss: 7.0166e-13\n",
1123 |             "Epoch 472/500\n",
1124 |             "6/6 [==============================] - 0s 403us/sample - loss: 7.0166e-13\n",
1125 |             "Epoch 473/500\n",
1126 |             "6/6 [==============================] - 0s 469us/sample - loss: 7.0166e-13\n",
1127 |             "Epoch 474/500\n",
1128 |             "6/6 [==============================] - 0s 401us/sample - loss: 7.0166e-13\n",
1129 |             "Epoch 475/500\n",
1130 |             "6/6 [==============================] - 0s 445us/sample - loss: 7.0166e-13\n",
1131 |             "Epoch 476/500\n",
1132 |             "6/6 [==============================] - 0s 291us/sample - loss: 7.0166e-13\n",
1133 |             "Epoch 477/500\n",
1134 |             "6/6 [==============================] - 0s 326us/sample - loss: 7.0166e-13\n",
1135 |             "Epoch 478/500\n",
1136 |             "6/6 [==============================] - 0s 475us/sample - loss: 7.0166e-13\n",
1137 |             "Epoch 479/500\n",
1138 |             "6/6 [==============================] - 0s 282us/sample - loss: 7.0166e-13\n",
1139 |             "Epoch 480/500\n",
1140 |             "6/6 [==============================] - 0s 288us/sample - loss: 7.0166e-13\n",
1141 |             "Epoch 481/500\n",
1142 |             "6/6 [==============================] - 0s 390us/sample - loss: 7.0166e-13\n",
1143 |             "Epoch 482/500\n",
1144 |             "6/6 [==============================] - 0s 304us/sample - loss: 7.0166e-13\n",
1145 |             "Epoch 483/500\n",
1146 |             "6/6 [==============================] - 0s 435us/sample - loss: 7.0166e-13\n",
1147 |             "Epoch 484/500\n",
1148 |             "6/6 [==============================] - 0s 298us/sample - loss: 7.0166e-13\n",
1149 |             "Epoch 485/500\n",
1150 |             "6/6 [==============================] - 0s 298us/sample - loss: 7.0166e-13\n",
1151 |             "Epoch 486/500\n",
1152 |             "6/6 [==============================] - 0s 286us/sample - loss: 7.0166e-13\n",
1153 |             "Epoch 487/500\n",
1154 |             "6/6 [==============================] - 0s 273us/sample - loss: 7.0166e-13\n",
1155 |             "Epoch 488/500\n",
1156 |             "6/6 [==============================] - 0s 290us/sample - loss: 7.0166e-13\n",
1157 |             "Epoch 489/500\n",
1158 |             "6/6 [==============================] - 0s 280us/sample - loss: 7.0166e-13\n",
1159 |             "Epoch 490/500\n",
1160 |             "6/6 [==============================] - 0s 248us/sample - loss: 7.0166e-13\n",
1161 |             "Epoch 491/500\n",
1162 |             "6/6 [==============================] - 0s 279us/sample - loss: 7.0166e-13\n",
1163 |             "Epoch 492/500\n",
1164 |             "6/6 [==============================] - 0s 280us/sample - loss: 7.0166e-13\n",
1165 |             "Epoch 493/500\n",
1166 |             "6/6 [==============================] - 0s 234us/sample - loss: 7.0166e-13\n",
1167 |             "Epoch 494/500\n",
1168 |             "6/6 [==============================] - 0s 192us/sample - loss: 7.0166e-13\n",
1169 |             "Epoch 495/500\n",
1170 |             "6/6 [==============================] - 0s 198us/sample - loss: 7.0166e-13\n",
1171 |             "Epoch 496/500\n",
1172 |             "6/6 [==============================] - 0s 287us/sample - loss: 7.0166e-13\n",
1173 |             "Epoch 497/500\n",
1174 |             "6/6 [==============================] - 0s 210us/sample - loss: 7.0166e-13\n",
1175 |             "Epoch 498/500\n",
1176 |             "6/6 [==============================] - 0s 209us/sample - loss: 7.0166e-13\n",
1177 |             "Epoch 499/500\n",
1178 |             "6/6 [==============================] - 0s 227us/sample - loss: 7.0166e-13\n",
1179 |             "Epoch 500/500\n",
1180 |             "6/6 [==============================] - 0s 313us/sample - loss: 7.0166e-13\n"
1181 |           ],
1182 |           "name": "stdout"
1183 |         },
1184 |         {
1185 |           "output_type": "execute_result",
1186 |           "data": {
1187 |             "text/plain": [
1188 |               "<tensorflow.python.keras.callbacks.History at 0x7ff495cedcc0>"
1189 |             ]
1190 |           },
1191 |           "metadata": {
1192 |             "tags": []
1193 |           },
1194 |           "execution_count": 6
1195 |         }
1196 |       ]
1197 |     },
1198 |     {
1199 |       "cell_type": "markdown",
1200 |       "metadata": {
1201 |         "id": "nVCCCLkjfeM7",
1202 |         "colab_type": "text"
1203 |       },
1204 |       "source": [
1205 |         "###Use model to predict responce with $x=3.5$ as input.\n",
1206 |         "###Based on formula $y=2x-1$ the predicted value should be close to 6."
1207 |       ]
1208 |     },
1209 |     {
1210 |       "cell_type": "code",
1211 |       "metadata": {
1212 |         "id": "cb6sV2vdfpZM",
1213 |         "colab_type": "code",
1214 |         "outputId": "7c9199ca-7ff6-4b2b-ec5d-f280ba22d97c",
1215 |         "colab": {
1216 |           "base_uri": "https://localhost:8080/",
1217 |           "height": 35
1218 |         }
1219 |       },
1220 |       "source": [
1221 |         "print(model.predict([3.5]))"
1222 |       ],
1223 |       "execution_count": 0,
1224 |       "outputs": [
1225 |         {
1226 |           "output_type": "stream",
1227 |           "text": [
1228 |             "[[5.9999995]]\n"
1229 |           ],
1230 |           "name": "stdout"
1231 |         }
1232 |       ]
1233 |     },
1234 |     {
1235 |       "cell_type": "markdown",
1236 |       "metadata": {
1237 |         "id": "WXNvJTWgf8o8",
1238 |         "colab_type": "text"
1239 |       },
1240 |       "source": [
1241 |         "###We can view a summary of the model "
1242 |       ]
1243 |     },
1244 |     {
1245 |       "cell_type": "code",
1246 |       "metadata": {
1247 |         "id": "erbBOBL8gCYa",
1248 |         "colab_type": "code",
1249 |         "outputId": "81c64390-c155-43f1-8b0c-95837aa8493c",
1250 |         "colab": {
1251 |           "base_uri": "https://localhost:8080/",
1252 |           "height": 237
1253 |         }
1254 |       },
1255 |       "source": [
1256 |         "model.summary()"
1257 |       ],
1258 |       "execution_count": 0,
1259 |       "outputs": [
1260 |         {
1261 |           "output_type": "stream",
1262 |           "text": [
1263 |             "Model: \"sequential\"\n",
1264 |             "_________________________________________________________________\n",
1265 |             "Layer (type)                 Output Shape              Param #   \n",
1266 |             "=================================================================\n",
1267 |             "dense (Dense)                (None, 1)                 2         \n",
1268 |             "_________________________________________________________________\n",
1269 |             "dense_1 (Dense)              (None, 1)                 2         \n",
1270 |             "=================================================================\n",
1271 |             "Total params: 4\n",
1272 |             "Trainable params: 4\n",
1273 |             "Non-trainable params: 0\n",
1274 |             "_________________________________________________________________\n"
1275 |           ],
1276 |           "name": "stdout"
1277 |         }
1278 |       ]
1279 |     }
1280 |   ]
1281 | }


--------------------------------------------------------------------------------
/Spacy_example.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "name": "Spacy example.ipynb",
  7 |       "provenance": [],
  8 |       "collapsed_sections": [],
  9 |       "include_colab_link": true
 10 |     },
 11 |     "kernelspec": {
 12 |       "name": "python3",
 13 |       "display_name": "Python 3"
 14 |     }
 15 |   },
 16 |   "cells": [
 17 |     {
 18 |       "cell_type": "markdown",
 19 |       "metadata": {
 20 |         "id": "view-in-github",
 21 |         "colab_type": "text"
 22 |       },
 23 |       "source": [
 24 |         "<a href=\"https://colab.research.google.com/github/dpanagop/ML_and_AI_examples/blob/master/Spacy_example.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
 25 |       ]
 26 |     },
 27 |     {
 28 |       "cell_type": "code",
 29 |       "metadata": {
 30 |         "id": "yIARIXVzyqZe",
 31 |         "colab_type": "code",
 32 |         "outputId": "7291a9ba-b53c-4613-ba63-414c247377ed",
 33 |         "colab": {
 34 |           "base_uri": "https://localhost:8080/",
 35 |           "height": 845
 36 |         }
 37 |       },
 38 |       "source": [
 39 |         "!pip install -U spacy"
 40 |       ],
 41 |       "execution_count": 1,
 42 |       "outputs": [
 43 |         {
 44 |           "output_type": "stream",
 45 |           "text": [
 46 |             "Collecting spacy\n",
 47 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/47/13/80ad28ef7a16e2a86d16d73e28588be5f1085afd3e85e4b9b912bd700e8a/spacy-2.2.3-cp36-cp36m-manylinux1_x86_64.whl (10.4MB)\n",
 48 |             "\u001b[K     |████████████████████████████████| 10.4MB 1.4MB/s \n",
 49 |             "\u001b[?25hRequirement already satisfied, skipping upgrade: setuptools in /usr/local/lib/python3.6/dist-packages (from spacy) (41.6.0)\n",
 50 |             "Requirement already satisfied, skipping upgrade: plac<1.2.0,>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from spacy) (0.9.6)\n",
 51 |             "Collecting preshed<3.1.0,>=3.0.2\n",
 52 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/db/6b/e07fad36913879757c90ba03d6fb7f406f7279e11dcefc105ee562de63ea/preshed-3.0.2-cp36-cp36m-manylinux1_x86_64.whl (119kB)\n",
 53 |             "\u001b[K     |████████████████████████████████| 122kB 52.2MB/s \n",
 54 |             "\u001b[?25hRequirement already satisfied, skipping upgrade: wasabi<1.1.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (0.4.0)\n",
 55 |             "Requirement already satisfied, skipping upgrade: requests<3.0.0,>=2.13.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (2.21.0)\n",
 56 |             "Requirement already satisfied, skipping upgrade: srsly<1.1.0,>=0.1.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (0.2.0)\n",
 57 |             "Requirement already satisfied, skipping upgrade: numpy>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (1.17.4)\n",
 58 |             "Requirement already satisfied, skipping upgrade: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.6/dist-packages (from spacy) (1.0.2)\n",
 59 |             "Collecting thinc<7.4.0,>=7.3.0\n",
 60 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/07/59/6bb553bc9a5f072d3cd479fc939fea0f6f682892f1f5cff98de5c9b615bb/thinc-7.3.1-cp36-cp36m-manylinux1_x86_64.whl (2.2MB)\n",
 61 |             "\u001b[K     |████████████████████████████████| 2.2MB 43.9MB/s \n",
 62 |             "\u001b[?25hCollecting catalogue<1.1.0,>=0.0.7\n",
 63 |             "  Downloading https://files.pythonhosted.org/packages/4f/d5/46ff975f0d7d055cf95557b944fd5d29d9dfb37a4341038e070f212b24fe/catalogue-0.0.8-py2.py3-none-any.whl\n",
 64 |             "Collecting blis<0.5.0,>=0.4.0\n",
 65 |             "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/41/19/f95c75562d18eb27219df3a3590b911e78d131b68466ad79fdf5847eaac4/blis-0.4.1-cp36-cp36m-manylinux1_x86_64.whl (3.7MB)\n",
 66 |             "\u001b[K     |████████████████████████████████| 3.7MB 48.6MB/s \n",
 67 |             "\u001b[?25hRequirement already satisfied, skipping upgrade: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy) (2.0.3)\n",
 68 |             "Requirement already satisfied, skipping upgrade: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (3.0.4)\n",
 69 |             "Requirement already satisfied, skipping upgrade: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (2.8)\n",
 70 |             "Requirement already satisfied, skipping upgrade: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (1.24.3)\n",
 71 |             "Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy) (2019.9.11)\n",
 72 |             "Requirement already satisfied, skipping upgrade: tqdm<5.0.0,>=4.10.0 in /usr/local/lib/python3.6/dist-packages (from thinc<7.4.0,>=7.3.0->spacy) (4.28.1)\n",
 73 |             "Requirement already satisfied, skipping upgrade: importlib-metadata>=0.20; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from catalogue<1.1.0,>=0.0.7->spacy) (0.23)\n",
 74 |             "Requirement already satisfied, skipping upgrade: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy) (0.6.0)\n",
 75 |             "Requirement already satisfied, skipping upgrade: more-itertools in /usr/local/lib/python3.6/dist-packages (from zipp>=0.5->importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy) (7.2.0)\n",
 76 |             "Installing collected packages: preshed, blis, thinc, catalogue, spacy\n",
 77 |             "  Found existing installation: preshed 2.0.1\n",
 78 |             "    Uninstalling preshed-2.0.1:\n",
 79 |             "      Successfully uninstalled preshed-2.0.1\n",
 80 |             "  Found existing installation: blis 0.2.4\n",
 81 |             "    Uninstalling blis-0.2.4:\n",
 82 |             "      Successfully uninstalled blis-0.2.4\n",
 83 |             "  Found existing installation: thinc 7.0.8\n",
 84 |             "    Uninstalling thinc-7.0.8:\n",
 85 |             "      Successfully uninstalled thinc-7.0.8\n",
 86 |             "  Found existing installation: spacy 2.1.9\n",
 87 |             "    Uninstalling spacy-2.1.9:\n",
 88 |             "      Successfully uninstalled spacy-2.1.9\n",
 89 |             "Successfully installed blis-0.4.1 catalogue-0.0.8 preshed-3.0.2 spacy-2.2.3 thinc-7.3.1\n"
 90 |           ],
 91 |           "name": "stdout"
 92 |         }
 93 |       ]
 94 |     },
 95 |     {
 96 |       "cell_type": "code",
 97 |       "metadata": {
 98 |         "id": "LXpmvaPijt4w",
 99 |         "colab_type": "code",
100 |         "outputId": "21263b86-21c2-48f9-e25b-bf2994de5526",
101 |         "colab": {
102 |           "base_uri": "https://localhost:8080/",
103 |           "height": 772
104 |         }
105 |       },
106 |       "source": [
107 |         "!python -m spacy download en"
108 |       ],
109 |       "execution_count": 2,
110 |       "outputs": [
111 |         {
112 |           "output_type": "stream",
113 |           "text": [
114 |             "Collecting en_core_web_sm==2.2.5\n",
115 |             "\u001b[?25l  Downloading https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-2.2.5/en_core_web_sm-2.2.5.tar.gz (12.0MB)\n",
116 |             "\u001b[K     |████████████████████████████████| 12.0MB 607kB/s \n",
117 |             "\u001b[?25hRequirement already satisfied: spacy>=2.2.2 in /usr/local/lib/python3.6/dist-packages (from en_core_web_sm==2.2.5) (2.2.3)\n",
118 |             "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (3.0.2)\n",
119 |             "Requirement already satisfied: plac<1.2.0,>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (0.9.6)\n",
120 |             "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (2.0.3)\n",
121 |             "Requirement already satisfied: requests<3.0.0,>=2.13.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (2.21.0)\n",
122 |             "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (1.0.2)\n",
123 |             "Requirement already satisfied: catalogue<1.1.0,>=0.0.7 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (0.0.8)\n",
124 |             "Requirement already satisfied: thinc<7.4.0,>=7.3.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (7.3.1)\n",
125 |             "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (41.6.0)\n",
126 |             "Requirement already satisfied: wasabi<1.1.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (0.4.0)\n",
127 |             "Requirement already satisfied: srsly<1.1.0,>=0.1.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (0.2.0)\n",
128 |             "Requirement already satisfied: blis<0.5.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (0.4.1)\n",
129 |             "Requirement already satisfied: numpy>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_sm==2.2.5) (1.17.4)\n",
130 |             "Requirement already satisfied: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_sm==2.2.5) (1.24.3)\n",
131 |             "Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_sm==2.2.5) (3.0.4)\n",
132 |             "Requirement already satisfied: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_sm==2.2.5) (2.8)\n",
133 |             "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_sm==2.2.5) (2019.9.11)\n",
134 |             "Requirement already satisfied: importlib-metadata>=0.20; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->en_core_web_sm==2.2.5) (0.23)\n",
135 |             "Requirement already satisfied: tqdm<5.0.0,>=4.10.0 in /usr/local/lib/python3.6/dist-packages (from thinc<7.4.0,>=7.3.0->spacy>=2.2.2->en_core_web_sm==2.2.5) (4.28.1)\n",
136 |             "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->en_core_web_sm==2.2.5) (0.6.0)\n",
137 |             "Requirement already satisfied: more-itertools in /usr/local/lib/python3.6/dist-packages (from zipp>=0.5->importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->en_core_web_sm==2.2.5) (7.2.0)\n",
138 |             "Building wheels for collected packages: en-core-web-sm\n",
139 |             "  Building wheel for en-core-web-sm (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
140 |             "  Created wheel for en-core-web-sm: filename=en_core_web_sm-2.2.5-cp36-none-any.whl size=12011741 sha256=3164e00eb94719adb5fc0229e4b43799309741207c8878e79d8cb1942f5d08d9\n",
141 |             "  Stored in directory: /tmp/pip-ephem-wheel-cache-7azvuuf_/wheels/6a/47/fb/6b5a0b8906d8e8779246c67d4658fd8a544d4a03a75520197a\n",
142 |             "Successfully built en-core-web-sm\n",
143 |             "Installing collected packages: en-core-web-sm\n",
144 |             "  Found existing installation: en-core-web-sm 2.1.0\n",
145 |             "    Uninstalling en-core-web-sm-2.1.0:\n",
146 |             "      Successfully uninstalled en-core-web-sm-2.1.0\n",
147 |             "Successfully installed en-core-web-sm-2.2.5\n",
148 |             "\u001b[38;5;2m✔ Download and installation successful\u001b[0m\n",
149 |             "You can now load the model via spacy.load('en_core_web_sm')\n",
150 |             "\u001b[38;5;2m✔ Linking successful\u001b[0m\n",
151 |             "/usr/local/lib/python3.6/dist-packages/en_core_web_sm -->\n",
152 |             "/usr/local/lib/python3.6/dist-packages/spacy/data/en\n",
153 |             "You can now load the model via spacy.load('en')\n"
154 |           ],
155 |           "name": "stdout"
156 |         }
157 |       ]
158 |     },
159 |     {
160 |       "cell_type": "code",
161 |       "metadata": {
162 |         "id": "2pask3OOs1Bf",
163 |         "colab_type": "code",
164 |         "outputId": "8d69587f-12ae-401b-fb14-e87d710f625f",
165 |         "colab": {
166 |           "base_uri": "https://localhost:8080/",
167 |           "height": 643
168 |         }
169 |       },
170 |       "source": [
171 |         "!python -m spacy download en_core_web_md"
172 |       ],
173 |       "execution_count": 3,
174 |       "outputs": [
175 |         {
176 |           "output_type": "stream",
177 |           "text": [
178 |             "Collecting en_core_web_md==2.2.5\n",
179 |             "\u001b[?25l  Downloading https://github.com/explosion/spacy-models/releases/download/en_core_web_md-2.2.5/en_core_web_md-2.2.5.tar.gz (96.4MB)\n",
180 |             "\u001b[K     |████████████████████████████████| 96.4MB 7.9MB/s \n",
181 |             "\u001b[?25hRequirement already satisfied: spacy>=2.2.2 in /usr/local/lib/python3.6/dist-packages (from en_core_web_md==2.2.5) (2.2.3)\n",
182 |             "Requirement already satisfied: requests<3.0.0,>=2.13.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (2.21.0)\n",
183 |             "Requirement already satisfied: numpy>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (1.17.4)\n",
184 |             "Requirement already satisfied: wasabi<1.1.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (0.4.0)\n",
185 |             "Requirement already satisfied: plac<1.2.0,>=0.9.6 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (0.9.6)\n",
186 |             "Requirement already satisfied: blis<0.5.0,>=0.4.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (0.4.1)\n",
187 |             "Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (1.0.2)\n",
188 |             "Requirement already satisfied: catalogue<1.1.0,>=0.0.7 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (0.0.8)\n",
189 |             "Requirement already satisfied: srsly<1.1.0,>=0.1.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (0.2.0)\n",
190 |             "Requirement already satisfied: preshed<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (3.0.2)\n",
191 |             "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (41.6.0)\n",
192 |             "Requirement already satisfied: cymem<2.1.0,>=2.0.2 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (2.0.3)\n",
193 |             "Requirement already satisfied: thinc<7.4.0,>=7.3.0 in /usr/local/lib/python3.6/dist-packages (from spacy>=2.2.2->en_core_web_md==2.2.5) (7.3.1)\n",
194 |             "Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_md==2.2.5) (3.0.4)\n",
195 |             "Requirement already satisfied: idna<2.9,>=2.5 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_md==2.2.5) (2.8)\n",
196 |             "Requirement already satisfied: urllib3<1.25,>=1.21.1 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_md==2.2.5) (1.24.3)\n",
197 |             "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.6/dist-packages (from requests<3.0.0,>=2.13.0->spacy>=2.2.2->en_core_web_md==2.2.5) (2019.9.11)\n",
198 |             "Requirement already satisfied: importlib-metadata>=0.20; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->en_core_web_md==2.2.5) (0.23)\n",
199 |             "Requirement already satisfied: tqdm<5.0.0,>=4.10.0 in /usr/local/lib/python3.6/dist-packages (from thinc<7.4.0,>=7.3.0->spacy>=2.2.2->en_core_web_md==2.2.5) (4.28.1)\n",
200 |             "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->en_core_web_md==2.2.5) (0.6.0)\n",
201 |             "Requirement already satisfied: more-itertools in /usr/local/lib/python3.6/dist-packages (from zipp>=0.5->importlib-metadata>=0.20; python_version < \"3.8\"->catalogue<1.1.0,>=0.0.7->spacy>=2.2.2->en_core_web_md==2.2.5) (7.2.0)\n",
202 |             "Building wheels for collected packages: en-core-web-md\n",
203 |             "  Building wheel for en-core-web-md (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
204 |             "  Created wheel for en-core-web-md: filename=en_core_web_md-2.2.5-cp36-none-any.whl size=98051304 sha256=fe0516b4fb10b1d02e66d83219266ba0a33bdabdaa7617a708ac952234ce8d75\n",
205 |             "  Stored in directory: /tmp/pip-ephem-wheel-cache-zwa3ymgo/wheels/df/94/ad/f5cf59224cea6b5686ac4fd1ad19c8a07bc026e13c36502d81\n",
206 |             "Successfully built en-core-web-md\n",
207 |             "Installing collected packages: en-core-web-md\n",
208 |             "Successfully installed en-core-web-md-2.2.5\n",
209 |             "\u001b[38;5;2m✔ Download and installation successful\u001b[0m\n",
210 |             "You can now load the model via spacy.load('en_core_web_md')\n"
211 |           ],
212 |           "name": "stdout"
213 |         }
214 |       ]
215 |     },
216 |     {
217 |       "cell_type": "code",
218 |       "metadata": {
219 |         "id": "Gc4rd8JJyS9Y",
220 |         "colab_type": "code",
221 |         "colab": {}
222 |       },
223 |       "source": [
224 |         "import spacy\n",
225 |         "import en_core_web_md\n",
226 |         "#nlp = spacy.load(\"en_core_web_md\") \n",
227 |         "#nlp= spacy.load(\"en\")\n",
228 |         "nlp=en_core_web_md.load()"
229 |       ],
230 |       "execution_count": 0,
231 |       "outputs": []
232 |     },
233 |     {
234 |       "cell_type": "code",
235 |       "metadata": {
236 |         "colab_type": "code",
237 |         "id": "p8dDdyS8kgXO",
238 |         "colab": {}
239 |       },
240 |       "source": [
241 |         "sample_text=\"This is an example of language processing. It is created by Dimitrios Panagopoulos in November of 2019 at Athens, Greece. You can execute in Google's Colab.\"\n",
242 |         "doc = nlp(sample_text)"
243 |       ],
244 |       "execution_count": 0,
245 |       "outputs": []
246 |     },
247 |     {
248 |       "cell_type": "code",
249 |       "metadata": {
250 |         "id": "ge3FRXSA0pQ0",
251 |         "colab_type": "code",
252 |         "outputId": "877f3d84-1135-4900-ba5b-5a70e243632f",
253 |         "colab": {
254 |           "base_uri": "https://localhost:8080/",
255 |           "height": 586
256 |         }
257 |       },
258 |       "source": [
259 |         "for token in doc:\n",
260 |         "    print(token.text, token.lemma_, token.pos_)"
261 |       ],
262 |       "execution_count": 6,
263 |       "outputs": [
264 |         {
265 |           "output_type": "stream",
266 |           "text": [
267 |             "This this DET\n",
268 |             "is be AUX\n",
269 |             "an an DET\n",
270 |             "example example NOUN\n",
271 |             "of of ADP\n",
272 |             "language language NOUN\n",
273 |             "processing processing NOUN\n",
274 |             ". . PUNCT\n",
275 |             "It -PRON- PRON\n",
276 |             "is be AUX\n",
277 |             "created create VERB\n",
278 |             "by by ADP\n",
279 |             "Dimitrios Dimitrios PROPN\n",
280 |             "Panagopoulos Panagopoulos PROPN\n",
281 |             "in in ADP\n",
282 |             "November November PROPN\n",
283 |             "of of ADP\n",
284 |             "2019 2019 NUM\n",
285 |             "at at ADP\n",
286 |             "Athens Athens PROPN\n",
287 |             ", , PUNCT\n",
288 |             "Greece Greece PROPN\n",
289 |             ". . PUNCT\n",
290 |             "You -PRON- PRON\n",
291 |             "can can VERB\n",
292 |             "execute execute VERB\n",
293 |             "in in ADP\n",
294 |             "Google Google PROPN\n",
295 |             "'s 's PART\n",
296 |             "Colab Colab PROPN\n",
297 |             ". . PUNCT\n"
298 |           ],
299 |           "name": "stdout"
300 |         }
301 |       ]
302 |     },
303 |     {
304 |       "cell_type": "code",
305 |       "metadata": {
306 |         "id": "uwp80jJtlYUE",
307 |         "colab_type": "code",
308 |         "outputId": "d2454316-4c35-4518-ce4e-69a38dd0675b",
309 |         "colab": {
310 |           "base_uri": "https://localhost:8080/",
311 |           "height": 122
312 |         }
313 |       },
314 |       "source": [
315 |         "from spacy import displacy\n",
316 |         "displacy.render(doc, style=\"ent\", jupyter=True)"
317 |       ],
318 |       "execution_count": 7,
319 |       "outputs": [
320 |         {
321 |           "output_type": "display_data",
322 |           "data": {
323 |             "text/html": [
324 |               "<div class=\"entities\" style=\"line-height: 2.5; direction: ltr\">This is an example of language processing. It is created by \n",
325 |               "<mark class=\"entity\" style=\"background: #7aecec; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
326 |               "    Dimitrios Panagopoulos\n",
327 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">ORG</span>\n",
328 |               "</mark>\n",
329 |               " in \n",
330 |               "<mark class=\"entity\" style=\"background: #bfe1d9; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
331 |               "    November of 2019\n",
332 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">DATE</span>\n",
333 |               "</mark>\n",
334 |               " at \n",
335 |               "<mark class=\"entity\" style=\"background: #feca74; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
336 |               "    Athens\n",
337 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">GPE</span>\n",
338 |               "</mark>\n",
339 |               ", \n",
340 |               "<mark class=\"entity\" style=\"background: #feca74; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
341 |               "    Greece\n",
342 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">GPE</span>\n",
343 |               "</mark>\n",
344 |               ". You can execute in \n",
345 |               "<mark class=\"entity\" style=\"background: #7aecec; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em;\">\n",
346 |               "    Google\n",
347 |               "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">ORG</span>\n",
348 |               "</mark>\n",
349 |               "'s Colab.</div>"
350 |             ],
351 |             "text/plain": [
352 |               "<IPython.core.display.HTML object>"
353 |             ]
354 |           },
355 |           "metadata": {
356 |             "tags": []
357 |           }
358 |         }
359 |       ]
360 |     },
361 |     {
362 |       "cell_type": "code",
363 |       "metadata": {
364 |         "id": "IZXd10Tnmefz",
365 |         "colab_type": "code",
366 |         "colab": {}
367 |       },
368 |       "source": [
369 |         "sample_words=\"boy girl man\""
370 |       ],
371 |       "execution_count": 0,
372 |       "outputs": []
373 |     },
374 |     {
375 |       "cell_type": "code",
376 |       "metadata": {
377 |         "id": "Msv3wlgbpk_y",
378 |         "colab_type": "code",
379 |         "colab": {}
380 |       },
381 |       "source": [
382 |         "tokens=nlp(sample_words)"
383 |       ],
384 |       "execution_count": 0,
385 |       "outputs": []
386 |     },
387 |     {
388 |       "cell_type": "code",
389 |       "metadata": {
390 |         "id": "-hacpD7rppfw",
391 |         "colab_type": "code",
392 |         "outputId": "e514cbb6-b2be-4c40-d048-c5770fee2a82",
393 |         "colab": {
394 |           "base_uri": "https://localhost:8080/",
395 |           "height": 35
396 |         }
397 |       },
398 |       "source": [
399 |         "print(tokens)"
400 |       ],
401 |       "execution_count": 86,
402 |       "outputs": [
403 |         {
404 |           "output_type": "stream",
405 |           "text": [
406 |             "boy girl man\n"
407 |           ],
408 |           "name": "stdout"
409 |         }
410 |       ]
411 |     },
412 |     {
413 |       "cell_type": "code",
414 |       "metadata": {
415 |         "id": "0uNtAMTNpq_i",
416 |         "colab_type": "code",
417 |         "outputId": "fba8d6a0-7451-4f0a-cb2e-49bef4f0ff10",
418 |         "colab": {
419 |           "base_uri": "https://localhost:8080/",
420 |           "height": 35
421 |         }
422 |       },
423 |       "source": [
424 |         "print(tokens[0].similarity(tokens[1]))"
425 |       ],
426 |       "execution_count": 41,
427 |       "outputs": [
428 |         {
429 |           "output_type": "stream",
430 |           "text": [
431 |             "0.7401745\n"
432 |           ],
433 |           "name": "stdout"
434 |         }
435 |       ]
436 |     },
437 |     {
438 |       "cell_type": "code",
439 |       "metadata": {
440 |         "id": "6nWtntI7puNh",
441 |         "colab_type": "code",
442 |         "outputId": "093e7184-7e56-4266-920a-b4c6a216b189",
443 |         "colab": {
444 |           "base_uri": "https://localhost:8080/",
445 |           "height": 35
446 |         }
447 |       },
448 |       "source": [
449 |         "print(tokens[0].similarity(tokens[2]))"
450 |       ],
451 |       "execution_count": 42,
452 |       "outputs": [
453 |         {
454 |           "output_type": "stream",
455 |           "text": [
456 |             "0.7045701\n"
457 |           ],
458 |           "name": "stdout"
459 |         }
460 |       ]
461 |     },
462 |     {
463 |       "cell_type": "code",
464 |       "metadata": {
465 |         "id": "0e5GklwfvWFy",
466 |         "colab_type": "code",
467 |         "colab": {}
468 |       },
469 |       "source": [
470 |         "import numpy as np\n",
471 |         "from scipy.spatial import distance"
472 |       ],
473 |       "execution_count": 0,
474 |       "outputs": []
475 |     },
476 |     {
477 |       "cell_type": "code",
478 |       "metadata": {
479 |         "id": "RW384Fjsxj9i",
480 |         "colab_type": "code",
481 |         "colab": {}
482 |       },
483 |       "source": [
484 |         "p=np.array([tokens[1].vector])-np.array([tokens[0].vector])+np.array([tokens[2].vector])\n",
485 |         "#p=np.array(nlp.vocab['boy'])"
486 |       ],
487 |       "execution_count": 0,
488 |       "outputs": []
489 |     },
490 |     {
491 |       "cell_type": "code",
492 |       "metadata": {
493 |         "id": "fvEaCyDWx4xV",
494 |         "colab_type": "code",
495 |         "colab": {}
496 |       },
497 |       "source": [
498 |         "# Format the vocabulary for use in the distance function\n",
499 |         "ids = [x for x in nlp.vocab.vectors.keys()]\n",
500 |         "vectors = [nlp.vocab.vectors[x] for x in ids]\n",
501 |         "vectors = np.array(vectors)"
502 |       ],
503 |       "execution_count": 0,
504 |       "outputs": []
505 |     },
506 |     {
507 |       "cell_type": "code",
508 |       "metadata": {
509 |         "id": "i0EPtc3QxyOq",
510 |         "colab_type": "code",
511 |         "colab": {}
512 |       },
513 |       "source": [
514 |         "closest_index = distance.cdist(p, vectors, metric='cosine').argmin() # change from  argmin() because distance calculate distance not similariy\n",
515 |         "word_id = ids[closest_index]\n",
516 |         "output_word = nlp.vocab[word_id].text"
517 |       ],
518 |       "execution_count": 0,
519 |       "outputs": []
520 |     },
521 |     {
522 |       "cell_type": "code",
523 |       "metadata": {
524 |         "id": "aZE8zAVAyI2Z",
525 |         "colab_type": "code",
526 |         "outputId": "5ac6d065-d377-40bc-a114-afa3b1b427dc",
527 |         "colab": {
528 |           "base_uri": "https://localhost:8080/",
529 |           "height": 35
530 |         }
531 |       },
532 |       "source": [
533 |         "print(output_word)"
534 |       ],
535 |       "execution_count": 89,
536 |       "outputs": [
537 |         {
538 |           "output_type": "stream",
539 |           "text": [
540 |             "woman\n"
541 |           ],
542 |           "name": "stdout"
543 |         }
544 |       ]
545 |     }
546 |   ]
547 | }


--------------------------------------------------------------------------------
/convolution_example.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/dpanagop/ML_and_AI_examples/675bada7b9fa4c62035b0891d08cc32d0f7a195e/convolution_example.xlsx


--------------------------------------------------------------------------------
/data/README.md:
--------------------------------------------------------------------------------
1 | 
2 | 


--------------------------------------------------------------------------------