machinelearning

Machine learning code
git clone git://git.laack.co/machinelearning.git
Log | Files | Refs

MNISTKerasFollowAlong.ipynb (16740B)

      1 {
      2  "cells": [
      3   {
      4    "cell_type": "code",
      5    "execution_count": 49,
      6    "metadata": {},
      7    "outputs": [],
      8    "source": [
      9     "from tensorflow.keras.datasets import mnist\n",
     10     "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()"
     11    ]
     12   },
     13   {
     14    "cell_type": "code",
     15    "execution_count": 50,
     16    "metadata": {},
     17    "outputs": [
     18     {
     19      "data": {
     20       "text/plain": [
     21        "dtype('uint8')"
     22       ]
     23      },
     24      "execution_count": 50,
     25      "metadata": {},
     26      "output_type": "execute_result"
     27     }
     28    ],
     29    "source": [
     30     "train_images.dtype"
     31    ]
     32   },
     33   {
     34    "cell_type": "code",
     35    "execution_count": 51,
     36    "metadata": {},
     37    "outputs": [],
     38    "source": [
     39     "import keras\n",
     40     "\n",
     41     "model = keras.Sequential(layers=[\n",
     42     "    keras.layers.Dense(512, 'relu'),\n",
     43     "    keras.layers.Dense(10, 'softmax')\n",
     44     "])"
     45    ]
     46   },
     47   {
     48    "cell_type": "code",
     49    "execution_count": 52,
     50    "metadata": {},
     51    "outputs": [],
     52    "source": [
     53     "model.compile(optimizer='rmsprop', loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
     54    ]
     55   },
     56   {
     57    "cell_type": "code",
     58    "execution_count": 53,
     59    "metadata": {},
     60    "outputs": [],
     61    "source": [
     62     "train_images = train_images.reshape((60000, 28*28))\n",
     63     "train_images = train_images.astype(\"float32\") / 255\n",
     64     "test_images = test_images.reshape((10000, 28*28))\n",
     65     "test_images = test_images.astype(\"float32\") / 255"
     66    ]
     67   },
     68   {
     69    "cell_type": "code",
     70    "execution_count": 54,
     71    "metadata": {},
     72    "outputs": [
     73     {
     74      "name": "stdout",
     75      "output_type": "stream",
     76      "text": [
     77       "Epoch 1/5\n",
     78       "\u001b[1m469/469\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 3ms/step - accuracy: 0.8704 - loss: 0.4457\n",
     79       "Epoch 2/5\n",
     80       "\u001b[1m469/469\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.9660 - loss: 0.1141\n",
     81       "Epoch 3/5\n",
     82       "\u001b[1m469/469\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.9785 - loss: 0.0732\n",
     83       "Epoch 4/5\n",
     84       "\u001b[1m469/469\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.9841 - loss: 0.0523\n",
     85       "Epoch 5/5\n",
     86       "\u001b[1m469/469\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.9894 - loss: 0.0366\n"
     87      ]
     88     },
     89     {
     90      "data": {
     91       "text/plain": [
     92        "<keras.src.callbacks.history.History at 0x7fa0fe4500d0>"
     93       ]
     94      },
     95      "execution_count": 54,
     96      "metadata": {},
     97      "output_type": "execute_result"
     98     }
     99    ],
    100    "source": [
    101     "model.fit(train_images, train_labels, epochs=5, batch_size=128)"
    102    ]
    103   },
    104   {
    105    "cell_type": "code",
    106    "execution_count": 55,
    107    "metadata": {},
    108    "outputs": [
    109     {
    110      "name": "stdout",
    111      "output_type": "stream",
    112      "text": [
    113       "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n"
    114      ]
    115     }
    116    ],
    117    "source": [
    118     "test_digits = test_images[0:10]\n",
    119     "predictions = model.predict(test_digits)"
    120    ]
    121   },
    122   {
    123    "cell_type": "code",
    124    "execution_count": 56,
    125    "metadata": {},
    126    "outputs": [
    127     {
    128      "data": {
    129       "text/plain": [
    130        "7"
    131       ]
    132      },
    133      "execution_count": 56,
    134      "metadata": {},
    135      "output_type": "execute_result"
    136     }
    137    ],
    138    "source": [
    139     "predictions[0].argmax()"
    140    ]
    141   },
    142   {
    143    "cell_type": "code",
    144    "execution_count": 62,
    145    "metadata": {},
    146    "outputs": [
    147     {
    148      "data": {
    149       "text/plain": [
    150        "<matplotlib.image.AxesImage at 0x7fa0fe283510>"
    151       ]
    152      },
    153      "execution_count": 62,
    154      "metadata": {},
    155      "output_type": "execute_result"
    156     },
    157     {
    158      "data": {
    159       "image/png": "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",
    160       "text/plain": [
    161        "<Figure size 640x480 with 1 Axes>"
    162       ]
    163      },
    164      "metadata": {},
    165      "output_type": "display_data"
    166     }
    167    ],
    168    "source": [
    169     "import matplotlib.pyplot as plt\n",
    170     "\n",
    171     "plt.imshow(test_digits[0].reshape(28,28), cmap='binary')"
    172    ]
    173   },
    174   {
    175    "cell_type": "code",
    176    "execution_count": 58,
    177    "metadata": {},
    178    "outputs": [
    179     {
    180      "name": "stdout",
    181      "output_type": "stream",
    182      "text": [
    183       "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 826us/step - accuracy: 0.9767 - loss: 0.0756\n"
    184      ]
    185     }
    186    ],
    187    "source": [
    188     "test_loss, test_acc = model.evaluate(test_images, test_labels)"
    189    ]
    190   },
    191   {
    192    "cell_type": "code",
    193    "execution_count": null,
    194    "metadata": {},
    195    "outputs": [
    196     {
    197      "data": {
    198       "text/plain": [
    199        "0.9807999730110168"
    200       ]
    201      },
    202      "execution_count": 59,
    203      "metadata": {},
    204      "output_type": "execute_result"
    205     }
    206    ],
    207    "source": [
    208     "test_acc"
    209    ]
    210   },
    211   {
    212    "cell_type": "code",
    213    "execution_count": 70,
    214    "metadata": {},
    215    "outputs": [
    216     {
    217      "data": {
    218       "text/plain": [
    219        "array([8.13043511, 9.1062374 , 6.10387794, 7.96415923, 6.70079762,\n",
    220        "       9.00702563, 7.36787942, 8.07919392, 7.57392299, 8.72668414,\n",
    221        "       7.86908924, 8.01073825, 8.05753397, 7.5316775 , 8.94442168,\n",
    222        "       7.8178981 , 7.64151114, 7.25169205, 8.33018161, 6.22029134,\n",
    223        "       6.81264197, 8.98350502, 8.26812517, 7.09232707, 6.67108924,\n",
    224        "       7.52754293, 8.72975814, 6.79826261, 8.97567685, 8.64827915,\n",
    225        "       7.1033046 , 8.20552579])"
    226       ]
    227      },
    228      "execution_count": 70,
    229      "metadata": {},
    230      "output_type": "execute_result"
    231     }
    232    ],
    233    "source": [
    234     "import numpy as np\n",
    235     "x = np.random.random((32, 32))\n",
    236     "y = np.random.random((32, 32))\n",
    237     "np.dot(x, y)[0]"
    238    ]
    239   },
    240   {
    241    "cell_type": "code",
    242    "execution_count": 71,
    243    "metadata": {},
    244    "outputs": [
    245     {
    246      "data": {
    247       "text/plain": [
    248        "array([8.13043511, 9.1062374 , 6.10387794, 7.96415923, 6.70079762,\n",
    249        "       9.00702563, 7.36787942, 8.07919392, 7.57392299, 8.72668414,\n",
    250        "       7.86908924, 8.01073825, 8.05753397, 7.5316775 , 8.94442168,\n",
    251        "       7.8178981 , 7.64151114, 7.25169205, 8.33018161, 6.22029134,\n",
    252        "       6.81264197, 8.98350502, 8.26812517, 7.09232707, 6.67108924,\n",
    253        "       7.52754293, 8.72975814, 6.79826261, 8.97567685, 8.64827915,\n",
    254        "       7.1033046 , 8.20552579])"
    255       ]
    256      },
    257      "execution_count": 71,
    258      "metadata": {},
    259      "output_type": "execute_result"
    260     }
    261    ],
    262    "source": [
    263     "(x @ y)[0]"
    264    ]
    265   }
    266  ],
    267  "metadata": {
    268   "kernelspec": {
    269    "display_name": ".venv",
    270    "language": "python",
    271    "name": "python3"
    272   },
    273   "language_info": {
    274    "codemirror_mode": {
    275     "name": "ipython",
    276     "version": 3
    277    },
    278    "file_extension": ".py",
    279    "mimetype": "text/x-python",
    280    "name": "python",
    281    "nbconvert_exporter": "python",
    282    "pygments_lexer": "ipython3",
    283    "version": "3.11.2"
    284   }
    285  },
    286  "nbformat": 4,
    287  "nbformat_minor": 2
    288 }