02_keras.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"TF_CPP_MIN_LOG_LEVEL\"] = '2'\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
      "11490434/11490434 [==============================] - 2s 0us/step\n",
      "[[[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]] [5 0 4 ... 5 6 8] [[[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]\n",
      "\n",
      " [[0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  ...\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]\n",
      "  [0 0 0 ... 0 0 0]]] [7 2 1 ... 4 5 6]\n"
     ]
    }
   ],
   "source": [
    "mnist = keras.datasets.mnist\n",
    "\n",
    "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
    "print(x_train, y_train, x_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Normalize data\n",
    "x_train, x_test = x_train / 255.0, x_test / 255.0\n",
    "\n",
    "for i in range(6):\n",
    "  plt.subplot(2, 3, i+1)\n",
    "  plt.imshow(x_train[i], cmap='gray')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_7\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " flatten_7 (Flatten)         (None, 784)               0         \n",
      "                                                                 \n",
      " dense_14 (Dense)            (None, 128)               100480    \n",
      "                                                                 \n",
      " dense_15 (Dense)            (None, 10)                1290      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 101,770\n",
      "Trainable params: 101,770\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#model\n",
    "model = keras.models.Sequential([\n",
    "  keras.layers.Flatten(input_shape=(28, 28)),\n",
    "  keras.layers.Dense(128, activation='relu'), # Fully connected layer\n",
    "  keras.layers.Dense(10),\n",
    "])\n",
    "\n",
    "model.build()\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "loss = keras.losses.SparseCategoricalCrossentropy(from_logits=True)\n",
    "optimizer = keras.optimizers.Adam(learning_rate=0.001)\n",
    "metrics = [\"accuracy\"]\n",
    "\n",
    "model.compile(loss=loss, optimizer=optimizer, metrics=metrics)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "469/469 - 3s - loss: 0.3562 - accuracy: 0.9025 - 3s/epoch - 5ms/step\n",
      "Epoch 2/10\n",
      "469/469 - 2s - loss: 0.1640 - accuracy: 0.9531 - 2s/epoch - 3ms/step\n",
      "Epoch 3/10\n",
      "469/469 - 2s - loss: 0.1161 - accuracy: 0.9668 - 2s/epoch - 4ms/step\n",
      "Epoch 4/10\n",
      "469/469 - 2s - loss: 0.0895 - accuracy: 0.9743 - 2s/epoch - 5ms/step\n",
      "Epoch 5/10\n",
      "469/469 - 2s - loss: 0.0724 - accuracy: 0.9790 - 2s/epoch - 4ms/step\n",
      "Epoch 6/10\n",
      "469/469 - 2s - loss: 0.0595 - accuracy: 0.9834 - 2s/epoch - 5ms/step\n",
      "Epoch 7/10\n",
      "469/469 - 2s - loss: 0.0496 - accuracy: 0.9859 - 2s/epoch - 5ms/step\n",
      "Epoch 8/10\n",
      "469/469 - 3s - loss: 0.0426 - accuracy: 0.9881 - 3s/epoch - 5ms/step\n",
      "Epoch 9/10\n",
      "469/469 - 2s - loss: 0.0358 - accuracy: 0.9898 - 2s/epoch - 5ms/step\n",
      "Epoch 10/10\n",
      "469/469 - 2s - loss: 0.0309 - accuracy: 0.9917 - 2s/epoch - 4ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x20251287b50>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "batch_size=128\n",
    "epochs=10\n",
    "\n",
    "model.fit(x_train, y_train, batch_size, epochs, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "79/79 - 0s - loss: 0.0740 - accuracy: 0.9769 - 455ms/epoch - 6ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.07403256744146347, 0.9768999814987183]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(x_test, y_test, batch_size, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(\n",
      "[1.0322480e-06 4.3681766e-08 1.2059590e-05 4.9117842e-04 5.2129967e-12\n",
      " 2.5690835e-07 4.4473165e-13 9.9946207e-01 3.5477246e-06 2.9740622e-05], shape=(10,), dtype=float32)\n"
     ]
    }
   ],
   "source": [
    "probability_model = keras.models.Sequential([\n",
    "  model,\n",
    "  keras.layers.Softmax()\n",
    "])\n",
    "\n",
    "predictions = probability_model(x_test)\n",
    "pred0 = predictions[0]\n",
    "print(pred0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7\n"
     ]
    }
   ],
   "source": [
    "label0 = np.argmax(pred0)\n",
    "print(label0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 0s 1ms/step\n",
      "tf.Tensor(\n",
      "[1.0322469e-06 4.3681766e-08 1.2059590e-05 4.9117749e-04 5.2130167e-12\n",
      " 2.5690787e-07 4.4473079e-13 9.9946207e-01 3.5477246e-06 2.9740566e-05], shape=(10,), dtype=float32)\n",
      "7\n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(x_test)\n",
    "predictions = tf.nn.softmax(predictions)\n",
    "pred0 = predictions[0]\n",
    "print(pred0)\n",
    "\n",
    "label0 = np.argmax(pred0)\n",
    "print(label0)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.6 ('venv': venv)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "fec9c3e99cf5c08fbdf195f91c54385bfa69aefe0d922a474aa56741c1f904bb"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}