tensorflow practice.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6bd6c70c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensorflow flow version:  2.12.0\n",
      "2.12.0\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "print(\"tensorflow flow version: \", tf.__version__)\n",
    "print(keras.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d46ca558",
   "metadata": {},
   "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"
     ]
    }
   ],
   "source": [
    "#load data\n",
    "from keras.datasets import mnist\n",
    "(X_train, y_train),(X_test,y_test)=mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c6995cbc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 28, 28)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5eff848d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000,)\n"
     ]
    }
   ],
   "source": [
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "74eb1d90",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10000, 28, 28)\n"
     ]
    }
   ],
   "source": [
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "54d416d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "#create mode, sequence model, \n",
    "mlp = keras.models.Sequential()\n",
    "mlp.add(keras.layers.Flatten(input_shape=[28,28]))\n",
    "mlp.add(keras.layers.Dense(300,activation='relu'))\n",
    "mlp.add(keras.layers.Dense(150,activation='relu'))\n",
    "mlp.add(keras.layers.Dense(10,activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4b3d0dee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " flatten_1 (Flatten)         (None, 784)               0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 300)               235500    \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 150)               45150     \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 10)                1510      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 282,160\n",
      "Trainable params: 282,160\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "mlp.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d139392f",
   "metadata": {},
   "outputs": [],
   "source": [
    "mlp.compile(\n",
    "        loss='sparse_categorical_crossentropy',\n",
    "        optimizer='adam',\n",
    "        metrics=['accuracy']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "8f74a284",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "1688/1688 [==============================] - 5s 3ms/step - loss: 1.5253 - accuracy: 0.8859 - val_loss: 0.2810 - val_accuracy: 0.9372\n",
      "Epoch 2/10\n",
      "1688/1688 [==============================] - 5s 3ms/step - loss: 0.2370 - accuracy: 0.9379 - val_loss: 0.2371 - val_accuracy: 0.9492\n",
      "Epoch 3/10\n",
      "1688/1688 [==============================] - 8s 5ms/step - loss: 0.1843 - accuracy: 0.9519 - val_loss: 0.1783 - val_accuracy: 0.9493\n",
      "Epoch 4/10\n",
      "1688/1688 [==============================] - 8s 5ms/step - loss: 0.1566 - accuracy: 0.9560 - val_loss: 0.1539 - val_accuracy: 0.9517\n",
      "Epoch 5/10\n",
      "1688/1688 [==============================] - 9s 5ms/step - loss: 0.1449 - accuracy: 0.9605 - val_loss: 0.1625 - val_accuracy: 0.9642\n",
      "Epoch 6/10\n",
      "1688/1688 [==============================] - 7s 4ms/step - loss: 0.1320 - accuracy: 0.9644 - val_loss: 0.1733 - val_accuracy: 0.9618\n",
      "Epoch 7/10\n",
      "1688/1688 [==============================] - 9s 5ms/step - loss: 0.1160 - accuracy: 0.9696 - val_loss: 0.1755 - val_accuracy: 0.9675\n",
      "Epoch 8/10\n",
      "1688/1688 [==============================] - 8s 5ms/step - loss: 0.1049 - accuracy: 0.9732 - val_loss: 0.1260 - val_accuracy: 0.9718\n",
      "Epoch 9/10\n",
      "1688/1688 [==============================] - 9s 6ms/step - loss: 0.1029 - accuracy: 0.9742 - val_loss: 0.1311 - val_accuracy: 0.9735\n",
      "Epoch 10/10\n",
      "1688/1688 [==============================] - 9s 5ms/step - loss: 0.1033 - accuracy: 0.9757 - val_loss: 0.1469 - val_accuracy: 0.9688\n"
     ]
    }
   ],
   "source": [
    "log = mlp.fit(X_train, y_train, epochs=10, validation_split=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "62444180",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: >"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "pd.DataFrame(log.history).plot()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7b2edcad",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 76ms/step\n"
     ]
    }
   ],
   "source": [
    "X_new = X_test[:10]\n",
    "y_hat = mlp.predict(X_new)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5427c420",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}