AutoKeras.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cf254b68",
   "metadata": {},
   "outputs": [],
   "source": [
    "# The user only specifies the input nodes and output heads.\n",
    "import autokeras as ak\n",
    "from keras.datasets import mnist\n",
    "from autokeras import ImageClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b595e403",
   "metadata": {},
   "outputs": [],
   "source": [
    "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
    "x_train = x_train.reshape(x_train.shape + (1,)) # (1,) denotes the channles which is 1 in this case\n",
    "x_test = x_test.reshape(x_test.shape + (1,)) # (1,) denotes the channles which is 1 in this case"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9800df04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 1 Complete [00h 35m 25s]\n",
      "val_loss: 0.040560998022556305\n",
      "\n",
      "Best val_loss So Far: 0.040560998022556305\n",
      "Total elapsed time: 00h 35m 25s\n",
      "\n",
      "Search: Running Trial #2\n",
      "\n",
      "Hyperparameter    |Value             |Best Value So Far \n",
      "image_block_1/b...|resnet            |vanilla           \n",
      "image_block_1/n...|True              |True              \n",
      "image_block_1/a...|True              |False             \n",
      "image_block_1/i...|True              |None              \n",
      "image_block_1/i...|True              |None              \n",
      "image_block_1/i...|0                 |None              \n",
      "image_block_1/i...|0                 |None              \n",
      "image_block_1/i...|0.1               |None              \n",
      "image_block_1/i...|0                 |None              \n",
      "image_block_1/r...|False             |None              \n",
      "image_block_1/r...|resnet50          |None              \n",
      "image_block_1/r...|True              |None              \n",
      "classification_...|global_avg        |flatten           \n",
      "classification_...|0                 |0.5               \n",
      "optimizer         |adam              |adam              \n",
      "learning_rate     |0.001             |0.001             \n",
      "\n",
      "Epoch 1/1000\n",
      " 128/1500 [=>............................] - ETA: 6:03:54 - loss: 1.3791 - accuracy: 0.5776"
     ]
    }
   ],
   "source": [
    "# Instantiate the ImageClassifier class\n",
    "clf = ImageClassifier(num_classes=9)\n",
    "# Fit the train set to the image classifier\n",
    "clf.fit(x_train, y_train)\n",
    "clf.final_fit(x_train, y_train, x_test, y_test, retrain=True)\n",
    "# Summarize the results\n",
    "y = clf.evaluate(x_test, y_test)\n",
    "print(y * 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "505e920d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "environment": {
   "name": "tf2-gpu.2-3.m76",
   "type": "gcloud",
   "uri": "gcr.io/deeplearning-platform-release/tf2-gpu.2-3:m76"
  },
  "kernelspec": {
   "display_name": "Python [conda env:root] *",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}