save-load-model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ذخیره کردن و بارگزاری وزن های مدل آموزش داده شده"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "به صورت پیشفرض مدل ها در تنسورفلو 2 با فرمت `SavedModel ` ذخیره می شوند. البته می توانیم مدل ها را به صورت فرمت h5 هم ذخیره کنیم."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Manually save weights\n",
    "You saw how to load the weights into a model. Manually saving them is just as simple with the Model.save_weights method. By default, tf.keras—and save_weights in particular—uses the TensorFlow checkpoint format with a .ckpt extension (saving in HDF5 with a .h5 extension is covered in the Save and serialize models guide):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Keras saves models by inspecting the architecture. This technique saves everything:\n",
    "\n",
    "The weight values\n",
    "The model's architecture\n",
    "The model's training configuration(what you passed to compile)\n",
    "The optimizer and its state, if any (this enables you to restart training where you left)\n",
    "\n",
    "Save the entire model\n",
    "Call model.save to save a model's architecture, weights, and training configuration in a single file/folder. This allows you to export a model so it can be used without access to the original Python code*. Since the optimizer-state is recovered, you can resume training from exactly where you left off."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "برای ذخیره سازی مدل ها در فرمت h5py نیاز به نصب کتابخانه h5py داریم.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#!pip install -q pyyaml h5py "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('my_model.h5') "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The restored model is compiled with the same arguments as the original model. Try running evaluate and predict with the loaded model:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### دخیره کردن مدلی که از توابع، لایه های که خودمان ساخته ایم استفاده می کند."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('my_model.h5') "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "منتها وقتی که مدل را بارگزاری می کنیم باید حواسمان باشد که "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model  = keras.models.load_model(\n",
    "'my_model.h5',\n",
    "    custom_objects = {'mape_fn':mape_fn})"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}