ConvAE_pwat_perpetual_Aug_2008-2017_1000epochs.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import load_model\n",
    "from mod_gendata import *\n",
    "import matplotlib.pyplot as plt\n",
    "from mod_cutregion import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "autoencoder = load_model(\"./pwat_convae1000_2knl_addchnl_peraug.h5\") # models9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 64, 64, 1)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 32, 32, 8)         40        \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 16, 16, 16)        528       \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 8, 8, 32)          2080      \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 4, 4, 64)          8256      \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 4, 4, 128)         32896     \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_1 (Conv2DTr (None, 8, 8, 64)          32832     \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 8, 8, 32)          8224      \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_2 (Conv2DTr (None, 16, 16, 32)        4128      \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 16, 16, 16)        2064      \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_3 (Conv2DTr (None, 32, 32, 8)         520       \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 32, 32, 8)         264       \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_4 (Conv2DTr (None, 64, 64, 4)         132       \n",
      "_________________________________________________________________\n",
      "conv2d_10 (Conv2D)           (None, 64, 64, 1)         17        \n",
      "=================================================================\n",
      "Total params: 91,981\n",
      "Trainable params: 91,981\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "# model archetecture \n",
    "print(autoencoder.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Max value   93.2\n"
     ]
    }
   ],
   "source": [
    "# load test dataset\n",
    "keyward = 'pwat'\n",
    "lat     = 181 ; lon = 360 \n",
    "testdir= '/home/kurihana/ml_model/work_mymodel/ex4/data/test_data'\n",
    "gd     = gen_grads_data()\n",
    "x_test = gd.load_key_data(testdir, keyward)\n",
    "x_test = x_test.reshape(x_test.shape[0],lat,lon)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80, 64, 64)\n"
     ]
    }
   ],
   "source": [
    "# Select Region\n",
    "sg = sl_region()\n",
    "\n",
    "# Descripition fot \"get_region\" API\n",
    "#\n",
    "# get_region(data, data_dimension, all_lat, all_lon. nlat, slat, wlon, elon)\n",
    "# all_lat/lon : original data's lat/lon\n",
    "# nlat = north latitude in selected region\n",
    "\n",
    "x_test  = sg.get_region(x_test, 3, lat, lon, 64, 0, 225, 289)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add channel dimension\n",
    "nlat = 64 # based on get_region number\n",
    "nlon = 64\n",
    "x_test = x_test.reshape(x_test.shape[0], nlat, nlon, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "decoded_imgs = autoencoder.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n = 2\n",
    "plt.figure(figsize=(16,8))\n",
    "image_list = [0, 12]\n",
    "#for i in range(n):\n",
    "for index, i in enumerate(image_list):\n",
    "     # test data\n",
    "     ax = plt.subplot(2,n,index+1)\n",
    "     plt.imshow(x_test[i].reshape(nlat,nlon))\n",
    "     plt.gray()\n",
    "     itime = int(i*6)\n",
    "     plt.title('+ %d hour'% itime )\n",
    "     ax.get_xaxis().set_visible(False)\n",
    "     ax.get_yaxis().set_visible(False)\n",
    "\n",
    "     # decoded data \n",
    "     ax = plt.subplot(2,n,index+1+n)\n",
    "     plt.imshow(decoded_imgs[i].reshape(nlat,nlon))\n",
    "     plt.gray()\n",
    "     ax.get_xaxis().set_visible(False)\n",
    "     ax.get_yaxis().set_visible(False)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compare with shallower network and channels\n",
    "# ====================  models8 vz models9 ====================\n",
    "autoencoder = load_model(\"./pwat_convae1000_2knldeeper_peraug.h5\")  # models8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 64, 64, 1)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 32, 32, 8)         40        \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 16, 16, 16)        528       \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 8, 8, 32)          2080      \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 8, 8, 64)          8256      \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_1 (Conv2DTr (None, 16, 16, 32)        8224      \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 16, 16, 16)        2064      \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_2 (Conv2DTr (None, 32, 32, 8)         520       \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 32, 32, 8)         264       \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_3 (Conv2DTr (None, 64, 64, 4)         132       \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 64, 64, 1)         17        \n",
      "=================================================================\n",
      "Total params: 22,125\n",
      "Trainable params: 22,125\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "# model archetecture \n",
    "print(autoencoder.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Max value   93.2\n"
     ]
    }
   ],
   "source": [
    "# load test dataset\n",
    "keyward = 'pwat'\n",
    "lat     = 181 ; lon = 360 \n",
    "testdir= '/home/kurihana/ml_model/work_mymodel/ex4/data/test_data'\n",
    "gd     = gen_grads_data()\n",
    "x_test = gd.load_key_data(testdir, keyward)\n",
    "x_test = x_test.reshape(x_test.shape[0],lat,lon)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(80, 64, 64)\n"
     ]
    }
   ],
   "source": [
    "# Select Region\n",
    "sg = sl_region()\n",
    "\n",
    "# Descripition fot \"get_region\" API\n",
    "#\n",
    "# get_region(data, data_dimension, all_lat, all_lon. nlat, slat, wlon, elon)\n",
    "# all_lat/lon : original data's lat/lon\n",
    "# nlat = north latitude in selected region\n",
    "\n",
    "x_test  = sg.get_region(x_test, 3, lat, lon, 64, 0, 225, 289)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add channel dimension\n",
    "nlat = 64 # based on get_region number\n",
    "nlon = 64\n",
    "x_test = x_test.reshape(x_test.shape[0], nlat, nlon, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "decoded_imgs = autoencoder.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n = 2\n",
    "plt.figure(figsize=(16,8))\n",
    "image_list = [0, 12]\n",
    "#for i in range(n):\n",
    "for index, i in enumerate(image_list):\n",
    "     # test data\n",
    "     ax = plt.subplot(2,n,index+1)\n",
    "     plt.imshow(x_test[i].reshape(nlat,nlon))\n",
    "     plt.gray()\n",
    "     itime = int(i*6)\n",
    "     plt.title('+ %d hour'% itime )\n",
    "     ax.get_xaxis().set_visible(False)\n",
    "     ax.get_yaxis().set_visible(False)\n",
    "\n",
    "     # decoded data \n",
    "     ax = plt.subplot(2,n,index+1+n)\n",
    "     plt.imshow(decoded_imgs[i].reshape(nlat,nlon))\n",
    "     plt.gray()\n",
    "     ax.get_xaxis().set_visible(False)\n",
    "     ax.get_yaxis().set_visible(False)\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}