s2_sharpening.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "pycharm": {
     "is_executing": false
    }
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "\n",
    "from tensorflow.keras.models import Sequential, Model\n",
    "from tensorflow.keras.layers import Conv2D, UpSampling2D, Input, BatchNormalization,Activation\n",
    "from tensorflow.keras.layers import LeakyReLU,Concatenate, Conv2DTranspose\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "# from models.unet import *\n",
    "from models.skipnet import *\n",
    "from models.resnet import *\n",
    "from utils.common import *\n",
    "from PIL import Image\n",
    "import matplotlib.pyplot as plt\n",
    "from skimage.transform import rescale, pyramid_reduce\n",
    "import os\n",
    "import scipy.io as sio\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6831\n"
     ]
    }
   ],
   "source": [
    "# Load data (data are preprocessed by Matlab)\n",
    "coastal_data=False\n",
    "if coastal_data:     #Costal data\n",
    "    dname=\"coastal\"\n",
    "    datapath=\"data/coastal.mat\"\n",
    "    coastal=sio.loadmat(datapath)\n",
    "    X=coastal[\"xim\"]\n",
    "    scale=np.max(X)\n",
    "    x=X/scale #ground truth (10 m band)\n",
    "    y=coastal[\"yD\"]/scale #input (20 m band)\n",
    "    xD=coastal[\"xU\"] #interpolated 20 m bands to spatial size of 10 m band\n",
    "else:                #Reykjavik\n",
    "    dname=\"rek\"\n",
    "    datapath=\"data/reykjavik.mat\"\n",
    "    data = sio.loadmat(datapath)\n",
    "    X=data[\"x20\"] #ground truth\n",
    "    scale=np.max(X)\n",
    "    x=X/scale\n",
    "    y=data[\"yD\"]/scale #input\n",
    "    \n",
    "#     We can use another downsampling technique instead of bicubic\n",
    "    # yD=pyramid_reduce(y)\n",
    "    # xD=rescale(pyramid_reduce(x),scale=r,order=5)\n",
    "print(scale)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(961, 16, 16, 10) (1, 256, 256, 6)\n"
     ]
    }
   ],
   "source": [
    "Y=GenerateCube(y,stride=8) #input\n",
    "X=GenerateCube(x,stride=8) #target\n",
    "Yt=y.reshape(-1,y.shape[0],y.shape[1],y.shape[2])\n",
    "Xt=x.reshape(-1,x.shape[0],x.shape[1],x.shape[2])\n",
    "print(Y.shape,Xt.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_epochs = 200\n",
    "\n",
    "skipnet=False\n",
    "\n",
    "if skipnet:\n",
    "    netname=\"skipnet\"\n",
    "    batch_size= 64\n",
    "    mymodel=skip(ndown=3)\n",
    "    myoptimizer=tf.keras.optimizers.Adam(learning_rate=0.0005)\n",
    "    mymodel.compile(optimizer=myoptimizer, loss=\"mse\")\n",
    "else:\n",
    "    netname=\"resnet\"\n",
    "    batch_size= 48\n",
    "    mymodel=residualnet(K=16)\n",
    "    myoptimizer=tf.keras.optimizers.Adam(learning_rate=0.0004)\n",
    "    mymodel.compile(optimizer=myoptimizer, loss=\"mse\")\n",
    "# mymodel.summary()\n",
    "modelpath=netname+\"bestmodel.hdf5\"\n",
    "checkpoint = ModelCheckpoint(modelpath, monitor='val_loss', verbose=1, save_best_only=True)\n",
    "callbacks_list = [checkpoint]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 961 samples, validate on 1 samples\n",
      "Epoch 1/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.1528\n",
      "Epoch 00001: val_loss improved from inf to 0.00325, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 4s 4ms/sample - loss: 0.1453 - val_loss: 0.0032\n",
      "Epoch 2/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 0.0021\n",
      "Epoch 00002: val_loss improved from 0.00325 to 0.00146, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 788us/sample - loss: 0.0020 - val_loss: 0.0015\n",
      "Epoch 3/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 0.0015\n",
      "Epoch 00003: val_loss improved from 0.00146 to 0.00131, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 803us/sample - loss: 0.0015 - val_loss: 0.0013\n",
      "Epoch 4/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 0.0014\n",
      "Epoch 00004: val_loss improved from 0.00131 to 0.00127, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 765us/sample - loss: 0.0013 - val_loss: 0.0013\n",
      "Epoch 5/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0013\n",
      "Epoch 00005: val_loss improved from 0.00127 to 0.00123, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 812us/sample - loss: 0.0013 - val_loss: 0.0012\n",
      "Epoch 6/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0013\n",
      "Epoch 00006: val_loss improved from 0.00123 to 0.00119, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 744us/sample - loss: 0.0013 - val_loss: 0.0012\n",
      "Epoch 7/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0012\n",
      "Epoch 00007: val_loss improved from 0.00119 to 0.00117, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 771us/sample - loss: 0.0012 - val_loss: 0.0012\n",
      "Epoch 8/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0012\n",
      "Epoch 00008: val_loss improved from 0.00117 to 0.00114, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 758us/sample - loss: 0.0012 - val_loss: 0.0011\n",
      "Epoch 9/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 0.0012\n",
      "Epoch 00009: val_loss improved from 0.00114 to 0.00109, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 776us/sample - loss: 0.0012 - val_loss: 0.0011\n",
      "Epoch 10/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0012\n",
      "Epoch 00010: val_loss improved from 0.00109 to 0.00106, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 748us/sample - loss: 0.0012 - val_loss: 0.0011\n",
      "Epoch 11/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0011\n",
      "Epoch 00011: val_loss improved from 0.00106 to 0.00102, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 789us/sample - loss: 0.0011 - val_loss: 0.0010\n",
      "Epoch 12/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0011\n",
      "Epoch 00012: val_loss improved from 0.00102 to 0.00096, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 744us/sample - loss: 0.0011 - val_loss: 9.5585e-04\n",
      "Epoch 13/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 0.0011\n",
      "Epoch 00013: val_loss improved from 0.00096 to 0.00089, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 759us/sample - loss: 0.0011 - val_loss: 8.9167e-04\n",
      "Epoch 14/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 0.0010 \n",
      "Epoch 00014: val_loss improved from 0.00089 to 0.00082, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 735us/sample - loss: 0.0010 - val_loss: 8.2130e-04\n",
      "Epoch 15/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 9.5501e-04\n",
      "Epoch 00015: val_loss improved from 0.00082 to 0.00077, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 771us/sample - loss: 9.5291e-04 - val_loss: 7.7236e-04\n",
      "Epoch 16/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 9.0799e-04\n",
      "Epoch 00016: val_loss improved from 0.00077 to 0.00072, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 746us/sample - loss: 9.0757e-04 - val_loss: 7.2102e-04\n",
      "Epoch 17/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 8.6898e-04\n",
      "Epoch 00017: val_loss improved from 0.00072 to 0.00067, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 772us/sample - loss: 8.6449e-04 - val_loss: 6.7084e-04\n",
      "Epoch 18/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 8.2296e-04\n",
      "Epoch 00018: val_loss improved from 0.00067 to 0.00063, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 741us/sample - loss: 8.2248e-04 - val_loss: 6.3376e-04\n",
      "Epoch 19/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 7.8694e-04\n",
      "Epoch 00019: val_loss improved from 0.00063 to 0.00060, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 770us/sample - loss: 7.8707e-04 - val_loss: 6.0125e-04\n",
      "Epoch 20/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 7.5931e-04\n",
      "Epoch 00020: val_loss improved from 0.00060 to 0.00058, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 722us/sample - loss: 7.5779e-04 - val_loss: 5.7539e-04\n",
      "Epoch 21/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 7.3216e-04\n",
      "Epoch 00021: val_loss improved from 0.00058 to 0.00055, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 767us/sample - loss: 7.2992e-04 - val_loss: 5.5371e-04\n",
      "Epoch 22/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 7.0339e-04\n",
      "Epoch 00022: val_loss improved from 0.00055 to 0.00053, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 745us/sample - loss: 7.0430e-04 - val_loss: 5.3414e-04\n",
      "Epoch 23/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 6.8971e-04\n",
      "Epoch 00023: val_loss improved from 0.00053 to 0.00052, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 777us/sample - loss: 6.8291e-04 - val_loss: 5.1935e-04\n",
      "Epoch 24/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 6.5849e-04\n",
      "Epoch 00024: val_loss improved from 0.00052 to 0.00050, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 739us/sample - loss: 6.6390e-04 - val_loss: 4.9609e-04\n",
      "Epoch 25/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 6.4373e-04\n",
      "Epoch 00025: val_loss improved from 0.00050 to 0.00048, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 788us/sample - loss: 6.4298e-04 - val_loss: 4.7924e-04\n",
      "Epoch 26/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 6.2858e-04\n",
      "Epoch 00026: val_loss improved from 0.00048 to 0.00046, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 739us/sample - loss: 6.2507e-04 - val_loss: 4.6436e-04\n",
      "Epoch 27/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 6.1256e-04\n",
      "Epoch 00027: val_loss improved from 0.00046 to 0.00045, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 780us/sample - loss: 6.0751e-04 - val_loss: 4.5207e-04\n",
      "Epoch 28/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 5.9266e-04\n",
      "Epoch 00028: val_loss improved from 0.00045 to 0.00044, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 752us/sample - loss: 5.9181e-04 - val_loss: 4.3793e-04\n",
      "Epoch 29/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 5.7845e-04\n",
      "Epoch 00029: val_loss improved from 0.00044 to 0.00042, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 789us/sample - loss: 5.7770e-04 - val_loss: 4.2417e-04\n",
      "Epoch 30/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 5.6551e-04\n",
      "Epoch 00030: val_loss improved from 0.00042 to 0.00042, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 742us/sample - loss: 5.6241e-04 - val_loss: 4.1682e-04\n",
      "Epoch 31/200\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "912/961 [===========================>..] - ETA: 0s - loss: 5.5330e-04\n",
      "Epoch 00031: val_loss improved from 0.00042 to 0.00040, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 764us/sample - loss: 5.4998e-04 - val_loss: 4.0236e-04\n",
      "Epoch 32/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 5.3626e-04\n",
      "Epoch 00032: val_loss improved from 0.00040 to 0.00039, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 753us/sample - loss: 5.3707e-04 - val_loss: 3.8834e-04\n",
      "Epoch 33/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 5.2017e-04\n",
      "Epoch 00033: val_loss improved from 0.00039 to 0.00038, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 767us/sample - loss: 5.2382e-04 - val_loss: 3.8440e-04\n",
      "Epoch 34/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 5.1404e-04\n",
      "Epoch 00034: val_loss improved from 0.00038 to 0.00038, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 763us/sample - loss: 5.1337e-04 - val_loss: 3.7695e-04\n",
      "Epoch 35/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 5.0143e-04\n",
      "Epoch 00035: val_loss improved from 0.00038 to 0.00037, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 756us/sample - loss: 5.0315e-04 - val_loss: 3.6558e-04\n",
      "Epoch 36/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 4.9312e-04\n",
      "Epoch 00036: val_loss improved from 0.00037 to 0.00036, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 749us/sample - loss: 4.9343e-04 - val_loss: 3.5751e-04\n",
      "Epoch 37/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.8340e-04\n",
      "Epoch 00037: val_loss improved from 0.00036 to 0.00035, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 744us/sample - loss: 4.8381e-04 - val_loss: 3.4767e-04\n",
      "Epoch 38/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.7489e-04\n",
      "Epoch 00038: val_loss improved from 0.00035 to 0.00034, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 755us/sample - loss: 4.7544e-04 - val_loss: 3.4081e-04\n",
      "Epoch 39/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.6768e-04\n",
      "Epoch 00039: val_loss improved from 0.00034 to 0.00034, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 740us/sample - loss: 4.6618e-04 - val_loss: 3.3564e-04\n",
      "Epoch 40/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.5728e-04\n",
      "Epoch 00040: val_loss improved from 0.00034 to 0.00033, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 747us/sample - loss: 4.5754e-04 - val_loss: 3.2856e-04\n",
      "Epoch 41/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.4730e-04\n",
      "Epoch 00041: val_loss improved from 0.00033 to 0.00032, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 748us/sample - loss: 4.4957e-04 - val_loss: 3.2233e-04\n",
      "Epoch 42/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.4300e-04\n",
      "Epoch 00042: val_loss improved from 0.00032 to 0.00032, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 754us/sample - loss: 4.4259e-04 - val_loss: 3.2017e-04\n",
      "Epoch 43/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.3556e-04\n",
      "Epoch 00043: val_loss improved from 0.00032 to 0.00031, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 793us/sample - loss: 4.3624e-04 - val_loss: 3.1237e-04\n",
      "Epoch 44/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.2693e-04\n",
      "Epoch 00044: val_loss improved from 0.00031 to 0.00031, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 757us/sample - loss: 4.2873e-04 - val_loss: 3.1084e-04\n",
      "Epoch 45/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 4.2401e-04\n",
      "Epoch 00045: val_loss improved from 0.00031 to 0.00030, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 797us/sample - loss: 4.2385e-04 - val_loss: 3.0457e-04\n",
      "Epoch 46/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 4.1552e-04\n",
      "Epoch 00046: val_loss improved from 0.00030 to 0.00030, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 757us/sample - loss: 4.1734e-04 - val_loss: 3.0329e-04\n",
      "Epoch 47/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 4.1340e-04\n",
      "Epoch 00047: val_loss improved from 0.00030 to 0.00030, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 773us/sample - loss: 4.1342e-04 - val_loss: 2.9510e-04\n",
      "Epoch 48/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 4.0639e-04\n",
      "Epoch 00048: val_loss improved from 0.00030 to 0.00029, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 743us/sample - loss: 4.0639e-04 - val_loss: 2.9259e-04\n",
      "Epoch 49/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 4.0580e-04\n",
      "Epoch 00049: val_loss improved from 0.00029 to 0.00029, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 783us/sample - loss: 4.0114e-04 - val_loss: 2.9103e-04\n",
      "Epoch 50/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.9578e-04\n",
      "Epoch 00050: val_loss improved from 0.00029 to 0.00029, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 728us/sample - loss: 3.9627e-04 - val_loss: 2.8594e-04\n",
      "Epoch 51/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 3.9065e-04\n",
      "Epoch 00051: val_loss improved from 0.00029 to 0.00028, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 791us/sample - loss: 3.9075e-04 - val_loss: 2.8314e-04\n",
      "Epoch 52/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.8672e-04\n",
      "Epoch 00052: val_loss improved from 0.00028 to 0.00028, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 749us/sample - loss: 3.8663e-04 - val_loss: 2.8245e-04\n",
      "Epoch 53/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.8232e-04\n",
      "Epoch 00053: val_loss improved from 0.00028 to 0.00028, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 787us/sample - loss: 3.8292e-04 - val_loss: 2.7737e-04\n",
      "Epoch 54/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.7864e-04\n",
      "Epoch 00054: val_loss improved from 0.00028 to 0.00028, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 748us/sample - loss: 3.7794e-04 - val_loss: 2.7580e-04\n",
      "Epoch 55/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.7488e-04\n",
      "Epoch 00055: val_loss improved from 0.00028 to 0.00028, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 778us/sample - loss: 3.7535e-04 - val_loss: 2.7504e-04\n",
      "Epoch 56/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.7107e-04\n",
      "Epoch 00056: val_loss improved from 0.00028 to 0.00027, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 761us/sample - loss: 3.7182e-04 - val_loss: 2.7026e-04\n",
      "Epoch 57/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.7123e-04\n",
      "Epoch 00057: val_loss improved from 0.00027 to 0.00027, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 791us/sample - loss: 3.6763e-04 - val_loss: 2.6811e-04\n",
      "Epoch 58/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.6114e-04\n",
      "Epoch 00058: val_loss improved from 0.00027 to 0.00027, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 849us/sample - loss: 3.6329e-04 - val_loss: 2.6725e-04\n",
      "Epoch 59/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.6074e-04\n",
      "Epoch 00059: val_loss improved from 0.00027 to 0.00027, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 719us/sample - loss: 3.6081e-04 - val_loss: 2.6539e-04\n",
      "Epoch 60/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 3.5669e-04\n",
      "Epoch 00060: val_loss improved from 0.00027 to 0.00026, saving model to resnetbestmodel.hdf5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "961/961 [==============================] - 1s 778us/sample - loss: 3.5658e-04 - val_loss: 2.6445e-04\n",
      "Epoch 61/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.5277e-04\n",
      "Epoch 00061: val_loss improved from 0.00026 to 0.00026, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 752us/sample - loss: 3.5438e-04 - val_loss: 2.6321e-04\n",
      "Epoch 62/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.5439e-04\n",
      "Epoch 00062: val_loss improved from 0.00026 to 0.00026, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 767us/sample - loss: 3.5108e-04 - val_loss: 2.6157e-04\n",
      "Epoch 63/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.5091e-04\n",
      "Epoch 00063: val_loss improved from 0.00026 to 0.00026, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 745us/sample - loss: 3.4956e-04 - val_loss: 2.6018e-04\n",
      "Epoch 64/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.4750e-04\n",
      "Epoch 00064: val_loss improved from 0.00026 to 0.00026, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 783us/sample - loss: 3.4844e-04 - val_loss: 2.5640e-04\n",
      "Epoch 65/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.4448e-04\n",
      "Epoch 00065: val_loss improved from 0.00026 to 0.00026, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 749us/sample - loss: 3.4317e-04 - val_loss: 2.5552e-04\n",
      "Epoch 66/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.4056e-04\n",
      "Epoch 00066: val_loss improved from 0.00026 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 793us/sample - loss: 3.4155e-04 - val_loss: 2.5405e-04\n",
      "Epoch 67/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.3927e-04\n",
      "Epoch 00067: val_loss improved from 0.00025 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 739us/sample - loss: 3.3893e-04 - val_loss: 2.5397e-04\n",
      "Epoch 68/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 3.3706e-04\n",
      "Epoch 00068: val_loss improved from 0.00025 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 779us/sample - loss: 3.3697e-04 - val_loss: 2.5212e-04\n",
      "Epoch 69/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.3308e-04\n",
      "Epoch 00069: val_loss improved from 0.00025 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 744us/sample - loss: 3.3433e-04 - val_loss: 2.5112e-04\n",
      "Epoch 70/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.3316e-04\n",
      "Epoch 00070: val_loss improved from 0.00025 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 777us/sample - loss: 3.3216e-04 - val_loss: 2.4928e-04\n",
      "Epoch 71/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.2931e-04\n",
      "Epoch 00071: val_loss improved from 0.00025 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 828us/sample - loss: 3.3003e-04 - val_loss: 2.4783e-04\n",
      "Epoch 72/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.2855e-04\n",
      "Epoch 00072: val_loss did not improve from 0.00025\n",
      "961/961 [==============================] - 1s 558us/sample - loss: 3.2704e-04 - val_loss: 2.4938e-04\n",
      "Epoch 73/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.2656e-04\n",
      "Epoch 00073: val_loss improved from 0.00025 to 0.00025, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 795us/sample - loss: 3.2552e-04 - val_loss: 2.4663e-04\n",
      "Epoch 74/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.2035e-04\n",
      "Epoch 00074: val_loss improved from 0.00025 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 774us/sample - loss: 3.2353e-04 - val_loss: 2.4482e-04\n",
      "Epoch 75/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.2360e-04\n",
      "Epoch 00075: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 829us/sample - loss: 3.2311e-04 - val_loss: 2.4424e-04\n",
      "Epoch 76/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.2025e-04\n",
      "Epoch 00076: val_loss did not improve from 0.00024\n",
      "961/961 [==============================] - 1s 553us/sample - loss: 3.2035e-04 - val_loss: 2.4446e-04\n",
      "Epoch 77/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.1788e-04\n",
      "Epoch 00077: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 841us/sample - loss: 3.1821e-04 - val_loss: 2.4206e-04\n",
      "Epoch 78/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.1691e-04\n",
      "Epoch 00078: val_loss did not improve from 0.00024\n",
      "961/961 [==============================] - 1s 560us/sample - loss: 3.1633e-04 - val_loss: 2.4407e-04\n",
      "Epoch 79/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.1636e-04\n",
      "Epoch 00079: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 772us/sample - loss: 3.1627e-04 - val_loss: 2.4055e-04\n",
      "Epoch 80/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 3.1288e-04\n",
      "Epoch 00080: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 781us/sample - loss: 3.1289e-04 - val_loss: 2.3870e-04\n",
      "Epoch 81/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.1012e-04\n",
      "Epoch 00081: val_loss did not improve from 0.00024\n",
      "961/961 [==============================] - 1s 560us/sample - loss: 3.1160e-04 - val_loss: 2.4059e-04\n",
      "Epoch 82/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.0957e-04\n",
      "Epoch 00082: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 773us/sample - loss: 3.1028e-04 - val_loss: 2.3805e-04\n",
      "Epoch 83/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.0883e-04\n",
      "Epoch 00083: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 751us/sample - loss: 3.0779e-04 - val_loss: 2.3668e-04\n",
      "Epoch 84/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 3.0714e-04\n",
      "Epoch 00084: val_loss improved from 0.00024 to 0.00024, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 773us/sample - loss: 3.0696e-04 - val_loss: 2.3637e-04\n",
      "Epoch 85/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.0705e-04\n",
      "Epoch 00085: val_loss improved from 0.00024 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 752us/sample - loss: 3.0509e-04 - val_loss: 2.3500e-04\n",
      "Epoch 86/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 3.0357e-04\n",
      "Epoch 00086: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 773us/sample - loss: 3.0347e-04 - val_loss: 2.3483e-04\n",
      "Epoch 87/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.0149e-04\n",
      "Epoch 00087: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 564us/sample - loss: 3.0201e-04 - val_loss: 2.3516e-04\n",
      "Epoch 88/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 3.0131e-04\n",
      "Epoch 00088: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 787us/sample - loss: 3.0231e-04 - val_loss: 2.3465e-04\n",
      "Epoch 89/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 3.0025e-04\n",
      "Epoch 00089: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 3.0057e-04 - val_loss: 2.3619e-04\n",
      "Epoch 90/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.9884e-04\n",
      "Epoch 00090: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 2.9892e-04 - val_loss: 2.3803e-04\n",
      "Epoch 91/200\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "912/961 [===========================>..] - ETA: 0s - loss: 3.0054e-04\n",
      "Epoch 00091: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 773us/sample - loss: 2.9996e-04 - val_loss: 2.3158e-04\n",
      "Epoch 92/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.9556e-04\n",
      "Epoch 00092: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 744us/sample - loss: 2.9681e-04 - val_loss: 2.3128e-04\n",
      "Epoch 93/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.9610e-04\n",
      "Epoch 00093: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 584us/sample - loss: 2.9573e-04 - val_loss: 2.3306e-04\n",
      "Epoch 94/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.9531e-04\n",
      "Epoch 00094: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 755us/sample - loss: 2.9531e-04 - val_loss: 2.2874e-04\n",
      "Epoch 95/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.9057e-04\n",
      "Epoch 00095: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 598us/sample - loss: 2.9276e-04 - val_loss: 2.3009e-04\n",
      "Epoch 96/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.9112e-04\n",
      "Epoch 00096: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 552us/sample - loss: 2.9119e-04 - val_loss: 2.2885e-04\n",
      "Epoch 97/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.9294e-04\n",
      "Epoch 00097: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 565us/sample - loss: 2.9141e-04 - val_loss: 2.2892e-04\n",
      "Epoch 98/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.9009e-04\n",
      "Epoch 00098: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 785us/sample - loss: 2.9009e-04 - val_loss: 2.2813e-04\n",
      "Epoch 99/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8884e-04\n",
      "Epoch 00099: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 559us/sample - loss: 2.8909e-04 - val_loss: 2.3150e-04\n",
      "Epoch 100/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.9025e-04\n",
      "Epoch 00100: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 778us/sample - loss: 2.8831e-04 - val_loss: 2.2767e-04\n",
      "Epoch 101/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8742e-04\n",
      "Epoch 00101: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 559us/sample - loss: 2.8746e-04 - val_loss: 2.2995e-04\n",
      "Epoch 102/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8653e-04\n",
      "Epoch 00102: val_loss improved from 0.00023 to 0.00023, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 812us/sample - loss: 2.8624e-04 - val_loss: 2.2585e-04\n",
      "Epoch 103/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8633e-04\n",
      "Epoch 00103: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 554us/sample - loss: 2.8508e-04 - val_loss: 2.2649e-04\n",
      "Epoch 104/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8309e-04\n",
      "Epoch 00104: val_loss did not improve from 0.00023\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 2.8443e-04 - val_loss: 2.2908e-04\n",
      "Epoch 105/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8486e-04\n",
      "Epoch 00105: val_loss improved from 0.00023 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 768us/sample - loss: 2.8581e-04 - val_loss: 2.2408e-04\n",
      "Epoch 106/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.8196e-04\n",
      "Epoch 00106: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 763us/sample - loss: 2.8346e-04 - val_loss: 2.2367e-04\n",
      "Epoch 107/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.8076e-04\n",
      "Epoch 00107: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 595us/sample - loss: 2.8118e-04 - val_loss: 2.3126e-04\n",
      "Epoch 108/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.8598e-04\n",
      "Epoch 00108: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 2.8549e-04 - val_loss: 2.2422e-04\n",
      "Epoch 109/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.8059e-04\n",
      "Epoch 00109: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 793us/sample - loss: 2.8044e-04 - val_loss: 2.2137e-04\n",
      "Epoch 110/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7869e-04\n",
      "Epoch 00110: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 746us/sample - loss: 2.7779e-04 - val_loss: 2.2129e-04\n",
      "Epoch 111/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7435e-04\n",
      "Epoch 00111: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 566us/sample - loss: 2.7731e-04 - val_loss: 2.2327e-04\n",
      "Epoch 112/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.7726e-04\n",
      "Epoch 00112: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 769us/sample - loss: 2.7737e-04 - val_loss: 2.2124e-04\n",
      "Epoch 113/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7689e-04\n",
      "Epoch 00113: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 559us/sample - loss: 2.7680e-04 - val_loss: 2.2133e-04\n",
      "Epoch 114/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7676e-04\n",
      "Epoch 00114: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 793us/sample - loss: 2.7527e-04 - val_loss: 2.1954e-04\n",
      "Epoch 115/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7132e-04\n",
      "Epoch 00115: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 2.7363e-04 - val_loss: 2.2109e-04\n",
      "Epoch 116/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.7395e-04\n",
      "Epoch 00116: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 777us/sample - loss: 2.7378e-04 - val_loss: 2.1795e-04\n",
      "Epoch 117/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.6823e-04\n",
      "Epoch 00117: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 745us/sample - loss: 2.7135e-04 - val_loss: 2.1776e-04\n",
      "Epoch 118/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6968e-04\n",
      "Epoch 00118: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 584us/sample - loss: 2.7144e-04 - val_loss: 2.1817e-04\n",
      "Epoch 119/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7187e-04\n",
      "Epoch 00119: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 759us/sample - loss: 2.7066e-04 - val_loss: 2.1719e-04\n",
      "Epoch 120/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.6982e-04\n",
      "Epoch 00120: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 848us/sample - loss: 2.6905e-04 - val_loss: 2.1633e-04\n",
      "Epoch 121/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7138e-04\n",
      "Epoch 00121: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 562us/sample - loss: 2.6875e-04 - val_loss: 2.1721e-04\n",
      "Epoch 122/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6719e-04\n",
      "Epoch 00122: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 563us/sample - loss: 2.6895e-04 - val_loss: 2.1765e-04\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 123/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7244e-04\n",
      "Epoch 00123: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 602us/sample - loss: 2.7038e-04 - val_loss: 2.1875e-04\n",
      "Epoch 124/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7028e-04\n",
      "Epoch 00124: val_loss did not improve from 0.00022\n",
      "961/961 [==============================] - 1s 561us/sample - loss: 2.7070e-04 - val_loss: 2.1798e-04\n",
      "Epoch 125/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.6584e-04\n",
      "Epoch 00125: val_loss improved from 0.00022 to 0.00022, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 825us/sample - loss: 2.6722e-04 - val_loss: 2.1623e-04\n",
      "Epoch 126/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.6677e-04\n",
      "Epoch 00126: val_loss improved from 0.00022 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 761us/sample - loss: 2.6671e-04 - val_loss: 2.1410e-04\n",
      "Epoch 127/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6445e-04\n",
      "Epoch 00127: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 544us/sample - loss: 2.6479e-04 - val_loss: 2.1776e-04\n",
      "Epoch 128/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6768e-04\n",
      "Epoch 00128: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 565us/sample - loss: 2.6775e-04 - val_loss: 2.2099e-04\n",
      "Epoch 129/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.7110e-04\n",
      "Epoch 00129: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 2.7128e-04 - val_loss: 2.1696e-04\n",
      "Epoch 130/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6729e-04\n",
      "Epoch 00130: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 561us/sample - loss: 2.6592e-04 - val_loss: 2.1579e-04\n",
      "Epoch 131/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6338e-04\n",
      "Epoch 00131: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 775us/sample - loss: 2.6325e-04 - val_loss: 2.1380e-04\n",
      "Epoch 132/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6026e-04\n",
      "Epoch 00132: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 560us/sample - loss: 2.6130e-04 - val_loss: 2.1426e-04\n",
      "Epoch 133/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6035e-04\n",
      "Epoch 00133: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 556us/sample - loss: 2.6177e-04 - val_loss: 2.1466e-04\n",
      "Epoch 134/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6840e-04\n",
      "Epoch 00134: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 555us/sample - loss: 2.6799e-04 - val_loss: 2.1389e-04\n",
      "Epoch 135/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6153e-04\n",
      "Epoch 00135: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 555us/sample - loss: 2.6191e-04 - val_loss: 2.1391e-04\n",
      "Epoch 136/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.6039e-04\n",
      "Epoch 00136: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 565us/sample - loss: 2.6047e-04 - val_loss: 2.1545e-04\n",
      "Epoch 137/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6159e-04\n",
      "Epoch 00137: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 560us/sample - loss: 2.6076e-04 - val_loss: 2.1695e-04\n",
      "Epoch 138/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.6232e-04\n",
      "Epoch 00138: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 766us/sample - loss: 2.6260e-04 - val_loss: 2.1105e-04\n",
      "Epoch 139/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5766e-04\n",
      "Epoch 00139: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 568us/sample - loss: 2.5808e-04 - val_loss: 2.1145e-04\n",
      "Epoch 140/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5727e-04\n",
      "Epoch 00140: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 812us/sample - loss: 2.5811e-04 - val_loss: 2.1095e-04\n",
      "Epoch 141/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5819e-04\n",
      "Epoch 00141: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 760us/sample - loss: 2.5734e-04 - val_loss: 2.1069e-04\n",
      "Epoch 142/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.5763e-04\n",
      "Epoch 00142: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 793us/sample - loss: 2.5686e-04 - val_loss: 2.0981e-04\n",
      "Epoch 143/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.5752e-04\n",
      "Epoch 00143: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 562us/sample - loss: 2.5537e-04 - val_loss: 2.1227e-04\n",
      "Epoch 144/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5737e-04\n",
      "Epoch 00144: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 597us/sample - loss: 2.5644e-04 - val_loss: 2.1086e-04\n",
      "Epoch 145/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5402e-04\n",
      "Epoch 00145: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 771us/sample - loss: 2.5376e-04 - val_loss: 2.0867e-04\n",
      "Epoch 146/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.5303e-04\n",
      "Epoch 00146: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 790us/sample - loss: 2.5310e-04 - val_loss: 2.0835e-04\n",
      "Epoch 147/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.5353e-04\n",
      "Epoch 00147: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 565us/sample - loss: 2.5360e-04 - val_loss: 2.0902e-04\n",
      "Epoch 148/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.5493e-04\n",
      "Epoch 00148: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 597us/sample - loss: 2.5501e-04 - val_loss: 2.1237e-04\n",
      "Epoch 149/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5929e-04\n",
      "Epoch 00149: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 755us/sample - loss: 2.5770e-04 - val_loss: 2.0686e-04\n",
      "Epoch 150/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.5271e-04\n",
      "Epoch 00150: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 803us/sample - loss: 2.5270e-04 - val_loss: 2.0667e-04\n",
      "Epoch 151/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5372e-04\n",
      "Epoch 00151: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 563us/sample - loss: 2.5257e-04 - val_loss: 2.0707e-04\n",
      "Epoch 152/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5083e-04\n",
      "Epoch 00152: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 564us/sample - loss: 2.5059e-04 - val_loss: 2.0879e-04\n",
      "Epoch 153/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.5317e-04\n",
      "Epoch 00153: val_loss improved from 0.00021 to 0.00021, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 797us/sample - loss: 2.5242e-04 - val_loss: 2.0634e-04\n",
      "Epoch 154/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5121e-04\n",
      "Epoch 00154: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 557us/sample - loss: 2.5064e-04 - val_loss: 2.0722e-04\n",
      "Epoch 155/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4800e-04\n",
      "Epoch 00155: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 596us/sample - loss: 2.4795e-04 - val_loss: 2.0918e-04\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 156/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4939e-04\n",
      "Epoch 00156: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 561us/sample - loss: 2.4921e-04 - val_loss: 2.0762e-04\n",
      "Epoch 157/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.5076e-04\n",
      "Epoch 00157: val_loss did not improve from 0.00021\n",
      "961/961 [==============================] - 1s 597us/sample - loss: 2.4891e-04 - val_loss: 2.0716e-04\n",
      "Epoch 158/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.5070e-04\n",
      "Epoch 00158: val_loss improved from 0.00021 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 774us/sample - loss: 2.5040e-04 - val_loss: 2.0461e-04\n",
      "Epoch 159/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4766e-04\n",
      "Epoch 00159: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 783us/sample - loss: 2.4768e-04 - val_loss: 2.0395e-04\n",
      "Epoch 160/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4596e-04\n",
      "Epoch 00160: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 746us/sample - loss: 2.4651e-04 - val_loss: 2.0364e-04\n",
      "Epoch 161/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4538e-04\n",
      "Epoch 00161: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 574us/sample - loss: 2.4536e-04 - val_loss: 2.0379e-04\n",
      "Epoch 162/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4388e-04\n",
      "Epoch 00162: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 747us/sample - loss: 2.4460e-04 - val_loss: 2.0210e-04\n",
      "Epoch 163/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4363e-04\n",
      "Epoch 00163: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 600us/sample - loss: 2.4368e-04 - val_loss: 2.0979e-04\n",
      "Epoch 164/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4848e-04\n",
      "Epoch 00164: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 562us/sample - loss: 2.4808e-04 - val_loss: 2.0631e-04\n",
      "Epoch 165/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.4934e-04\n",
      "Epoch 00165: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 566us/sample - loss: 2.4845e-04 - val_loss: 2.0369e-04\n",
      "Epoch 166/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4554e-04\n",
      "Epoch 00166: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 592us/sample - loss: 2.4493e-04 - val_loss: 2.0362e-04\n",
      "Epoch 167/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4305e-04\n",
      "Epoch 00167: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 754us/sample - loss: 2.4270e-04 - val_loss: 2.0204e-04\n",
      "Epoch 168/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4210e-04\n",
      "Epoch 00168: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 562us/sample - loss: 2.4346e-04 - val_loss: 2.0493e-04\n",
      "Epoch 169/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4312e-04\n",
      "Epoch 00169: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 807us/sample - loss: 2.4298e-04 - val_loss: 2.0077e-04\n",
      "Epoch 170/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.4044e-04\n",
      "Epoch 00170: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 558us/sample - loss: 2.4039e-04 - val_loss: 2.0215e-04\n",
      "Epoch 171/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4380e-04\n",
      "Epoch 00171: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 601us/sample - loss: 2.4386e-04 - val_loss: 2.0112e-04\n",
      "Epoch 172/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4079e-04\n",
      "Epoch 00172: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 573us/sample - loss: 2.4083e-04 - val_loss: 2.0204e-04\n",
      "Epoch 173/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.4030e-04\n",
      "Epoch 00173: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 569us/sample - loss: 2.4133e-04 - val_loss: 2.0362e-04\n",
      "Epoch 174/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4186e-04\n",
      "Epoch 00174: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 816us/sample - loss: 2.4191e-04 - val_loss: 1.9913e-04\n",
      "Epoch 175/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3935e-04\n",
      "Epoch 00175: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 563us/sample - loss: 2.3979e-04 - val_loss: 2.0081e-04\n",
      "Epoch 176/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3795e-04\n",
      "Epoch 00176: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 798us/sample - loss: 2.3826e-04 - val_loss: 1.9831e-04\n",
      "Epoch 177/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.3921e-04\n",
      "Epoch 00177: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 562us/sample - loss: 2.3853e-04 - val_loss: 1.9901e-04\n",
      "Epoch 178/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3860e-04\n",
      "Epoch 00178: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 592us/sample - loss: 2.3869e-04 - val_loss: 1.9913e-04\n",
      "Epoch 179/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3784e-04\n",
      "Epoch 00179: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 571us/sample - loss: 2.3976e-04 - val_loss: 2.0729e-04\n",
      "Epoch 180/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.4151e-04\n",
      "Epoch 00180: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 835us/sample - loss: 2.4144e-04 - val_loss: 1.9787e-04\n",
      "Epoch 181/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3567e-04\n",
      "Epoch 00181: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 753us/sample - loss: 2.3591e-04 - val_loss: 1.9761e-04\n",
      "Epoch 182/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3614e-04\n",
      "Epoch 00182: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 814us/sample - loss: 2.3613e-04 - val_loss: 1.9711e-04\n",
      "Epoch 183/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3470e-04\n",
      "Epoch 00183: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 819us/sample - loss: 2.3562e-04 - val_loss: 1.9668e-04\n",
      "Epoch 184/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3627e-04\n",
      "Epoch 00184: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 627us/sample - loss: 2.3645e-04 - val_loss: 1.9766e-04\n",
      "Epoch 185/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3655e-04\n",
      "Epoch 00185: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 599us/sample - loss: 2.3653e-04 - val_loss: 1.9715e-04\n",
      "Epoch 186/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3450e-04\n",
      "Epoch 00186: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 594us/sample - loss: 2.3494e-04 - val_loss: 2.1964e-04\n",
      "Epoch 187/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.4591e-04\n",
      "Epoch 00187: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 563us/sample - loss: 2.4628e-04 - val_loss: 1.9675e-04\n",
      "Epoch 188/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.3714e-04\n",
      "Epoch 00188: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "961/961 [==============================] - 1s 766us/sample - loss: 2.3667e-04 - val_loss: 1.9622e-04\n",
      "Epoch 189/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3471e-04\n",
      "Epoch 00189: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 804us/sample - loss: 2.3399e-04 - val_loss: 1.9590e-04\n",
      "Epoch 190/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3300e-04\n",
      "Epoch 00190: val_loss improved from 0.00020 to 0.00020, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 778us/sample - loss: 2.3293e-04 - val_loss: 1.9530e-04\n",
      "Epoch 191/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.3225e-04\n",
      "Epoch 00191: val_loss did not improve from 0.00020\n",
      "961/961 [==============================] - 1s 597us/sample - loss: 2.3210e-04 - val_loss: 1.9552e-04\n",
      "Epoch 192/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3301e-04\n",
      "Epoch 00192: val_loss improved from 0.00020 to 0.00019, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 766us/sample - loss: 2.3219e-04 - val_loss: 1.9372e-04\n",
      "Epoch 193/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3150e-04\n",
      "Epoch 00193: val_loss did not improve from 0.00019\n",
      "961/961 [==============================] - 1s 590us/sample - loss: 2.3086e-04 - val_loss: 1.9504e-04\n",
      "Epoch 194/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3377e-04\n",
      "Epoch 00194: val_loss did not improve from 0.00019\n",
      "961/961 [==============================] - 1s 561us/sample - loss: 2.3293e-04 - val_loss: 1.9685e-04\n",
      "Epoch 195/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3444e-04\n",
      "Epoch 00195: val_loss improved from 0.00019 to 0.00019, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 859us/sample - loss: 2.3454e-04 - val_loss: 1.9366e-04\n",
      "Epoch 196/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.3077e-04\n",
      "Epoch 00196: val_loss did not improve from 0.00019\n",
      "961/961 [==============================] - 1s 558us/sample - loss: 2.3037e-04 - val_loss: 1.9372e-04\n",
      "Epoch 197/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.2871e-04\n",
      "Epoch 00197: val_loss improved from 0.00019 to 0.00019, saving model to resnetbestmodel.hdf5\n",
      "961/961 [==============================] - 1s 844us/sample - loss: 2.2879e-04 - val_loss: 1.9279e-04\n",
      "Epoch 198/200\n",
      "960/961 [============================>.] - ETA: 0s - loss: 2.2888e-04\n",
      "Epoch 00198: val_loss did not improve from 0.00019\n",
      "961/961 [==============================] - 1s 571us/sample - loss: 2.2892e-04 - val_loss: 1.9395e-04\n",
      "Epoch 199/200\n",
      "864/961 [=========================>....] - ETA: 0s - loss: 2.3114e-04\n",
      "Epoch 00199: val_loss did not improve from 0.00019\n",
      "961/961 [==============================] - 1s 563us/sample - loss: 2.2924e-04 - val_loss: 1.9293e-04\n",
      "Epoch 200/200\n",
      "912/961 [===========================>..] - ETA: 0s - loss: 2.2964e-04\n",
      "Epoch 00200: val_loss did not improve from 0.00019\n",
      "961/961 [==============================] - 1s 588us/sample - loss: 2.2860e-04 - val_loss: 1.9280e-04\n"
     ]
    }
   ],
   "source": [
    "h=mymodel.fit(x=Y,y=X,batch_size=batch_size,epochs=num_epochs,validation_data=(Yt,Xt),callbacks=callbacks_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig1=plt.figure()\n",
    "plt.plot(range(num_epochs),h.history[\"loss\"],label=\"train loss\")\n",
    "plt.plot(range(num_epochs),h.history[\"val_loss\"],label=\"val loss\")\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "# fig1.savefig(\"skipnetloss_rek.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "bestmodel=tf.keras.models.load_model(modelpath)\n",
    "Ypr=bestmodel.predict(y.reshape(-1,y.shape[0],y.shape[1],y.shape[2]))\n",
    "Ypr=Ypr*scale"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Each band [24.0342051  27.27951604 27.57427314 28.39492495 26.4581451  22.21935589]\n",
      "Mean SRE:  25.993403369688085\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "msre,sre=sreCal(x*scale,Ypr[0])\n",
    "# bmsre,bsre=sreCal(x,xD)\n",
    "print(\"Each band\",sre)\n",
    "print(\"Mean SRE: \",msre)\n",
    "# print(\"Each band bicubic\",bsre)\n",
    "# print(\"Mean SRE bicubic: \",bmsre)\n",
    "plt.figure(figsize=(15,15))\n",
    "band=0\n",
    "plt.subplot(131)\n",
    "plt.imshow(Ypr[0][:,:,1],cmap=\"gray\")\n",
    "plt.title(\"Predict band 5\")\n",
    "plt.subplot(132)\n",
    "plt.imshow(x[:,:,1],cmap=\"gray\")\n",
    "plt.title(\"True band 5\")\n",
    "# plt.subplot(133)\n",
    "# plt.imshow(xD[:,:,showband],cmap=\"gray\")\n",
    "# plt.title(\"Bicubic band 5\")\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": "3.7.6"
  },
  "pycharm": {
   "stem_cell": {
    "cell_type": "raw",
    "metadata": {
     "collapsed": false
    },
    "source": []
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}