fashion-mnist-katib.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[I 200222 04:06:30 config:123] Using preprocessor: <kubeflow.fairing.preprocessors.converted_notebook.ConvertNotebookPreprocessor object at 0x7fe087472898>\n",
      "[I 200222 04:06:30 config:125] Using builder: <kubeflow.fairing.builders.append.append.AppendBuilder object at 0x7fe035db02e8>\n",
      "[I 200222 04:06:30 config:127] Using deployer: <kubeflow.fairing.builders.append.append.AppendBuilder object at 0x7fe035db02e8>\n",
      "[W 200222 04:06:30 append:50] Building image using Append builder...\n",
      "[I 200222 04:06:30 base:105] Creating docker context: /tmp/fairing_context_yah1yor_\n",
      "[I 200222 04:06:30 converted_notebook:127] Converting fashion-mnist-katib.ipynb to fashion-mnist-katib.py\n",
      "[I 200222 04:06:30 docker_creds_:234] Loading Docker credentials for repository 'brightfly/kubeflow-jupyter-lab:tf2.0-cpu'\n",
      "[W 200222 04:06:32 append:54] Image successfully built in 2.3115801259991713s.\n",
      "[W 200222 04:06:32 append:94] Pushing image kubeflow-registry.default.svc.cluster.local:30000/fairing-job:506FE0D9...\n",
      "[I 200222 04:06:32 docker_creds_:234] Loading Docker credentials for repository 'kubeflow-registry.default.svc.cluster.local:30000/fairing-job:506FE0D9'\n",
      "[W 200222 04:06:32 append:81] Uploading kubeflow-registry.default.svc.cluster.local:30000/fairing-job:506FE0D9\n",
      "[I 200222 04:06:33 docker_session_:284] Layer sha256:d83811f270d56d34a208f721f3dbf1b9242d1900ad8981fc7071339681998a31 pushed.\n",
      "[I 200222 04:06:33 docker_session_:284] Layer sha256:66257906239d377cd700c566b27f12895ccefcd8d95eae7377f600208151d8e0 pushed.\n",
      "[I 200222 04:06:33 docker_session_:284] Layer sha256:7acbfba21e38804c20c4da292d2f99d88d317b83440fb2f3de0a620edf2b9595 pushed.\n",
      "[I 200222 04:06:33 docker_session_:284] Layer sha256:adaed07043891659935eb0bf7ce4bfa93d631223548781e6bf5f7ec286cddaa3 pushed.\n",
      "[I 200222 04:06:34 docker_session_:284] Layer sha256:7fc152dfb3a6b5c9a436b49ff6cd72ed7eb5f1fd349128b50ee04c3c5c2355fb pushed.\n",
      "[I 200222 04:06:34 docker_session_:284] Layer sha256:25d0c0b5c0e2b38808b9702d211eb0dbfecfab54e44020b7e20c2f5c727feda0 pushed.\n",
      "[I 200222 04:06:35 docker_session_:284] Layer sha256:af228bb15b867acbee0d6df28dbb42a8218948fdde64833bd021a255a99a2690 pushed.\n",
      "[I 200222 04:06:36 docker_session_:284] Layer sha256:fb9676c4102241f220b864e98c53676e27c407c6ecfbb13c2cbc1fa4de8f1811 pushed.\n",
      "[I 200222 04:06:36 docker_session_:284] Layer sha256:ee671aafb583e2321880e275c94d49a49185006730e871435cd851f42d2a775d pushed.\n",
      "[I 200222 04:06:37 docker_session_:284] Layer sha256:1c60aaf19cc4c142934f390281975f61956475c205b5895711f839a91f927779 pushed.\n",
      "[I 200222 04:06:37 docker_session_:284] Layer sha256:d846ead9aa38bc0d35dd500ada5349fe42efb96c89db08b8abcb9ad949332c2a pushed.\n",
      "[I 200222 04:06:39 docker_session_:284] Layer sha256:d0d480fe0d0ab233a90471379d9d3789d4153ad4611d101cfefdfc0c507d9356 pushed.\n",
      "[I 200222 04:06:47 docker_session_:284] Layer sha256:50f1fe304a9431ea025aa6ee5e8e45826fa62640a83e55be8bd3c227119cbcd2 pushed.\n",
      "[I 200222 04:06:47 docker_session_:284] Layer sha256:df80607e3e38dc3f117e967a36620c42f791afc9ca0568900d13d84c1e3a38bb pushed.\n",
      "[I 200222 04:06:48 docker_session_:284] Layer sha256:6ac505cbabeccba798e608e1bb80ae66f497c2c856733800e1db060e3c875917 pushed.\n",
      "[I 200222 04:06:49 docker_session_:284] Layer sha256:5667fdb72017d1fb364744ca1abf7b6f3bbe9c98c3786f294a461c2866db69ab pushed.\n",
      "[I 200222 04:06:54 docker_session_:284] Layer sha256:fd781ca227d0199735cceb67afa81e2a073b5ef5867ca6e96ad46e002a03252a pushed.\n",
      "[I 200222 04:07:09 docker_session_:284] Layer sha256:485ea80d093de507a5ccd36effaf8dad03b67a1063edbd427fedfe667c9155bc pushed.\n",
      "[I 200222 04:07:10 docker_session_:284] Layer sha256:872efc07c3d848e0f55b7cbce9866a8dea7b76d8c1f131801934ea0d856b15c5 pushed.\n",
      "[I 200222 04:07:31 docker_session_:284] Layer sha256:069de100aaecb1299b09dab29a4eb39a2ca6a5f8c0a033b65ff0e15e9ea39a24 pushed.\n",
      "[I 200222 04:07:34 docker_session_:284] Layer sha256:8f64ecfab6c102a62dd5d912d96113603663ad8b436701bf2dbd1aa9630882e7 pushed.\n",
      "[I 200222 04:07:42 docker_session_:284] Layer sha256:8e7d3ffcd500d3ed0b2ee2449a36a44601b8e282cc6d541ab93f1189d4a2cf93 pushed.\n",
      "[I 200222 04:07:54 docker_session_:284] Layer sha256:c2f81144f815902f9c6e6f7883068544e35924f0efe2275bc41c3d0c558f394c pushed.\n",
      "[I 200222 04:07:54 docker_session_:334] Finished upload of: kubeflow-registry.default.svc.cluster.local:30000/fairing-job:506FE0D9\n",
      "[W 200222 04:07:54 append:99] Pushed image kubeflow-registry.default.svc.cluster.local:30000/fairing-job:506FE0D9 in 81.79007339099189s.\n",
      "[W 200222 04:07:54 job:90] The job fairing-job-jwnl4 launched.\n",
      "[W 200222 04:07:54 manager:227] Waiting for fairing-job-jwnl4-kkmn9 to start...\n",
      "[W 200222 04:07:54 manager:227] Waiting for fairing-job-jwnl4-kkmn9 to start...\n",
      "[W 200222 04:07:54 manager:227] Waiting for fairing-job-jwnl4-kkmn9 to start...\n",
      "[W 200222 04:07:56 manager:227] Waiting for fairing-job-jwnl4-kkmn9 to start...\n",
      "[I 200222 04:07:58 manager:233] Pod started running True\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz\n",
      "32768/29515 [=================================] - 0s 1us/step\n",
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz\n",
      " 2785280/26421880 [==>...........................] - ETA: 70\n",
      " 6651904/26421880 [======>.......................] - ETA: \n",
      "11059200/26421880 [===========>..................] - ETA: \n",
      "14958592/26421880 [===============>..............] - ETA: 1\n",
      "19513344/26421880 [====================>.........] - ETA: \n",
      "23322624/26421880 [=========================>....] - ETA: \n",
      "26427392/26421880 [==============================] - 3s 0us/step\n",
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz\n",
      "8192/5148 [===============================================] - 0s 0us/step\n",
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz\n",
      "2842624/4422102 [==================>...........] - ETA: 0\n",
      "4423680/4422102 [==============================] - 1s 0us/step\n",
      "2020-02-22 04:08:04.219654: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
      "2020-02-22 04:08:04.226354: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3593190000 Hz\n",
      "2020-02-22 04:08:04.226973: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x4fddb10 executing computations on platform Host. Devices:\n",
      "2020-02-22 04:08:04.227013: I tensorflow/compiler/xla/service/service.cc:175]   StreamExecutor device (0): Host, Default Version\n",
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #\n",
      "=================================================================\n",
      "flatten (Flatten)            (None, 784)               0\n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 128)               100480\n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 128)               0\n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                1290\n",
      "=================================================================\n",
      "Total params: 101,770\n",
      "Trainable params: 101,770\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "epoch 0:\n",
      "batch=0 accuracy=0.125 loss=2.5681076\n",
      "batch=1 accuracy=0.1875 loss=2.1206436\n",
      "batch=2 accuracy=0.21875 loss=2.3287926\n",
      "batch=3 accuracy=0.1875 loss=2.4069507\n",
      "batch=4 accuracy=0.20625 loss=2.0534747\n",
      "batch=5 accuracy=0.20833333 loss=2.1179285\n",
      "batch=6 accuracy=0.23214285 loss=1.9193498\n",
      "batch=7 accuracy=0.2265625 loss=2.0583556\n",
      "batch=8 accuracy=0.24305555 loss=1.9798673\n",
      "batch=9 accuracy=0.2625 loss=1.6924738\n",
      "batch=10 accuracy=0.27272728 loss=1.7517884\n",
      "batch=11 accuracy=0.29427084 loss=1.6274675\n",
      "batch=12 accuracy=0.3125 loss=1.5028646\n",
      "batch=13 accuracy=0.34375 loss=1.2844216\n",
      "batch=14 accuracy=0.36458334 loss=1.2661712\n",
      "batch=15 accuracy=0.3828125 loss=1.2886682\n",
      "batch=16 accuracy=0.39338234 loss=1.2432353\n",
      "batch=17 accuracy=0.39583334 loss=1.4442044\n",
      "batch=18 accuracy=0.40131578 loss=1.2461569\n",
      "batch=19 accuracy=0.415625 loss=1.2020655\n",
      "batch=20 accuracy=0.42410713 loss=1.1726625\n",
      "batch=21 accuracy=0.4275568 loss=1.2376397\n",
      "batch=22 accuracy=0.44293478 loss=0.9086225\n",
      "batch=23 accuracy=0.44921875 loss=0.99452937\n",
      "batch=24 accuracy=0.46 loss=0.9242365\n",
      "batch=25 accuracy=0.46754807 loss=1.1532617\n",
      "batch=26 accuracy=0.4664352 loss=1.2440512\n",
      "batch=27 accuracy=0.46986607 loss=1.3048594\n",
      "batch=28 accuracy=0.47737068 loss=0.9412961\n",
      "batch=29 accuracy=0.47916666 loss=0.9768911\n",
      "batch=30 accuracy=0.48387095 loss=1.0549443\n",
      "batch=31 accuracy=0.4892578 loss=0.87268496\n",
      "batch=32 accuracy=0.49810606 loss=0.84744465\n",
      "batch=33 accuracy=0.5018382 loss=0.8850215\n",
      "batch=34 accuracy=0.51160717 loss=0.7609725\n",
      "batch=35 accuracy=0.5173611 loss=0.86959517\n",
      "batch=36 accuracy=0.5236486 loss=0.8299762\n",
      "batch=37 accuracy=0.5263158 loss=1.0902741\n",
      "batch=38 accuracy=0.53044873 loss=0.9614005\n",
      "batch=39 accuracy=0.5359375 loss=0.716655\n",
      "batch=40 accuracy=0.53963417 loss=0.9407904\n",
      "batch=41 accuracy=0.5431548 loss=0.81419045\n",
      "batch=42 accuracy=0.54651165 loss=1.0200577\n",
      "batch=43 accuracy=0.54971594 loss=0.8794763\n",
      "batch=44 accuracy=0.55277777 loss=0.9087976\n",
      "batch=45 accuracy=0.5557065 loss=0.7544688\n",
      "batch=46 accuracy=0.5578458 loss=0.7981615\n",
      "batch=47 accuracy=0.5605469 loss=0.9656665\n",
      "batch=48 accuracy=0.56377554 loss=0.73368454\n",
      "batch=49 accuracy=0.5675 loss=0.7997851\n",
      "batch=50 accuracy=0.5729167 loss=0.629359\n",
      "batch=51 accuracy=0.57632214 loss=0.887509\n",
      "batch=52 accuracy=0.5795991 loss=0.795642\n",
      "batch=53 accuracy=0.5821759 loss=0.77451414\n",
      "batch=54 accuracy=0.58579546 loss=0.54670763\n",
      "batch=55 accuracy=0.5859375 loss=0.96665263\n",
      "batch=56 accuracy=0.5855263 loss=1.2347646\n",
      "batch=57 accuracy=0.5872845 loss=1.106189\n",
      "batch=58 accuracy=0.5895127 loss=0.7430196\n",
      "batch=59 accuracy=0.59166664 loss=0.88049203\n",
      "batch=60 accuracy=0.5927254 loss=0.92079324\n",
      "batch=61 accuracy=0.59375 loss=1.3101388\n",
      "batch=62 accuracy=0.5952381 loss=0.9651842\n",
      "batch=63 accuracy=0.5961914 loss=0.7396469\n",
      "batch=64 accuracy=0.59759617 loss=0.8686681\n",
      "batch=65 accuracy=0.6003788 loss=0.70830333\n",
      "batch=66 accuracy=0.60307837 loss=0.6494886\n",
      "batch=67 accuracy=0.60294116 loss=0.98812205\n",
      "batch=68 accuracy=0.60461956 loss=0.80506384\n",
      "batch=69 accuracy=0.6075893 loss=0.7835567\n",
      "batch=70 accuracy=0.6078345 loss=1.0448959\n",
      "batch=71 accuracy=0.6111111 loss=0.8302816\n",
      "batch=72 accuracy=0.6121575 loss=1.0015035\n",
      "batch=73 accuracy=0.6140203 loss=0.69450045\n",
      "batch=74 accuracy=0.6141667 loss=0.90414804\n",
      "batch=75 accuracy=0.61513156 loss=0.8442034\n",
      "batch=76 accuracy=0.6168831 loss=0.66335773\n",
      "batch=77 accuracy=0.61778843 loss=0.77038044\n",
      "batch=78 accuracy=0.6190665 loss=0.72104216\n",
      "batch=79 accuracy=0.6214844 loss=0.47949725\n",
      "batch=80 accuracy=0.6246142 loss=0.49958122\n",
      "batch=81 accuracy=0.6253811 loss=0.8908566\n",
      "batch=82 accuracy=0.6261295 loss=0.80352634\n",
      "batch=83 accuracy=0.62723213 loss=0.64967\n",
      "batch=84 accuracy=0.62757355 loss=0.7987986\n",
      "batch=85 accuracy=0.62681687 loss=1.2920783\n",
      "batch=86 accuracy=0.62679595 loss=0.8644213\n",
      "batch=87 accuracy=0.628196 loss=0.8473968\n",
      "batch=88 accuracy=0.62851125 loss=0.84895647\n",
      "batch=89 accuracy=0.62951386 loss=0.8563082\n",
      "batch=90 accuracy=0.63255495 loss=0.6112212\n",
      "batch=91 accuracy=0.6314538 loss=1.076471\n",
      "batch=92 accuracy=0.6317204 loss=0.91470146\n",
      "batch=93 accuracy=0.6313165 loss=0.952371\n",
      "batch=94 accuracy=0.6325658 loss=0.6121182\n",
      "batch=95 accuracy=0.63411456 loss=0.7870654\n",
      "batch=96 accuracy=0.63563144 loss=0.6428796\n",
      "batch=97 accuracy=0.63616073 loss=0.95265377\n",
      "batch=98 accuracy=0.6366793 loss=0.91701514\n",
      "batch=99 accuracy=0.6371875 loss=0.9893533\n",
      "batch=100 accuracy=0.63675743 loss=1.0281937\n",
      "batch=101 accuracy=0.6372549 loss=0.8135197\n",
      "batch=102 accuracy=0.63834953 loss=0.70500433\n",
      "batch=103 accuracy=0.63882214 loss=0.9622747\n",
      "batch=104 accuracy=0.63988096 loss=0.88532454\n",
      "batch=105 accuracy=0.6400354 loss=1.0671117\n",
      "batch=106 accuracy=0.6401869 loss=0.78653467\n",
      "batch=107 accuracy=0.640625 loss=0.80191094\n",
      "batch=108 accuracy=0.64162844 loss=0.6419003\n",
      "batch=109 accuracy=0.64204544 loss=0.63981056\n",
      "batch=110 accuracy=0.6421734 loss=0.78898036\n",
      "batch=111 accuracy=0.64313614 loss=0.8314222\n",
      "batch=112 accuracy=0.6449115 loss=0.665616\n",
      "batch=113 accuracy=0.64638156 loss=0.61445725\n",
      "batch=114 accuracy=0.6467391 loss=0.78056633\n",
      "batch=115 accuracy=0.6465517 loss=0.83120376\n",
      "batch=116 accuracy=0.647703 loss=0.8234837\n",
      "batch=117 accuracy=0.6490996 loss=0.54256964\n",
      "batch=118 accuracy=0.64889705 loss=0.9056465\n",
      "batch=119 accuracy=0.6481771 loss=1.0520048\n",
      "batch=120 accuracy=0.64927685 loss=0.6611284\n",
      "batch=121 accuracy=0.6503586 loss=0.5281464\n",
      "batch=122 accuracy=0.6511687 loss=0.6503133\n",
      "batch=123 accuracy=0.65120965 loss=0.91975754\n",
      "batch=124 accuracy=0.65125 loss=0.78918654\n",
      "batch=125 accuracy=0.6512897 loss=1.1448369\n",
      "batch=126 accuracy=0.65206695 loss=0.77634597\n",
      "batch=127 accuracy=0.65356445 loss=0.513331\n",
      "batch=128 accuracy=0.65455425 loss=0.44909775\n",
      "batch=129 accuracy=0.6545673 loss=0.98005736\n",
      "batch=130 accuracy=0.6545802 loss=0.79908544\n",
      "batch=131 accuracy=0.65625 loss=0.47845656\n",
      "batch=132 accuracy=0.65789473 loss=0.56253433\n",
      "batch=133 accuracy=0.6588153 loss=0.6826075\n",
      "batch=134 accuracy=0.65810186 loss=0.94532675\n",
      "batch=135 accuracy=0.6587776 loss=0.7535622\n",
      "batch=136 accuracy=0.66012776 loss=0.66243887\n",
      "batch=137 accuracy=0.6614583 loss=0.5606285\n",
      "batch=138 accuracy=0.6620953 loss=0.7442091\n",
      "batch=139 accuracy=0.66272324 loss=0.6485254\n",
      "batch=140 accuracy=0.6626773 loss=0.77991533\n",
      "batch=141 accuracy=0.6621919 loss=0.8110698\n",
      "batch=142 accuracy=0.6623689 loss=0.79073495\n",
      "batch=143 accuracy=0.6623264 loss=0.8089435\n",
      "batch=144 accuracy=0.6627155 loss=0.91912884\n",
      "batch=145 accuracy=0.6637414 loss=0.63196856\n",
      "batch=146 accuracy=0.6651786 loss=0.47639716\n",
      "batch=147 accuracy=0.6651182 loss=0.7641686\n",
      "batch=148 accuracy=0.6652685 loss=0.77291846\n",
      "batch=149 accuracy=0.66583335 loss=0.7256652\n",
      "batch=150 accuracy=0.6663907 loss=0.73115075\n",
      "batch=151 accuracy=0.66735196 loss=0.5379677\n",
      "batch=152 accuracy=0.66789216 loss=0.8160052\n",
      "batch=153 accuracy=0.6686283 loss=0.5623418\n",
      "batch=154 accuracy=0.6697581 loss=0.64900637\n",
      "batch=155 accuracy=0.6708734 loss=0.5259553\n",
      "batch=156 accuracy=0.6709793 loss=0.8478115\n",
      "batch=157 accuracy=0.6722706 loss=0.46505237\n",
      "batch=158 accuracy=0.6733491 loss=0.43913203\n",
      "batch=159 accuracy=0.6732422 loss=1.0042561\n",
      "batch=160 accuracy=0.6733307 loss=0.78545773\n",
      "batch=161 accuracy=0.6736111 loss=0.6215239\n",
      "batch=162 accuracy=0.6733129 loss=1.0757977\n",
      "batch=163 accuracy=0.67358994 loss=0.78811723\n",
      "batch=164 accuracy=0.6748106 loss=0.44191897\n",
      "batch=165 accuracy=0.67469877 loss=0.9507098\n",
      "batch=166 accuracy=0.67552394 loss=0.5639377\n",
      "batch=167 accuracy=0.67689735 loss=0.539762\n",
      "batch=168 accuracy=0.67788464 loss=0.63635504\n",
      "batch=169 accuracy=0.678125 loss=0.7997353\n",
      "batch=170 accuracy=0.6785453 loss=0.7328057\n",
      "batch=171 accuracy=0.67896074 loss=0.54994416\n",
      "batch=172 accuracy=0.679552 loss=0.81098723\n",
      "batch=173 accuracy=0.6794181 loss=0.9390428\n",
      "batch=174 accuracy=0.67964286 loss=0.7709953\n",
      "batch=175 accuracy=0.67986506 loss=0.8430088\n",
      "batch=176 accuracy=0.68008476 loss=0.74897027\n",
      "batch=177 accuracy=0.6806531 loss=0.62500393\n",
      "batch=178 accuracy=0.6805168 loss=1.075212\n",
      "batch=179 accuracy=0.68072915 loss=0.6493772\n",
      "batch=180 accuracy=0.6811119 loss=0.66430175\n",
      "batch=181 accuracy=0.68149036 loss=0.48221046\n",
      "batch=182 accuracy=0.6820355 loss=0.6168963\n",
      "batch=183 accuracy=0.68342394 loss=0.28852233\n",
      "batch=184 accuracy=0.6841216 loss=0.54945624\n",
      "batch=185 accuracy=0.6846438 loss=0.5748917\n",
      "batch=186 accuracy=0.684492 loss=0.7218969\n",
      "batch=187 accuracy=0.6846742 loss=0.7868124\n",
      "batch=188 accuracy=0.68584657 loss=0.46735984\n",
      "batch=189 accuracy=0.6861842 loss=0.629288\n",
      "batch=190 accuracy=0.68684554 loss=0.7001759\n",
      "batch=191 accuracy=0.6871745 loss=0.60597193\n",
      "batch=192 accuracy=0.6879858 loss=0.53237736\n",
      "batch=193 accuracy=0.68830544 loss=0.50992477\n",
      "batch=194 accuracy=0.68830127 loss=0.77762926\n",
      "batch=195 accuracy=0.68813777 loss=0.84239644\n",
      "batch=196 accuracy=0.68845177 loss=0.5866749\n",
      "batch=197 accuracy=0.6895518 loss=0.4812005\n",
      "batch=198 accuracy=0.6893844 loss=0.7372949\n",
      "batch=199 accuracy=0.6896875 loss=0.7542126\n",
      "batch=200 accuracy=0.68998754 loss=0.57466185\n",
      "batch=201 accuracy=0.69012994 loss=0.6545949\n",
      "batch=202 accuracy=0.6908867 loss=0.57548654\n",
      "batch=203 accuracy=0.6919424 loss=0.5863943\n",
      "batch=204 accuracy=0.69207317 loss=0.7240062\n",
      "batch=205 accuracy=0.69280946 loss=0.43329278\n",
      "batch=206 accuracy=0.69278383 loss=0.70027506\n",
      "batch=207 accuracy=0.69290864 loss=0.80334127\n",
      "batch=208 accuracy=0.6936304 loss=0.5442042\n",
      "batch=209 accuracy=0.6941964 loss=0.6035783\n",
      "batch=210 accuracy=0.6941647 loss=0.7391397\n",
      "batch=211 accuracy=0.6945755 loss=0.666914\n",
      "batch=212 accuracy=0.69483566 loss=0.8720749\n",
      "batch=213 accuracy=0.6958236 loss=0.33366787\n",
      "batch=214 accuracy=0.69593024 loss=0.5825633\n",
      "batch=215 accuracy=0.69690394 loss=0.40935385\n",
      "batch=216 accuracy=0.69743663 loss=0.68728137\n",
      "batch=217 accuracy=0.69767773 loss=0.7254218\n",
      "batch=218 accuracy=0.6980594 loss=0.55064625\n",
      "batch=219 accuracy=0.69815344 loss=0.67634165\n",
      "batch=220 accuracy=0.698388 loss=0.7860924\n",
      "batch=221 accuracy=0.6984797 loss=0.66316277\n",
      "batch=222 accuracy=0.69885087 loss=0.5596706\n",
      "batch=223 accuracy=0.69977677 loss=0.4001946\n",
      "batch=224 accuracy=0.6998611 loss=0.81491673\n",
      "batch=225 accuracy=0.70022124 loss=0.6251459\n",
      "batch=226 accuracy=0.7005782 loss=0.69026613\n",
      "batch=227 accuracy=0.70106906 loss=0.4802792\n",
      "batch=228 accuracy=0.7018286 loss=0.5150988\n",
      "batch=229 accuracy=0.7024456 loss=0.44563484\n",
      "batch=230 accuracy=0.7027868 loss=0.59083176\n",
      "batch=231 accuracy=0.7028556 loss=0.6324985\n",
      "batch=232 accuracy=0.70292383 loss=1.0255826\n",
      "batch=233 accuracy=0.7032586 loss=0.4780367\n",
      "batch=234 accuracy=0.7039894 loss=0.44817907\n",
      "batch=235 accuracy=0.70458156 loss=0.46581805\n",
      "batch=236 accuracy=0.7051688 loss=0.546839\n",
      "batch=237 accuracy=0.70548844 loss=0.66145647\n",
      "batch=238 accuracy=0.70554394 loss=0.73312426\n",
      "batch=239 accuracy=0.70585936 loss=0.6844007\n",
      "batch=240 accuracy=0.7065612 loss=0.42137563\n",
      "batch=241 accuracy=0.70699894 loss=0.5138163\n",
      "batch=242 accuracy=0.7069187 loss=0.7915798\n",
      "batch=243 accuracy=0.70722336 loss=0.53614587\n",
      "batch=244 accuracy=0.70790815 loss=0.40265125\n",
      "batch=245 accuracy=0.7080793 loss=0.63802433\n",
      "batch=246 accuracy=0.7076164 loss=1.0429175\n",
      "batch=247 accuracy=0.70753527 loss=0.69970435\n",
      "batch=248 accuracy=0.7078313 loss=0.67482555\n",
      "batch=249 accuracy=0.7085 loss=0.34371582\n",
      "batch=250 accuracy=0.70891434 loss=0.6208298\n",
      "batch=251 accuracy=0.7089534 loss=0.705752\n",
      "batch=252 accuracy=0.7097332 loss=0.39837718\n",
      "batch=253 accuracy=0.7097687 loss=0.58240473\n",
      "batch=254 accuracy=0.7101716 loss=0.6219133\n",
      "batch=255 accuracy=0.71032715 loss=0.6177678\n",
      "batch=256 accuracy=0.7106031 loss=0.68481326\n",
      "batch=257 accuracy=0.71087694 loss=0.7061736\n",
      "batch=258 accuracy=0.7107867 loss=0.769949\n",
      "batch=259 accuracy=0.7109375 loss=0.710726\n",
      "batch=260 accuracy=0.7113266 loss=0.77757853\n",
      "batch=261 accuracy=0.7115935 loss=0.4276979\n",
      "batch=262 accuracy=0.71162075 loss=0.62224174\n",
      "batch=263 accuracy=0.7112926 loss=1.139437\n",
      "batch=264 accuracy=0.7110849 loss=0.7267959\n",
      "batch=265 accuracy=0.71123123 loss=0.8552765\n",
      "batch=266 accuracy=0.7110253 loss=0.909599\n",
      "batch=267 accuracy=0.7112873 loss=0.50300723\n",
      "batch=268 accuracy=0.7115474 loss=0.7557207\n",
      "batch=269 accuracy=0.712037 loss=0.39847425\n",
      "batch=270 accuracy=0.7127537 loss=0.56280005\n",
      "batch=271 accuracy=0.7132353 loss=0.441159\n",
      "batch=272 accuracy=0.713141 loss=0.80677646\n",
      "batch=273 accuracy=0.71316147 loss=0.6733967\n",
      "batch=274 accuracy=0.7130682 loss=0.8102349\n",
      "batch=275 accuracy=0.71342844 loss=0.48929265\n",
      "batch=276 accuracy=0.7142374 loss=0.28191352\n",
      "batch=277 accuracy=0.71459085 loss=0.71215546\n",
      "batch=278 accuracy=0.71460575 loss=0.6091076\n",
      "batch=279 accuracy=0.71473217 loss=0.70840925\n",
      "batch=280 accuracy=0.71463525 loss=0.8554923\n",
      "batch=281 accuracy=0.7143174 loss=0.8553549\n",
      "batch=282 accuracy=0.7145539 loss=0.5564417\n",
      "batch=283 accuracy=0.7150088 loss=0.4195895\n",
      "batch=284 accuracy=0.7150219 loss=0.67830306\n",
      "batch=285 accuracy=0.7152535 loss=0.71317923\n",
      "batch=286 accuracy=0.7151568 loss=0.84475815\n",
      "batch=287 accuracy=0.7156033 loss=0.4873712\n",
      "batch=288 accuracy=0.7156142 loss=0.63612145\n",
      "batch=289 accuracy=0.7161638 loss=0.49394047\n",
      "batch=290 accuracy=0.71638745 loss=0.6154473\n",
      "batch=291 accuracy=0.7169306 loss=0.43662316\n",
      "batch=292 accuracy=0.71768343 loss=0.5404318\n",
      "batch=293 accuracy=0.71779335 loss=0.62889564\n",
      "batch=294 accuracy=0.71800846 loss=0.65457183\n",
      "batch=295 accuracy=0.71801096 loss=0.59640175\n",
      "batch=296 accuracy=0.71801347 loss=0.6821197\n",
      "batch=297 accuracy=0.71822566 loss=0.60933495\n",
      "batch=298 accuracy=0.71833193 loss=0.5528558\n",
      "batch=299 accuracy=0.7190625 loss=0.5258521\n",
      "batch=300 accuracy=0.7195806 loss=0.48512483\n",
      "batch=301 accuracy=0.71968126 loss=0.6605216\n",
      "batch=302 accuracy=0.71998763 loss=0.47906762\n",
      "batch=303 accuracy=0.72039473 loss=0.42533645\n",
      "batch=304 accuracy=0.7204918 loss=0.5768213\n",
      "batch=305 accuracy=0.720384 loss=0.6499653\n",
      "batch=306 accuracy=0.72068405 loss=0.48976958\n",
      "batch=307 accuracy=0.7205763 loss=1.091543\n",
      "batch=308 accuracy=0.7208738 loss=0.6782802\n",
      "batch=309 accuracy=0.7214718 loss=0.2899105\n",
      "batch=310 accuracy=0.72196543 loss=0.5106101\n",
      "batch=311 accuracy=0.7225561 loss=0.3554115\n",
      "batch=312 accuracy=0.72274363 loss=0.5034184\n",
      "batch=313 accuracy=0.72332805 loss=0.23497358\n",
      "batch=314 accuracy=0.72400796 loss=0.2934247\n",
      "batch=315 accuracy=0.7240902 loss=0.51813114\n",
      "batch=316 accuracy=0.7244677 loss=0.5193397\n",
      "batch=317 accuracy=0.7248428 loss=0.42922366\n",
      "batch=318 accuracy=0.7253135 loss=0.377361\n",
      "batch=319 accuracy=0.72558594 loss=0.49048924\n",
      "batch=320 accuracy=0.7257593 loss=0.7336093\n",
      "batch=321 accuracy=0.7258346 loss=0.8956044\n",
      "batch=322 accuracy=0.7259095 loss=0.82032704\n",
      "batch=323 accuracy=0.7259838 loss=0.5760094\n",
      "batch=324 accuracy=0.72625 loss=0.507066\n",
      "batch=325 accuracy=0.7263228 loss=0.577802\n",
      "batch=326 accuracy=0.72610855 loss=0.9108792\n",
      "batch=327 accuracy=0.7265625 loss=0.44878364\n",
      "batch=328 accuracy=0.7268237 loss=0.60502565\n",
      "batch=329 accuracy=0.7266098 loss=0.7228793\n",
      "batch=330 accuracy=0.727247 loss=0.35282722\n",
      "batch=331 accuracy=0.72703314 loss=0.81069183\n",
      "batch=332 accuracy=0.7272898 loss=0.57344866\n",
      "batch=333 accuracy=0.7273578 loss=0.61200947\n",
      "batch=334 accuracy=0.72770524 loss=0.5892185\n",
      "batch=335 accuracy=0.7280506 loss=0.569204\n",
      "batch=336 accuracy=0.72830117 loss=0.42773694\n",
      "batch=337 accuracy=0.72836536 loss=0.74809396\n",
      "batch=338 accuracy=0.7285214 loss=0.7524204\n",
      "batch=339 accuracy=0.7287684 loss=0.5224056\n",
      "batch=340 accuracy=0.7291972 loss=0.4585392\n",
      "batch=341 accuracy=0.72889256 loss=0.8619792\n",
      "batch=342 accuracy=0.7292274 loss=0.49991146\n",
      "batch=343 accuracy=0.7294695 loss=0.5589335\n",
      "batch=344 accuracy=0.72961956 loss=0.44216067\n",
      "batch=345 accuracy=0.7302204 loss=0.33572116\n",
      "batch=346 accuracy=0.73009723 loss=0.76265764\n",
      "batch=347 accuracy=0.73015445 loss=0.6208107\n",
      "batch=348 accuracy=0.73047996 loss=0.590639\n",
      "batch=349 accuracy=0.7305357 loss=0.5884895\n",
      "batch=350 accuracy=0.7307692 loss=0.54068816\n",
      "batch=351 accuracy=0.73073506 loss=0.7837602\n",
      "batch=352 accuracy=0.7308782 loss=0.6325827\n",
      "batch=353 accuracy=0.7313736 loss=0.40472066\n",
      "batch=354 accuracy=0.7315141 loss=0.48405415\n",
      "batch=355 accuracy=0.73191714 loss=0.64856863\n",
      "batch=356 accuracy=0.7320553 loss=0.7391896\n",
      "batch=357 accuracy=0.73219275 loss=0.60556966\n",
      "batch=358 accuracy=0.73224235 loss=0.6601708\n",
      "batch=359 accuracy=0.73220485 loss=0.5359715\n",
      "batch=360 accuracy=0.732081 loss=0.8787967\n",
      "batch=361 accuracy=0.7321305 loss=0.62515175\n",
      "batch=362 accuracy=0.73217976 loss=0.5929769\n",
      "batch=363 accuracy=0.73214287 loss=0.7062036\n",
      "batch=364 accuracy=0.7322774 loss=0.5868438\n",
      "batch=365 accuracy=0.7324112 loss=0.60487175\n",
      "batch=366 accuracy=0.73237395 loss=0.7099694\n",
      "batch=367 accuracy=0.7325068 loss=0.72453904\n",
      "batch=368 accuracy=0.7324695 loss=0.7658018\n",
      "batch=369 accuracy=0.73277026 loss=0.46278265\n",
      "batch=370 accuracy=0.7330694 loss=0.3921923\n",
      "batch=371 accuracy=0.7331149 loss=0.78927207\n",
      "batch=372 accuracy=0.73299265 loss=0.68703467\n",
      "batch=373 accuracy=0.732871 loss=0.64683795\n",
      "batch=374 accuracy=0.7328333 loss=0.67273533\n",
      "batch=375 accuracy=0.7329621 loss=0.64870447\n",
      "batch=376 accuracy=0.733256 loss=0.47328657\n",
      "batch=377 accuracy=0.73363096 loss=0.30689976\n",
      "batch=378 accuracy=0.7339215 loss=0.5263915\n",
      "batch=379 accuracy=0.7338816 loss=0.82159793\n",
      "batch=380 accuracy=0.7340059 loss=0.7008175\n",
      "batch=381 accuracy=0.7344568 loss=0.39662874\n",
      "batch=382 accuracy=0.73482376 loss=0.42507595\n",
      "batch=383 accuracy=0.7348633 loss=0.76429397\n",
      "batch=384 accuracy=0.73482144 loss=0.6713113\n",
      "batch=385 accuracy=0.7348608 loss=0.630602\n",
      "batch=386 accuracy=0.7351421 loss=0.4607946\n",
      "batch=387 accuracy=0.73518044 loss=0.72991025\n",
      "batch=388 accuracy=0.73537916 loss=0.52027583\n",
      "batch=389 accuracy=0.73525643 loss=0.8660767\n",
      "batch=390 accuracy=0.7352941 loss=0.6276661\n",
      "batch=391 accuracy=0.7354911 loss=0.5607534\n",
      "batch=392 accuracy=0.73576653 loss=0.48584688\n",
      "batch=393 accuracy=0.7359613 loss=0.6667969\n",
      "batch=394 accuracy=0.7361551 loss=0.45951948\n",
      "batch=395 accuracy=0.73634785 loss=0.39038074\n",
      "batch=396 accuracy=0.7361461 loss=0.63250345\n",
      "batch=397 accuracy=0.7367305 loss=0.24291807\n",
      "batch=398 accuracy=0.73668545 loss=0.5948617\n",
      "batch=399 accuracy=0.7370312 loss=0.44715434\n",
      "batch=400 accuracy=0.73729736 loss=0.5127121\n",
      "batch=401 accuracy=0.7371735 loss=0.58653605\n",
      "batch=402 accuracy=0.73705024 loss=0.5890035\n",
      "batch=403 accuracy=0.73715967 loss=0.63149184\n",
      "batch=404 accuracy=0.7371142 loss=0.72336704\n",
      "batch=405 accuracy=0.73729986 loss=0.58379865\n",
      "batch=406 accuracy=0.7373311 loss=0.5254511\n",
      "batch=407 accuracy=0.7375919 loss=0.38879317\n",
      "batch=408 accuracy=0.73762226 loss=0.6676431\n",
      "batch=409 accuracy=0.7378049 loss=0.3768615\n",
      "batch=410 accuracy=0.73768246 loss=0.855461\n",
      "batch=411 accuracy=0.7375607 loss=0.7202548\n",
      "batch=412 accuracy=0.7378178 loss=0.32024455\n",
      "batch=413 accuracy=0.73777175 loss=0.7307569\n",
      "batch=414 accuracy=0.7379518 loss=0.5531736\n",
      "batch=415 accuracy=0.738131 loss=0.6980269\n",
      "batch=416 accuracy=0.7383093 loss=0.58279884\n",
      "batch=417 accuracy=0.7384121 loss=0.49470016\n",
      "batch=418 accuracy=0.7388126 loss=0.35510373\n",
      "batch=419 accuracy=0.73913693 loss=0.47191352\n",
      "batch=420 accuracy=0.73938537 loss=0.40357792\n",
      "batch=421 accuracy=0.7394105 loss=0.6042004\n",
      "batch=422 accuracy=0.7395833 loss=0.5815421\n",
      "batch=423 accuracy=0.73975533 loss=0.4854058\n",
      "batch=424 accuracy=0.7400735 loss=0.33226213\n",
      "batch=425 accuracy=0.740537 loss=0.33719277\n",
      "batch=426 accuracy=0.7407055 loss=0.7628186\n",
      "batch=427 accuracy=0.74072725 loss=0.7688726\n",
      "batch=428 accuracy=0.74082166 loss=0.57307494\n",
      "batch=429 accuracy=0.74106103 loss=0.44524294\n",
      "batch=430 accuracy=0.7411543 loss=0.8549963\n",
      "batch=431 accuracy=0.74117476 loss=0.5495777\n",
      "batch=432 accuracy=0.7413395 loss=0.67238426\n",
      "batch=433 accuracy=0.7415035 loss=0.49315566\n",
      "batch=434 accuracy=0.7411638 loss=0.79522055\n",
      "batch=435 accuracy=0.7410407 loss=0.80069774\n",
      "batch=436 accuracy=0.7410612 loss=0.51311773\n",
      "batch=437 accuracy=0.7409389 loss=0.67193973\n",
      "batch=438 accuracy=0.7412443 loss=0.4011768\n",
      "batch=439 accuracy=0.7415483 loss=0.37004298\n",
      "batch=440 accuracy=0.7414966 loss=0.9686492\n",
      "batch=441 accuracy=0.74172795 loss=0.58593804\n",
      "batch=442 accuracy=0.7418877 loss=0.48659894\n",
      "batch=443 accuracy=0.7418356 loss=0.89323485\n",
      "batch=444 accuracy=0.7422051 loss=0.5394596\n",
      "batch=445 accuracy=0.7424327 loss=0.65214574\n",
      "batch=446 accuracy=0.7425895 loss=0.5988083\n",
      "batch=447 accuracy=0.7428153 loss=0.54219687\n",
      "batch=448 accuracy=0.7431097 loss=0.50154865\n",
      "batch=449 accuracy=0.74319446 loss=0.54280925\n",
      "batch=450 accuracy=0.7432788 loss=0.5530865\n",
      "batch=451 accuracy=0.7435011 loss=0.52992153\n",
      "batch=452 accuracy=0.7437914 loss=0.5765809\n",
      "batch=453 accuracy=0.74421805 loss=0.27341074\n",
      "batch=454 accuracy=0.7440934 loss=0.9143107\n",
      "batch=455 accuracy=0.7441749 loss=0.60745275\n",
      "batch=456 accuracy=0.7443244 loss=0.5407591\n",
      "batch=457 accuracy=0.74454147 loss=0.53825945\n",
      "batch=458 accuracy=0.74475765 loss=0.5367071\n",
      "batch=459 accuracy=0.7449728 loss=0.4989591\n",
      "batch=460 accuracy=0.74498373 loss=0.6490945\n",
      "batch=461 accuracy=0.74506223 loss=0.5389902\n",
      "batch=462 accuracy=0.7451404 loss=0.44014204\n",
      "batch=463 accuracy=0.7452856 loss=0.770028\n",
      "batch=464 accuracy=0.7454973 loss=0.37546748\n",
      "batch=465 accuracy=0.7458423 loss=0.40960938\n",
      "batch=466 accuracy=0.74598503 loss=0.70844615\n",
      "batch=467 accuracy=0.7459268 loss=0.7576155\n",
      "batch=468 accuracy=0.74626863 loss=0.3286243\n",
      "batch=469 accuracy=0.7463431 loss=0.60989463\n",
      "batch=470 accuracy=0.74661624 loss=0.46492448\n",
      "batch=471 accuracy=0.74702066 loss=0.29832515\n",
      "batch=472 accuracy=0.74715906 loss=0.48947787\n",
      "batch=473 accuracy=0.74703324 loss=1.0576513\n",
      "batch=474 accuracy=0.74723685 loss=0.41590652\n",
      "batch=475 accuracy=0.7474396 loss=0.59696734\n",
      "batch=476 accuracy=0.7475105 loss=0.49672663\n",
      "batch=477 accuracy=0.74764645 loss=0.5545091\n",
      "batch=478 accuracy=0.74752086 loss=0.7379403\n",
      "batch=479 accuracy=0.74759114 loss=0.7042282\n",
      "batch=480 accuracy=0.74759614 loss=0.73470056\n",
      "batch=481 accuracy=0.7477308 loss=0.4894261\n",
      "batch=482 accuracy=0.74780023 loss=0.51990986\n",
      "batch=483 accuracy=0.7475465 loss=0.97689474\n",
      "batch=484 accuracy=0.7474227 loss=0.8746206\n",
      "batch=485 accuracy=0.74755657 loss=0.50870776\n",
      "batch=486 accuracy=0.7472408 loss=0.8657117\n",
      "batch=487 accuracy=0.7473745 loss=0.54628456\n",
      "batch=488 accuracy=0.7477633 loss=0.2892055\n",
      "batch=489 accuracy=0.74783164 loss=0.42546356\n",
      "batch=490 accuracy=0.74802697 loss=0.34904516\n",
      "batch=491 accuracy=0.74815804 loss=0.6265742\n",
      "batch=492 accuracy=0.7479716 loss=0.89431995\n",
      "batch=493 accuracy=0.7482287 loss=0.35653192\n",
      "batch=494 accuracy=0.7482323 loss=0.70233536\n",
      "batch=495 accuracy=0.7481729 loss=0.644238\n",
      "batch=496 accuracy=0.7485538 loss=0.2643938\n",
      "batch=497 accuracy=0.7486822 loss=0.6167303\n",
      "batch=498 accuracy=0.7486849 loss=0.5465984\n",
      "batch=499 accuracy=0.74875 loss=0.53234446\n",
      "batch=500 accuracy=0.74893963 loss=0.42572486\n",
      "batch=501 accuracy=0.749004 loss=0.5699904\n",
      "batch=502 accuracy=0.7490681 loss=0.46221218\n",
      "batch=503 accuracy=0.74900794 loss=0.70969987\n",
      "batch=504 accuracy=0.7490718 loss=0.6623541\n",
      "batch=505 accuracy=0.7494442 loss=0.32792947\n",
      "batch=506 accuracy=0.7495685 loss=0.5847865\n",
      "batch=507 accuracy=0.74950784 loss=0.6407883\n",
      "batch=508 accuracy=0.7498158 loss=0.32639432\n",
      "batch=509 accuracy=0.7498162 loss=0.5205115\n",
      "batch=510 accuracy=0.74993885 loss=0.57801616\n",
      "batch=511 accuracy=0.75006104 loss=0.5176378\n",
      "batch=512 accuracy=0.75012183 loss=0.53914297\n",
      "batch=513 accuracy=0.750304 loss=0.54709196\n",
      "batch=514 accuracy=0.75042474 loss=0.5165181\n",
      "batch=515 accuracy=0.75048447 loss=0.5719578\n",
      "batch=516 accuracy=0.7503022 loss=0.7050616\n",
      "batch=517 accuracy=0.7504826 loss=0.5435599\n",
      "batch=518 accuracy=0.7504817 loss=0.7457087\n",
      "batch=519 accuracy=0.75072116 loss=0.41010422\n",
      "batch=520 accuracy=0.75101966 loss=0.5240942\n",
      "batch=521 accuracy=0.7510776 loss=0.5753012\n",
      "batch=522 accuracy=0.7512548 loss=0.7171525\n",
      "batch=523 accuracy=0.7511331 loss=0.64199686\n",
      "batch=524 accuracy=0.7511905 loss=0.5520054\n",
      "batch=525 accuracy=0.7514853 loss=0.3340202\n",
      "batch=526 accuracy=0.75160104 loss=0.5668137\n",
      "batch=527 accuracy=0.75177556 loss=0.41578755\n",
      "batch=528 accuracy=0.7520085 loss=0.41743505\n",
      "batch=529 accuracy=0.75182784 loss=0.74496186\n",
      "batch=530 accuracy=0.75188327 loss=0.6684824\n",
      "batch=531 accuracy=0.751821 loss=0.6175832\n",
      "batch=532 accuracy=0.7517003 loss=0.66903865\n",
      "batch=533 accuracy=0.7517556 loss=0.6531769\n",
      "batch=534 accuracy=0.75192755 loss=0.49549592\n",
      "batch=535 accuracy=0.7519823 loss=0.59159\n",
      "batch=536 accuracy=0.75215316 loss=0.3013693\n",
      "batch=537 accuracy=0.75232345 loss=0.41947055\n",
      "batch=538 accuracy=0.752551 loss=0.35312682\n",
      "batch=539 accuracy=0.75248843 loss=0.64716005\n",
      "batch=540 accuracy=0.7527149 loss=0.36575085\n",
      "batch=541 accuracy=0.7529982 loss=0.25205916\n",
      "batch=542 accuracy=0.7530502 loss=0.55608034\n",
      "batch=543 accuracy=0.7532169 loss=0.6891588\n",
      "batch=544 accuracy=0.753211 loss=0.5266914\n",
      "batch=545 accuracy=0.75337684 loss=0.5719296\n",
      "batch=546 accuracy=0.75337064 loss=0.75443524\n",
      "batch=547 accuracy=0.75364965 loss=0.31323916\n",
      "batch=548 accuracy=0.75387067 loss=0.43504104\n",
      "batch=549 accuracy=0.7540341 loss=0.35066515\n",
      "batch=550 accuracy=0.75419694 loss=0.69537616\n",
      "batch=551 accuracy=0.7543591 loss=0.4561131\n",
      "batch=552 accuracy=0.75446427 loss=0.70315874\n",
      "batch=553 accuracy=0.7547383 loss=0.3472193\n",
      "batch=554 accuracy=0.75472975 loss=0.67548877\n",
      "batch=555 accuracy=0.75483364 loss=0.6231906\n",
      "batch=556 accuracy=0.75499326 loss=0.40941155\n",
      "batch=557 accuracy=0.75537634 loss=0.13436103\n",
      "batch=558 accuracy=0.75570214 loss=0.41100585\n",
      "batch=559 accuracy=0.75563616 loss=0.5937595\n",
      "batch=560 accuracy=0.7557932 loss=0.2912472\n",
      "batch=561 accuracy=0.75594974 loss=0.3956687\n",
      "batch=562 accuracy=0.7559392 loss=0.4816509\n",
      "batch=563 accuracy=0.75615025 loss=0.50603867\n",
      "batch=564 accuracy=0.75625 loss=0.61116505\n",
      "batch=565 accuracy=0.7564046 loss=0.6409245\n",
      "batch=566 accuracy=0.7564484 loss=0.68711734\n",
      "batch=567 accuracy=0.75638205 loss=0.77637833\n",
      "batch=568 accuracy=0.7565905 loss=0.43969443\n",
      "batch=569 accuracy=0.75685304 loss=0.4004915\n",
      "batch=570 accuracy=0.75706 loss=0.38798696\n",
      "batch=571 accuracy=0.75726616 loss=0.29378313\n",
      "batch=572 accuracy=0.75725347 loss=0.8718411\n",
      "batch=573 accuracy=0.7574042 loss=0.39127392\n",
      "batch=574 accuracy=0.7571739 loss=0.798072\n",
      "batch=575 accuracy=0.75716144 loss=0.6631733\n",
      "batch=576 accuracy=0.75704074 loss=0.57919884\n",
      "batch=577 accuracy=0.757407 loss=0.29834273\n",
      "batch=578 accuracy=0.7576101 loss=0.35439867\n",
      "batch=579 accuracy=0.75738144 loss=0.8964193\n",
      "batch=580 accuracy=0.75747633 loss=0.6228863\n",
      "batch=581 accuracy=0.7574098 loss=0.9184893\n",
      "batch=582 accuracy=0.7576115 loss=0.4159209\n",
      "batch=583 accuracy=0.7578125 loss=0.36780122\n",
      "batch=584 accuracy=0.75790596 loss=0.47806364\n",
      "batch=585 accuracy=0.7578925 loss=0.8590381\n",
      "batch=586 accuracy=0.75798553 loss=0.4430673\n",
      "batch=587 accuracy=0.75818455 loss=0.3991701\n",
      "batch=588 accuracy=0.75827676 loss=0.53919196\n",
      "batch=589 accuracy=0.7584216 loss=0.51277256\n",
      "batch=590 accuracy=0.75867176 loss=0.4421046\n",
      "batch=591 accuracy=0.7587099 loss=0.5301552\n",
      "batch=592 accuracy=0.7588533 loss=0.6548015\n",
      "batch=593 accuracy=0.75883836 loss=0.8235589\n",
      "batch=594 accuracy=0.7588235 loss=0.51800126\n",
      "batch=595 accuracy=0.7590709 loss=0.4943263\n",
      "batch=596 accuracy=0.75926507 loss=0.30622455\n",
      "batch=597 accuracy=0.7594586 loss=0.5528748\n",
      "batch=598 accuracy=0.7594428 loss=0.65534854\n",
      "batch=599 accuracy=0.7596354 loss=0.50051147\n",
      "batch=600 accuracy=0.7597754 loss=0.413203\n",
      "batch=601 accuracy=0.7599668 loss=0.3617974\n",
      "batch=602 accuracy=0.7600539 loss=0.58849263\n",
      "batch=603 accuracy=0.7599855 loss=0.773224\n",
      "batch=604 accuracy=0.7600207 loss=0.47276995\n",
      "batch=605 accuracy=0.7603135 loss=0.26079538\n",
      "batch=606 accuracy=0.760348 loss=0.48934937\n",
      "batch=607 accuracy=0.7604852 loss=0.6351609\n",
      "batch=608 accuracy=0.76051927 loss=0.5722244\n",
      "batch=609 accuracy=0.76070696 loss=0.56537473\n",
      "batch=610 accuracy=0.7607406 loss=0.6306581\n",
      "batch=611 accuracy=0.7607741 loss=0.61619735\n",
      "batch=612 accuracy=0.76090944 loss=0.4696918\n",
      "batch=613 accuracy=0.7608408 loss=0.7211967\n",
      "batch=614 accuracy=0.7610772 loss=0.3872139\n",
      "batch=615 accuracy=0.76121145 loss=0.44222417\n",
      "batch=616 accuracy=0.7613959 loss=0.38118005\n",
      "batch=617 accuracy=0.7615797 loss=0.5138384\n",
      "batch=618 accuracy=0.76171243 loss=0.42487332\n",
      "batch=619 accuracy=0.76174396 loss=0.6089908\n",
      "batch=620 accuracy=0.7618257 loss=0.5358664\n",
      "batch=621 accuracy=0.7619574 loss=0.38641235\n",
      "batch=622 accuracy=0.76213884 loss=0.4611044\n",
      "batch=623 accuracy=0.7621695 loss=0.5185801\n",
      "batch=624 accuracy=0.7624 loss=0.33473122\n",
      "batch=625 accuracy=0.76243013 loss=0.49354887\n",
      "batch=626 accuracy=0.76260966 loss=0.36803013\n",
      "batch=627 accuracy=0.7627886 loss=0.44703782\n",
      "batch=628 accuracy=0.7627186 loss=0.84055734\n",
      "batch=629 accuracy=0.76289684 loss=0.45065564\n",
      "batch=630 accuracy=0.76282686 loss=0.85909367\n",
      "batch=631 accuracy=0.7629055 loss=0.4988094\n",
      "batch=632 accuracy=0.7629838 loss=0.51329875\n",
      "batch=633 accuracy=0.7631112 loss=0.58053935\n",
      "batch=634 accuracy=0.76299214 loss=0.7162589\n",
      "batch=635 accuracy=0.7631191 loss=0.48160374\n",
      "batch=636 accuracy=0.76329476 loss=0.6236565\n",
      "batch=637 accuracy=0.7633719 loss=0.43868905\n",
      "batch=638 accuracy=0.7634488 loss=0.6722958\n",
      "batch=639 accuracy=0.76357424 loss=0.6267524\n",
      "batch=640 accuracy=0.7636993 loss=0.5586594\n",
      "batch=641 accuracy=0.763824 loss=0.4834448\n",
      "batch=642 accuracy=0.7638511 loss=0.5302727\n",
      "batch=643 accuracy=0.7638296 loss=0.58046776\n",
      "batch=644 accuracy=0.7637597 loss=0.7604481\n",
      "batch=645 accuracy=0.76383513 loss=0.4852152\n",
      "batch=646 accuracy=0.764007 loss=0.3835523\n",
      "batch=647 accuracy=0.7640818 loss=0.56635517\n",
      "batch=648 accuracy=0.7641564 loss=0.49849808\n",
      "batch=649 accuracy=0.7644712 loss=0.18800844\n",
      "batch=650 accuracy=0.7645449 loss=0.3695371\n",
      "batch=651 accuracy=0.76471436 loss=0.4281561\n",
      "batch=652 accuracy=0.76464397 loss=0.87883604\n",
      "batch=653 accuracy=0.7647649 loss=0.5642084\n",
      "batch=654 accuracy=0.7649809 loss=0.2659446\n",
      "batch=655 accuracy=0.7648628 loss=0.6985283\n",
      "batch=656 accuracy=0.76488775 loss=0.4991585\n",
      "batch=657 accuracy=0.76515007 loss=0.22863427\n",
      "batch=658 accuracy=0.76512706 loss=0.77560437\n",
      "batch=659 accuracy=0.7652462 loss=0.4721165\n",
      "batch=660 accuracy=0.765365 loss=0.5842219\n",
      "batch=661 accuracy=0.7655778 loss=0.33902037\n",
      "batch=662 accuracy=0.76564854 loss=0.5109874\n",
      "batch=663 accuracy=0.7657662 loss=0.4209131\n",
      "batch=664 accuracy=0.7658365 loss=0.54855835\n",
      "batch=665 accuracy=0.7658127 loss=0.5020244\n",
      "batch=666 accuracy=0.765789 loss=0.85273427\n",
      "batch=667 accuracy=0.76599926 loss=0.49923342\n",
      "batch=668 accuracy=0.7661155 loss=0.59553325\n",
      "batch=669 accuracy=0.76623136 loss=0.6330023\n",
      "batch=670 accuracy=0.7663003 loss=0.49731973\n",
      "batch=671 accuracy=0.7665086 loss=0.33827955\n",
      "batch=672 accuracy=0.76666975 loss=0.38040257\n",
      "batch=673 accuracy=0.76678413 loss=0.46718395\n",
      "batch=674 accuracy=0.7667593 loss=0.6359886\n",
      "batch=675 accuracy=0.7668269 loss=0.47679237\n",
      "batch=676 accuracy=0.7668482 loss=0.5834473\n",
      "batch=677 accuracy=0.76691556 loss=0.5705736\n",
      "batch=678 accuracy=0.76693666 loss=0.4888556\n",
      "batch=679 accuracy=0.7669577 loss=0.77703846\n",
      "batch=680 accuracy=0.7669787 loss=0.45502868\n",
      "batch=681 accuracy=0.76677054 loss=0.77174246\n",
      "batch=682 accuracy=0.766929 loss=0.3041285\n",
      "batch=683 accuracy=0.767087 loss=0.34298486\n",
      "batch=684 accuracy=0.7669708 loss=0.9302355\n",
      "batch=685 accuracy=0.76708275 loss=0.43590778\n",
      "batch=686 accuracy=0.76714885 loss=0.51699126\n",
      "batch=687 accuracy=0.7673056 loss=0.5008657\n",
      "batch=688 accuracy=0.76741654 loss=0.37184727\n",
      "batch=689 accuracy=0.76757246 loss=0.3740208\n",
      "batch=690 accuracy=0.7676375 loss=0.38448614\n",
      "batch=691 accuracy=0.76774746 loss=0.5335141\n",
      "batch=692 accuracy=0.76776695 loss=0.37600172\n",
      "batch=693 accuracy=0.76787645 loss=0.34855646\n",
      "batch=694 accuracy=0.76794064 loss=0.44765162\n",
      "batch=695 accuracy=0.7679149 loss=0.4303711\n",
      "batch=696 accuracy=0.76797885 loss=0.4705148\n",
      "batch=697 accuracy=0.7679531 loss=0.5787289\n",
      "batch=698 accuracy=0.7678827 loss=0.7378566\n",
      "batch=699 accuracy=0.7679464 loss=0.639347\n",
      "batch=700 accuracy=0.76800996 loss=0.5202683\n",
      "batch=701 accuracy=0.7682514 loss=0.26842713\n",
      "batch=702 accuracy=0.76849216 loss=0.42099857\n",
      "batch=703 accuracy=0.7685991 loss=0.71358734\n",
      "batch=704 accuracy=0.76870567 loss=0.38830584\n",
      "batch=705 accuracy=0.76881194 loss=0.54962176\n",
      "batch=706 accuracy=0.76896214 loss=0.43918258\n",
      "batch=707 accuracy=0.7690678 loss=0.5393132\n",
      "batch=708 accuracy=0.769085 loss=0.6291646\n",
      "batch=709 accuracy=0.76919013 loss=0.39136586\n",
      "batch=710 accuracy=0.7694708 loss=0.22142377\n",
      "batch=711 accuracy=0.7694874 loss=0.6536272\n",
      "batch=712 accuracy=0.7695477 loss=0.51666\n",
      "batch=713 accuracy=0.76973915 loss=0.25888178\n",
      "batch=714 accuracy=0.76971155 loss=0.9773836\n",
      "batch=715 accuracy=0.76981497 loss=0.35906976\n",
      "batch=716 accuracy=0.76974374 loss=0.7311863\n",
      "batch=717 accuracy=0.7698903 loss=0.32576317\n",
      "batch=718 accuracy=0.76986265 loss=0.4313964\n",
      "batch=719 accuracy=0.76983505 loss=0.7761622\n",
      "batch=720 accuracy=0.7699809 loss=0.37109786\n",
      "batch=721 accuracy=0.7701264 loss=0.48549244\n",
      "batch=722 accuracy=0.7701418 loss=0.5985287\n",
      "batch=723 accuracy=0.7702866 loss=0.493546\n",
      "batch=724 accuracy=0.77038795 loss=0.42796347\n",
      "batch=725 accuracy=0.77044594 loss=0.86932135\n",
      "batch=726 accuracy=0.77058977 loss=0.45927253\n",
      "batch=727 accuracy=0.7707332 loss=0.51035494\n",
      "batch=728 accuracy=0.7709191 loss=0.34710696\n",
      "batch=729 accuracy=0.7708904 loss=0.97876674\n",
      "batch=730 accuracy=0.77094734 loss=0.5249564\n",
      "batch=731 accuracy=0.7710041 loss=0.42534786\n",
      "batch=732 accuracy=0.7710181 loss=0.5676296\n",
      "batch=733 accuracy=0.77115977 loss=0.5343307\n",
      "batch=734 accuracy=0.7713435 loss=0.30799088\n",
      "batch=735 accuracy=0.7715268 loss=0.32789773\n",
      "batch=736 accuracy=0.77141285 loss=0.77669847\n",
      "batch=737 accuracy=0.77159554 loss=0.37225276\n",
      "batch=738 accuracy=0.77169317 loss=0.413395\n",
      "batch=739 accuracy=0.77162164 loss=0.6535162\n",
      "batch=740 accuracy=0.77159244 loss=0.66643304\n",
      "batch=741 accuracy=0.7716476 loss=0.65922356\n",
      "batch=742 accuracy=0.7717446 loss=0.3915363\n",
      "batch=743 accuracy=0.7718414 loss=0.42073235\n",
      "batch=744 accuracy=0.7719379 loss=0.44718158\n",
      "batch=745 accuracy=0.7720761 loss=0.42719042\n",
      "batch=746 accuracy=0.77208835 loss=0.5187539\n",
      "batch=747 accuracy=0.77214235 loss=0.5439384\n",
      "batch=748 accuracy=0.77223796 loss=0.48354587\n",
      "batch=749 accuracy=0.77229166 loss=0.5300602\n",
      "batch=750 accuracy=0.772262 loss=0.68806404\n",
      "batch=751 accuracy=0.77248174 loss=0.30615252\n",
      "batch=752 accuracy=0.7726179 loss=0.8663972\n",
      "batch=753 accuracy=0.77258784 loss=0.53251946\n",
      "batch=754 accuracy=0.7726407 loss=0.44794497\n",
      "batch=755 accuracy=0.77269346 loss=0.48748127\n",
      "batch=756 accuracy=0.7727048 loss=0.6176576\n",
      "batch=757 accuracy=0.7728397 loss=0.30350298\n",
      "batch=758 accuracy=0.7726861 loss=0.6509161\n",
      "batch=759 accuracy=0.7727796 loss=0.36381868\n",
      "batch=760 accuracy=0.7726675 loss=0.8279885\n",
      "batch=761 accuracy=0.7727608 loss=0.36079752\n",
      "batch=762 accuracy=0.77285385 loss=0.6315297\n",
      "batch=763 accuracy=0.77294666 loss=0.59753376\n",
      "batch=764 accuracy=0.7730392 loss=0.6243484\n",
      "batch=765 accuracy=0.7731723 loss=0.399659\n",
      "batch=766 accuracy=0.77318287 loss=0.45915204\n",
      "batch=767 accuracy=0.7733968 loss=0.2330836\n",
      "batch=768 accuracy=0.77352893 loss=0.3818658\n",
      "batch=769 accuracy=0.77357954 loss=0.480036\n",
      "batch=770 accuracy=0.7734679 loss=0.7226724\n",
      "batch=771 accuracy=0.77359945 loss=0.3836763\n",
      "batch=772 accuracy=0.77377105 loss=0.28874493\n",
      "batch=773 accuracy=0.7738614 loss=0.4025511\n",
      "batch=774 accuracy=0.7739113 loss=0.45988685\n",
      "batch=775 accuracy=0.77404153 loss=0.37897575\n",
      "batch=776 accuracy=0.7739704 loss=0.650903\n",
      "batch=777 accuracy=0.77389944 loss=0.6209913\n",
      "batch=778 accuracy=0.7740693 loss=0.45196873\n",
      "batch=779 accuracy=0.77415866 loss=0.4401301\n",
      "batch=780 accuracy=0.77424777 loss=0.37972778\n",
      "batch=781 accuracy=0.77433664 loss=0.562251\n",
      "batch=782 accuracy=0.7744652 loss=0.4400655\n",
      "batch=783 accuracy=0.7743543 loss=0.6863823\n",
      "batch=784 accuracy=0.7744825 loss=0.3942479\n",
      "batch=785 accuracy=0.77461034 loss=0.37709993\n",
      "batch=786 accuracy=0.7746188 loss=0.5126306\n",
      "batch=787 accuracy=0.7746669 loss=0.551079\n",
      "batch=788 accuracy=0.7746356 loss=0.7878357\n",
      "batch=789 accuracy=0.77464396 loss=0.4532174\n",
      "batch=790 accuracy=0.77477086 loss=0.35845295\n",
      "batch=791 accuracy=0.77493685 loss=0.3683398\n",
      "batch=792 accuracy=0.77498424 loss=0.48939297\n",
      "batch=793 accuracy=0.77522826 loss=0.26488513\n",
      "batch=794 accuracy=0.77519655 loss=0.73057437\n",
      "batch=795 accuracy=0.7753219 loss=0.3602846\n",
      "batch=796 accuracy=0.77521175 loss=0.84784806\n",
      "batch=797 accuracy=0.77529764 loss=0.48581254\n",
      "batch=798 accuracy=0.7754224 loss=0.5245609\n",
      "batch=799 accuracy=0.77542967 loss=0.41845256\n",
      "batch=800 accuracy=0.775554 loss=0.36288798\n",
      "batch=801 accuracy=0.775678 loss=0.37745768\n",
      "batch=802 accuracy=0.7755682 loss=0.7205578\n",
      "batch=803 accuracy=0.775653 loss=0.3489775\n",
      "batch=804 accuracy=0.7757764 loss=0.35486287\n",
      "batch=805 accuracy=0.775822 loss=0.48298374\n",
      "batch=806 accuracy=0.7759448 loss=0.55198175\n",
      "batch=807 accuracy=0.77591276 loss=0.72975016\n",
      "batch=808 accuracy=0.7759193 loss=0.54254854\n",
      "batch=809 accuracy=0.77600306 loss=0.37655392\n",
      "batch=810 accuracy=0.77612513 loss=0.34683067\n",
      "batch=811 accuracy=0.77613145 loss=0.6725658\n",
      "batch=812 accuracy=0.7762531 loss=0.49731582\n",
      "batch=813 accuracy=0.7762976 loss=0.7370634\n",
      "batch=814 accuracy=0.77638036 loss=0.43404835\n",
      "batch=815 accuracy=0.7765395 loss=0.34992602\n",
      "batch=816 accuracy=0.7765453 loss=0.575266\n",
      "batch=817 accuracy=0.7766274 loss=0.3831433\n",
      "batch=818 accuracy=0.77667123 loss=0.44892395\n",
      "batch=819 accuracy=0.77682924 loss=0.3061214\n",
      "batch=820 accuracy=0.77683467 loss=0.51982665\n",
      "batch=821 accuracy=0.77684003 loss=0.38584864\n",
      "batch=822 accuracy=0.77692133 loss=0.39200515\n",
      "batch=823 accuracy=0.7767749 loss=0.8295521\n",
      "batch=824 accuracy=0.7768939 loss=0.31624818\n",
      "batch=825 accuracy=0.77697486 loss=0.5891869\n",
      "batch=826 accuracy=0.7770934 loss=0.35598916\n",
      "batch=827 accuracy=0.77713615 loss=0.38989067\n",
      "batch=828 accuracy=0.7771788 loss=0.52590644\n",
      "batch=829 accuracy=0.77725905 loss=0.7793188\n",
      "batch=830 accuracy=0.7772638 loss=0.46647364\n",
      "batch=831 accuracy=0.77734375 loss=0.39903313\n",
      "batch=832 accuracy=0.77734846 loss=0.82681257\n",
      "batch=833 accuracy=0.7773531 loss=0.5541079\n",
      "batch=834 accuracy=0.77758235 loss=0.24999575\n",
      "batch=835 accuracy=0.77762413 loss=0.44330466\n",
      "batch=836 accuracy=0.7777031 loss=0.61062247\n",
      "batch=837 accuracy=0.77774465 loss=0.42129585\n",
      "batch=838 accuracy=0.7778605 loss=0.33567083\n",
      "batch=839 accuracy=0.77793896 loss=0.602757\n",
      "batch=840 accuracy=0.7777943 loss=0.87869024\n",
      "batch=841 accuracy=0.7779468 loss=0.33356953\n",
      "batch=842 accuracy=0.77809906 loss=0.30309057\n",
      "batch=843 accuracy=0.77806574 loss=0.51165783\n",
      "batch=844 accuracy=0.7781805 loss=0.71281415\n",
      "batch=845 accuracy=0.778221 loss=0.4846538\n",
      "batch=846 accuracy=0.7782984 loss=0.42382896\n",
      "batch=847 accuracy=0.7783019 loss=0.7468718\n",
      "batch=848 accuracy=0.7784526 loss=0.2715258\n",
      "batch=849 accuracy=0.77841914 loss=0.5135158\n",
      "batch=850 accuracy=0.7784959 loss=0.38426626\n",
      "batch=851 accuracy=0.7785725 loss=0.49327105\n",
      "batch=852 accuracy=0.77861226 loss=0.7807648\n",
      "batch=853 accuracy=0.77865195 loss=0.41363776\n",
      "batch=854 accuracy=0.7786184 loss=0.6064234\n",
      "batch=855 accuracy=0.7787675 loss=0.3827371\n",
      "batch=856 accuracy=0.7788798 loss=0.30915454\n",
      "batch=857 accuracy=0.778919 loss=0.3943099\n",
      "batch=858 accuracy=0.77903086 loss=0.3372854\n",
      "batch=859 accuracy=0.77914244 loss=0.49346194\n",
      "batch=860 accuracy=0.7791086 loss=0.7927376\n",
      "batch=861 accuracy=0.7793286 loss=0.23322767\n",
      "batch=862 accuracy=0.77940327 loss=0.4088288\n",
      "batch=863 accuracy=0.77944154 loss=0.5352853\n",
      "batch=864 accuracy=0.77955204 loss=0.34992576\n",
      "batch=865 accuracy=0.77966225 loss=0.487194\n",
      "batch=866 accuracy=0.7796641 loss=0.6130115\n",
      "batch=867 accuracy=0.7795939 loss=0.6963666\n",
      "batch=868 accuracy=0.77966774 loss=0.41464195\n",
      "batch=869 accuracy=0.7796695 loss=0.52699184\n",
      "batch=870 accuracy=0.7796714 loss=0.5715296\n",
      "batch=871 accuracy=0.77967316 loss=0.4475478\n",
      "batch=872 accuracy=0.7797466 loss=0.43688232\n",
      "batch=873 accuracy=0.7798198 loss=0.5118427\n",
      "batch=874 accuracy=0.77992857 loss=0.335168\n",
      "batch=875 accuracy=0.77996576 loss=0.49663815\n",
      "batch=876 accuracy=0.78000283 loss=0.45023766\n",
      "batch=877 accuracy=0.7801111 loss=0.5515498\n",
      "batch=878 accuracy=0.7800768 loss=0.6396696\n",
      "batch=879 accuracy=0.7802557 loss=0.34633362\n",
      "batch=880 accuracy=0.7802923 loss=0.56369424\n",
      "batch=881 accuracy=0.7803997 loss=0.279479\n",
      "batch=882 accuracy=0.7804714 loss=0.37244636\n",
      "batch=883 accuracy=0.7805076 loss=0.44168627\n",
      "batch=884 accuracy=0.7805438 loss=0.8266022\n",
      "batch=885 accuracy=0.7805093 loss=0.5853821\n",
      "batch=886 accuracy=0.78054535 loss=0.5950657\n",
      "batch=887 accuracy=0.7805462 loss=0.6125623\n",
      "batch=888 accuracy=0.78058213 loss=0.423346\n",
      "batch=889 accuracy=0.7806882 loss=0.32144296\n",
      "batch=890 accuracy=0.78061867 loss=0.61516607\n",
      "batch=891 accuracy=0.78065443 loss=0.66280687\n",
      "batch=892 accuracy=0.7806901 loss=0.44445714\n",
      "batch=893 accuracy=0.7806907 loss=0.40374434\n",
      "batch=894 accuracy=0.7808659 loss=0.3433694\n",
      "batch=895 accuracy=0.7808663 loss=0.64385474\n",
      "batch=896 accuracy=0.78100616 loss=0.3910774\n",
      "batch=897 accuracy=0.7809716 loss=0.5411908\n",
      "batch=898 accuracy=0.7811805 loss=0.15131679\n",
      "batch=899 accuracy=0.78118056 loss=0.52577263\n",
      "batch=900 accuracy=0.78114593 loss=0.6172128\n",
      "batch=901 accuracy=0.78125 loss=0.2416406\n",
      "batch=902 accuracy=0.78128463 loss=0.5410814\n",
      "batch=903 accuracy=0.78131914 loss=0.6203221\n",
      "batch=904 accuracy=0.7812845 loss=0.5515091\n",
      "batch=905 accuracy=0.78131896 loss=0.4155161\n",
      "batch=906 accuracy=0.78135335 loss=0.4801494\n",
      "batch=907 accuracy=0.7814565 loss=0.37018013\n",
      "batch=908 accuracy=0.7814563 loss=0.4782865\n",
      "batch=909 accuracy=0.78155905 loss=0.32133135\n",
      "batch=910 accuracy=0.7814558 loss=0.4434911\n",
      "batch=911 accuracy=0.7815241 loss=0.38275832\n",
      "batch=912 accuracy=0.78155804 loss=0.3835143\n",
      "batch=913 accuracy=0.78179705 loss=0.18796423\n",
      "batch=914 accuracy=0.7818989 loss=0.38349432\n",
      "batch=915 accuracy=0.7818982 loss=0.5146781\n",
      "batch=916 accuracy=0.7820679 loss=0.40415972\n",
      "batch=917 accuracy=0.782067 loss=0.59705436\n",
      "batch=918 accuracy=0.7821681 loss=0.36663067\n",
      "batch=919 accuracy=0.782269 loss=0.44603005\n",
      "batch=920 accuracy=0.78233576 loss=0.3584666\n",
      "batch=921 accuracy=0.7823346 loss=0.42542735\n",
      "batch=922 accuracy=0.78253657 loss=0.3415688\n",
      "batch=923 accuracy=0.7824337 loss=0.7729952\n",
      "batch=924 accuracy=0.7825 loss=0.43103376\n",
      "batch=925 accuracy=0.78259987 loss=0.40326896\n",
      "batch=926 accuracy=0.78259844 loss=0.4982165\n",
      "batch=927 accuracy=0.78269804 loss=0.50618625\n",
      "batch=928 accuracy=0.78279734 loss=0.3775464\n",
      "batch=929 accuracy=0.7828293 loss=0.51845104\n",
      "batch=930 accuracy=0.78296185 loss=0.3737256\n",
      "batch=931 accuracy=0.78296006 loss=0.5548259\n",
      "batch=932 accuracy=0.7829917 loss=0.49179083\n",
      "batch=933 accuracy=0.7830902 loss=0.36611116\n",
      "batch=934 accuracy=0.78312165 loss=0.3523996\n",
      "batch=935 accuracy=0.7830863 loss=0.7607386\n",
      "batch=936 accuracy=0.78318435 loss=0.3915456\n",
      "batch=937 accuracy=0.78321564 loss=0.4767061\n",
      "batch=938 accuracy=0.7833799 loss=0.27677155\n",
      "batch=939 accuracy=0.78357714 loss=0.25984168\n",
      "batch=940 accuracy=0.7836743 loss=0.33047605\n",
      "batch=941 accuracy=0.78367174 loss=0.5443013\n",
      "batch=942 accuracy=0.7838017 loss=0.41470683\n",
      "batch=943 accuracy=0.7838652 loss=0.37202764\n",
      "batch=944 accuracy=0.7837963 loss=0.60644704\n",
      "batch=945 accuracy=0.78385967 loss=0.41202164\n",
      "batch=946 accuracy=0.7837579 loss=0.7953602\n",
      "batch=947 accuracy=0.78378826 loss=0.57292104\n",
      "batch=948 accuracy=0.7837526 loss=0.47894722\n",
      "batch=949 accuracy=0.7838158 loss=0.522566\n",
      "batch=950 accuracy=0.7838788 loss=0.66201603\n",
      "batch=951 accuracy=0.7839745 loss=0.4089477\n",
      "batch=952 accuracy=0.7839061 loss=0.45904595\n",
      "batch=953 accuracy=0.7839361 loss=0.5326108\n",
      "batch=954 accuracy=0.7840641 loss=0.3605125\n",
      "batch=955 accuracy=0.7841266 loss=0.44814897\n",
      "batch=956 accuracy=0.78409094 loss=0.7036568\n",
      "batch=957 accuracy=0.78408796 loss=0.7427351\n",
      "batch=958 accuracy=0.7841827 loss=0.62053967\n",
      "batch=959 accuracy=0.7841797 loss=0.60608834\n",
      "batch=960 accuracy=0.7842417 loss=0.3902545\n",
      "batch=961 accuracy=0.7842386 loss=0.35755354\n",
      "batch=962 accuracy=0.7843004 loss=0.45807084\n",
      "batch=963 accuracy=0.78439444 loss=0.35110903\n",
      "batch=964 accuracy=0.7844236 loss=0.50271565\n",
      "batch=965 accuracy=0.784485 loss=0.55157244\n",
      "batch=966 accuracy=0.78444934 loss=0.5641378\n",
      "batch=967 accuracy=0.78454286 loss=0.39638963\n",
      "batch=968 accuracy=0.78460395 loss=0.39659527\n",
      "batch=969 accuracy=0.7846649 loss=0.3653965\n",
      "batch=970 accuracy=0.7846936 loss=0.7686379\n",
      "batch=971 accuracy=0.78481865 loss=0.3049723\n",
      "batch=972 accuracy=0.784815 loss=0.5155753\n",
      "batch=973 accuracy=0.78484344 loss=0.45145622\n",
      "batch=974 accuracy=0.78496796 loss=0.45026913\n",
      "batch=975 accuracy=0.7849001 loss=0.55102444\n",
      "batch=976 accuracy=0.7849603 loss=0.4779538\n",
      "batch=977 accuracy=0.78508437 loss=0.33060557\n",
      "batch=978 accuracy=0.78527194 loss=0.28540456\n",
      "batch=979 accuracy=0.7853635 loss=0.3429165\n",
      "batch=980 accuracy=0.7852956 loss=0.65064764\n",
      "batch=981 accuracy=0.785387 loss=0.5639\n",
      "batch=982 accuracy=0.7854781 loss=0.26531404\n",
      "batch=983 accuracy=0.78553736 loss=0.46220878\n",
      "batch=984 accuracy=0.78550124 loss=0.5839808\n",
      "batch=985 accuracy=0.78552866 loss=0.47034317\n",
      "batch=986 accuracy=0.7856193 loss=0.36434263\n",
      "batch=987 accuracy=0.7856149 loss=0.7548756\n",
      "batch=988 accuracy=0.78557885 loss=0.7567711\n",
      "batch=989 accuracy=0.7856692 loss=0.3207168\n",
      "batch=990 accuracy=0.78579086 loss=0.24777389\n",
      "batch=991 accuracy=0.7859123 loss=0.36591065\n",
      "batch=992 accuracy=0.78597057 loss=0.39030755\n",
      "batch=993 accuracy=0.7860287 loss=0.431808\n",
      "batch=994 accuracy=0.78602386 loss=0.58659416\n",
      "batch=995 accuracy=0.78605044 loss=0.5474669\n",
      "batch=996 accuracy=0.786171 loss=0.3188843\n",
      "batch=997 accuracy=0.78632265 loss=0.3188252\n",
      "batch=998 accuracy=0.7864114 loss=0.30383182\n",
      "batch=999 accuracy=0.7864063 loss=0.7489281\n",
      "batch=1000 accuracy=0.78646356 loss=0.46949288\n",
      "batch=1001 accuracy=0.7865519 loss=0.30011243\n",
      "batch=1002 accuracy=0.7865466 loss=0.579559\n",
      "batch=1003 accuracy=0.78654134 loss=0.44637293\n",
      "batch=1004 accuracy=0.78656715 loss=0.43022865\n",
      "batch=1005 accuracy=0.7864687 loss=0.740939\n",
      "batch=1006 accuracy=0.7864635 loss=0.63655853\n",
      "batch=1007 accuracy=0.7864583 loss=0.48810464\n",
      "batch=1008 accuracy=0.7864532 loss=0.670396\n",
      "batch=1009 accuracy=0.7865099 loss=0.43838027\n",
      "batch=1010 accuracy=0.7866902 loss=0.24646106\n",
      "batch=1011 accuracy=0.78677744 loss=0.4818818\n",
      "batch=1012 accuracy=0.78683364 loss=0.3813753\n",
      "batch=1013 accuracy=0.7869514 loss=0.28860044\n",
      "batch=1014 accuracy=0.78706896 loss=0.35085636\n",
      "batch=1015 accuracy=0.78718626 loss=0.31545094\n",
      "batch=1016 accuracy=0.7873341 loss=0.26647195\n",
      "batch=1017 accuracy=0.7873895 loss=0.47592956\n",
      "batch=1018 accuracy=0.78741413 loss=0.5123122\n",
      "batch=1019 accuracy=0.78753066 loss=0.3497399\n",
      "batch=1020 accuracy=0.7876469 loss=0.29105756\n",
      "batch=1021 accuracy=0.78773236 loss=0.43133867\n",
      "batch=1022 accuracy=0.7876955 loss=0.48231044\n",
      "batch=1023 accuracy=0.7878113 loss=0.23955512\n",
      "batch=1024 accuracy=0.7878659 loss=0.40922737\n",
      "batch=1025 accuracy=0.78779846 loss=0.76035863\n",
      "batch=1026 accuracy=0.78770083 loss=0.7574031\n",
      "batch=1027 accuracy=0.7876946 loss=0.53310895\n",
      "batch=1028 accuracy=0.7878401 loss=0.16990066\n",
      "batch=1029 accuracy=0.7878944 loss=0.54522973\n",
      "batch=1030 accuracy=0.7880092 loss=0.2803248\n",
      "batch=1031 accuracy=0.7879724 loss=0.51469743\n",
      "batch=1032 accuracy=0.7879356 loss=0.49422225\n",
      "batch=1033 accuracy=0.78801984 loss=0.4738158\n",
      "batch=1034 accuracy=0.78807366 loss=0.51394695\n",
      "batch=1035 accuracy=0.7881274 loss=0.35537922\n",
      "batch=1036 accuracy=0.78818107 loss=0.44001886\n",
      "batch=1037 accuracy=0.7882948 loss=0.4387657\n",
      "batch=1038 accuracy=0.7883181 loss=0.36668026\n",
      "batch=1039 accuracy=0.7883714 loss=0.39534125\n",
      "batch=1040 accuracy=0.7883946 loss=0.4987946\n",
      "batch=1041 accuracy=0.7884477 loss=0.45093358\n",
      "batch=1042 accuracy=0.7885307 loss=0.42526138\n",
      "batch=1043 accuracy=0.7886434 loss=0.22460894\n",
      "batch=1044 accuracy=0.78869617 loss=0.28773198\n",
      "batch=1045 accuracy=0.78877866 loss=0.38872296\n",
      "batch=1046 accuracy=0.7888312 loss=0.5999699\n",
      "batch=1047 accuracy=0.7889134 loss=0.3280227\n",
      "batch=1048 accuracy=0.78899544 loss=0.38565117\n",
      "batch=1049 accuracy=0.7889881 loss=0.7643194\n",
      "batch=1050 accuracy=0.7891591 loss=0.2343004\n",
      "batch=1051 accuracy=0.7892407 loss=0.3868642\n",
      "batch=1052 accuracy=0.78926283 loss=0.3357113\n",
      "batch=1053 accuracy=0.7891663 loss=0.64095646\n",
      "batch=1054 accuracy=0.7891884 loss=0.56512403\n",
      "batch=1055 accuracy=0.7891809 loss=0.44639033\n",
      "batch=1056 accuracy=0.78926206 loss=0.4154448\n",
      "batch=1057 accuracy=0.78931355 loss=0.41989344\n",
      "batch=1058 accuracy=0.7893059 loss=0.4756654\n",
      "batch=1059 accuracy=0.7893573 loss=0.45245013\n",
      "batch=1060 accuracy=0.789438 loss=0.32168967\n",
      "batch=1061 accuracy=0.78954804 loss=0.2933665\n",
      "batch=1062 accuracy=0.7895696 loss=0.6078203\n",
      "batch=1063 accuracy=0.7896205 loss=0.53763807\n",
      "batch=1064 accuracy=0.78961265 loss=0.453393\n",
      "batch=1065 accuracy=0.78960484 loss=0.60504395\n",
      "batch=1066 accuracy=0.789597 loss=0.6869345\n",
      "batch=1067 accuracy=0.7896477 loss=0.53677046\n",
      "batch=1068 accuracy=0.7896691 loss=0.4089837\n",
      "batch=1069 accuracy=0.78974885 loss=0.30191147\n",
      "batch=1070 accuracy=0.7898284 loss=0.51975214\n",
      "batch=1071 accuracy=0.78982043 loss=0.5645814\n",
      "batch=1072 accuracy=0.7899872 loss=0.31573212\n",
      "batch=1073 accuracy=0.7900081 loss=0.4771042\n",
      "batch=1074 accuracy=0.79002905 loss=0.3794558\n",
      "batch=1075 accuracy=0.7900209 loss=0.3595647\n",
      "batch=1076 accuracy=0.7900708 loss=0.50097483\n",
      "batch=1077 accuracy=0.79003364 loss=0.535843\n",
      "batch=1078 accuracy=0.7901124 loss=0.41213387\n",
      "batch=1079 accuracy=0.7899884 loss=0.56745076\n",
      "batch=1080 accuracy=0.78995144 loss=0.48940235\n",
      "batch=1081 accuracy=0.78985673 loss=0.68777585\n",
      "batch=1082 accuracy=0.7899642 loss=0.34005162\n",
      "batch=1083 accuracy=0.79001385 loss=0.42354366\n",
      "batch=1084 accuracy=0.78994817 loss=0.8123521\n",
      "batch=1085 accuracy=0.7900265 loss=0.45501822\n",
      "batch=1086 accuracy=0.7901046 loss=0.6742346\n",
      "batch=1087 accuracy=0.79003906 loss=0.6759984\n",
      "batch=1088 accuracy=0.79008836 loss=0.36345276\n",
      "batch=1089 accuracy=0.79016626 loss=0.38561684\n",
      "batch=1090 accuracy=0.7903013 loss=0.2715466\n",
      "batch=1091 accuracy=0.79037887 loss=0.2818038\n",
      "batch=1092 accuracy=0.79042774 loss=0.39255714\n",
      "batch=1093 accuracy=0.7904479 loss=0.5346189\n",
      "batch=1094 accuracy=0.7904966 loss=0.48903888\n",
      "batch=1095 accuracy=0.7906022 loss=0.3725322\n",
      "batch=1096 accuracy=0.7905652 loss=0.60041344\n",
      "batch=1097 accuracy=0.79058516 loss=0.40330216\n",
      "batch=1098 accuracy=0.79063356 loss=0.6395843\n",
      "batch=1099 accuracy=0.7906534 loss=0.5589367\n",
      "batch=1100 accuracy=0.7907016 loss=0.60974926\n",
      "batch=1101 accuracy=0.7907498 loss=0.5182562\n",
      "batch=1102 accuracy=0.7908545 loss=0.40784392\n",
      "batch=1103 accuracy=0.790959 loss=0.48258147\n",
      "batch=1104 accuracy=0.7909785 loss=0.42218304\n",
      "batch=1105 accuracy=0.7910545 loss=0.26575783\n",
      "batch=1106 accuracy=0.79110205 loss=0.48920095\n",
      "batch=1107 accuracy=0.7911496 loss=0.43429965\n",
      "batch=1108 accuracy=0.7911407 loss=0.5843472\n",
      "batch=1109 accuracy=0.7911599 loss=0.35563266\n",
      "batch=1110 accuracy=0.7910947 loss=0.5867216\n",
      "batch=1111 accuracy=0.7911702 loss=0.44972372\n",
      "batch=1112 accuracy=0.7911332 loss=0.59645915\n",
      "batch=1113 accuracy=0.79112434 loss=0.5287746\n",
      "batch=1114 accuracy=0.7911435 loss=0.6232166\n",
      "batch=1115 accuracy=0.7910786 loss=0.55880743\n",
      "batch=1116 accuracy=0.7911538 loss=0.3281563\n",
      "batch=1117 accuracy=0.79117286 loss=0.5863794\n",
      "batch=1118 accuracy=0.7912478 loss=0.29920217\n",
      "batch=1119 accuracy=0.79126674 loss=0.555071\n",
      "batch=1120 accuracy=0.7913414 loss=0.37523085\n",
      "batch=1121 accuracy=0.7913603 loss=0.58074987\n",
      "batch=1122 accuracy=0.7913791 loss=0.57632875\n",
      "batch=1123 accuracy=0.7913979 loss=0.76050186\n",
      "batch=1124 accuracy=0.79141665 loss=0.38490313\n",
      "batch=1125 accuracy=0.79137987 loss=0.6302941\n",
      "batch=1126 accuracy=0.79142636 loss=0.32213137\n",
      "batch=1127 accuracy=0.791445 loss=0.6006824\n",
      "batch=1128 accuracy=0.79151905 loss=0.45998958\n",
      "batch=1129 accuracy=0.7915099 loss=0.5518044\n",
      "batch=1130 accuracy=0.7915561 loss=0.6675807\n",
      "batch=1131 accuracy=0.79160225 loss=0.24992949\n",
      "batch=1132 accuracy=0.79170346 loss=0.3073368\n",
      "batch=1133 accuracy=0.7916942 loss=0.43741196\n",
      "batch=1134 accuracy=0.79174006 loss=0.44120535\n",
      "batch=1135 accuracy=0.79175836 loss=0.53999853\n",
      "batch=1136 accuracy=0.79183155 loss=0.38308674\n",
      "batch=1137 accuracy=0.79176736 loss=0.58777297\n",
      "batch=1138 accuracy=0.791813 loss=0.41185868\n",
      "batch=1139 accuracy=0.7918037 loss=0.5930394\n",
      "batch=1140 accuracy=0.7918766 loss=0.35426962\n",
      "batch=1141 accuracy=0.79194945 loss=0.43835664\n",
      "batch=1142 accuracy=0.79199475 loss=0.40584886\n",
      "batch=1143 accuracy=0.7920673 loss=0.27894467\n",
      "batch=1144 accuracy=0.792167 loss=0.45009416\n",
      "batch=1145 accuracy=0.79229385 loss=0.34240997\n",
      "batch=1146 accuracy=0.7923387 loss=0.38419142\n",
      "batch=1147 accuracy=0.79235625 loss=0.3740866\n",
      "batch=1148 accuracy=0.792401 loss=0.3426227\n",
      "batch=1149 accuracy=0.7923913 loss=0.6322302\n",
      "batch=1150 accuracy=0.79240876 loss=0.5606906\n",
      "batch=1151 accuracy=0.79245335 loss=0.32595056\n",
      "batch=1152 accuracy=0.7924978 loss=0.5849848\n",
      "batch=1153 accuracy=0.7925422 loss=0.33902502\n",
      "batch=1154 accuracy=0.79253244 loss=0.54432213\n",
      "batch=1155 accuracy=0.7925768 loss=0.28227323\n",
      "batch=1156 accuracy=0.792594 loss=0.36621606\n",
      "batch=1157 accuracy=0.7926112 loss=0.59422815\n",
      "batch=1158 accuracy=0.7926823 loss=0.34421182\n",
      "batch=1159 accuracy=0.79278016 loss=0.2313421\n",
      "batch=1160 accuracy=0.792851 loss=0.31398693\n",
      "batch=1161 accuracy=0.7929486 loss=0.28279024\n",
      "batch=1162 accuracy=0.793046 loss=0.24372816\n",
      "batch=1163 accuracy=0.79303586 loss=0.78381515\n",
      "batch=1164 accuracy=0.7929989 loss=0.60717624\n",
      "batch=1165 accuracy=0.79298884 loss=0.4718409\n",
      "batch=1166 accuracy=0.79297876 loss=0.58961904\n",
      "batch=1167 accuracy=0.792942 loss=0.6137738\n",
      "batch=1168 accuracy=0.7930122 loss=0.37736407\n",
      "batch=1169 accuracy=0.79302883 loss=0.3415954\n",
      "batch=1170 accuracy=0.7931522 loss=0.27576637\n",
      "batch=1171 accuracy=0.79308873 loss=0.7591373\n",
      "batch=1172 accuracy=0.79307866 loss=0.5987114\n",
      "batch=1173 accuracy=0.7930952 loss=0.3672815\n",
      "batch=1174 accuracy=0.7931649 loss=0.5202186\n",
      "batch=1175 accuracy=0.7931548 loss=0.6824366\n",
      "batch=1176 accuracy=0.7932243 loss=0.38306457\n",
      "batch=1177 accuracy=0.7932937 loss=0.29445076\n",
      "batch=1178 accuracy=0.7933895 loss=0.22949547\n",
      "batch=1179 accuracy=0.7934587 loss=0.41745073\n",
      "batch=1180 accuracy=0.79342186 loss=0.58258146\n",
      "batch=1181 accuracy=0.7934909 loss=0.29961646\n",
      "batch=1182 accuracy=0.7935334 loss=0.6316304\n",
      "batch=1183 accuracy=0.7936286 loss=0.31121355\n",
      "batch=1184 accuracy=0.7936445 loss=0.44374937\n",
      "batch=1185 accuracy=0.7936077 loss=0.56305754\n",
      "batch=1186 accuracy=0.79367626 loss=0.38846758\n",
      "batch=1187 accuracy=0.7936132 loss=0.81526816\n",
      "batch=1188 accuracy=0.7936817 loss=0.40084746\n",
      "batch=1189 accuracy=0.79364496 loss=0.73331845\n",
      "batch=1190 accuracy=0.7937133 loss=0.47864562\n",
      "batch=1191 accuracy=0.7938077 loss=0.2866725\n",
      "batch=1192 accuracy=0.79382336 loss=0.59730685\n",
      "batch=1193 accuracy=0.7937343 loss=0.7025442\n",
      "batch=1194 accuracy=0.79377615 loss=0.4531024\n",
      "batch=1195 accuracy=0.79379183 loss=0.6691172\n",
      "batch=1196 accuracy=0.79385966 loss=0.25028536\n",
      "batch=1197 accuracy=0.79390126 loss=0.48341337\n",
      "batch=1198 accuracy=0.7938907 loss=0.57235384\n",
      "batch=1199 accuracy=0.7938542 loss=0.54832697\n",
      "batch=1200 accuracy=0.7938437 loss=0.45887718\n",
      "batch=1201 accuracy=0.7938852 loss=0.45117915\n",
      "batch=1202 accuracy=0.7939786 loss=0.28689134\n",
      "batch=1203 accuracy=0.793968 loss=0.44824058\n",
      "batch=1204 accuracy=0.79403526 loss=0.4511096\n",
      "batch=1205 accuracy=0.79410243 loss=0.4687755\n",
      "batch=1206 accuracy=0.79403996 loss=0.78481853\n",
      "batch=1207 accuracy=0.7940553 loss=0.681629\n",
      "batch=1208 accuracy=0.7941739 loss=0.27668458\n",
      "batch=1209 accuracy=0.7942407 loss=0.393457\n",
      "batch=1210 accuracy=0.7943332 loss=0.36682612\n",
      "batch=1211 accuracy=0.794374 loss=0.45059183\n",
      "batch=1212 accuracy=0.7944662 loss=0.38878727\n",
      "batch=1213 accuracy=0.7945068 loss=0.50237465\n",
      "batch=1214 accuracy=0.79457307 loss=0.38783002\n",
      "batch=1215 accuracy=0.79463917 loss=0.342082\n",
      "batch=1216 accuracy=0.7946795 loss=0.6265546\n",
      "batch=1217 accuracy=0.79464287 loss=0.5591341\n",
      "batch=1218 accuracy=0.7946319 loss=0.7444992\n",
      "batch=1219 accuracy=0.79472333 loss=0.3326035\n",
      "batch=1220 accuracy=0.7947123 loss=0.3979624\n",
      "batch=1221 accuracy=0.79470134 loss=0.7245498\n",
      "batch=1222 accuracy=0.7946392 loss=0.7929858\n",
      "batch=1223 accuracy=0.7946538 loss=0.5251124\n",
      "batch=1224 accuracy=0.7946684 loss=0.47394532\n",
      "batch=1225 accuracy=0.79470843 loss=0.33850205\n",
      "batch=1226 accuracy=0.7947993 loss=0.31473905\n",
      "batch=1227 accuracy=0.79476285 loss=0.6162342\n",
      "batch=1228 accuracy=0.7948281 loss=0.38705766\n",
      "batch=1229 accuracy=0.7948171 loss=0.5873829\n",
      "batch=1230 accuracy=0.79485685 loss=0.29955494\n",
      "batch=1231 accuracy=0.7948965 loss=0.4635659\n",
      "batch=1232 accuracy=0.79483473 loss=0.74932146\n",
      "batch=1233 accuracy=0.79479843 loss=0.5306479\n",
      "batch=1234 accuracy=0.7948887 loss=0.25380462\n",
      "batch=1235 accuracy=0.79487765 loss=0.40107352\n",
      "batch=1236 accuracy=0.7948666 loss=0.4659003\n",
      "batch=1237 accuracy=0.79498184 loss=0.34021464\n",
      "batch=1238 accuracy=0.7949203 loss=0.59487617\n",
      "batch=1239 accuracy=0.79493445 loss=0.48187578\n",
      "batch=1240 accuracy=0.79502416 loss=0.28445607\n",
      "batch=1241 accuracy=0.7950886 loss=0.32390743\n",
      "batch=1242 accuracy=0.79507744 loss=0.61506444\n",
      "batch=1243 accuracy=0.7950412 loss=0.73129\n",
      "batch=1244 accuracy=0.79513055 loss=0.3159324\n",
      "batch=1245 accuracy=0.7951194 loss=0.45797342\n",
      "batch=1246 accuracy=0.7951834 loss=0.3873847\n",
      "batch=1247 accuracy=0.7951723 loss=0.4903322\n",
      "batch=1248 accuracy=0.7952112 loss=0.335043\n",
      "batch=1249 accuracy=0.7953 loss=0.39398772\n",
      "batch=1250 accuracy=0.7952388 loss=0.559454\n",
      "batch=1251 accuracy=0.7951528 loss=0.78588647\n",
      "batch=1252 accuracy=0.7951915 loss=0.39456332\n",
      "batch=1253 accuracy=0.7952552 loss=0.44881374\n",
      "batch=1254 accuracy=0.7952938 loss=0.3964963\n",
      "batch=1255 accuracy=0.7954319 loss=0.16857323\n",
      "batch=1256 accuracy=0.7954952 loss=0.5102277\n",
      "batch=1257 accuracy=0.7953597 loss=0.5626456\n",
      "batch=1258 accuracy=0.7953733 loss=0.6122554\n",
      "batch=1259 accuracy=0.7953621 loss=0.6350367\n",
      "batch=1260 accuracy=0.79542524 loss=0.5007412\n",
      "batch=1261 accuracy=0.79533976 loss=0.64624393\n",
      "batch=1262 accuracy=0.79532856 loss=0.62743264\n",
      "batch=1263 accuracy=0.7953916 loss=0.3858174\n",
      "batch=1264 accuracy=0.79552865 loss=0.25914207\n",
      "batch=1265 accuracy=0.7956408 loss=0.24444798\n",
      "batch=1266 accuracy=0.7957034 loss=0.23364948\n",
      "batch=1267 accuracy=0.7956674 loss=0.63225496\n",
      "batch=1268 accuracy=0.79568064 loss=0.49376595\n",
      "batch=1269 accuracy=0.79579234 loss=0.22173549\n",
      "batch=1270 accuracy=0.7958546 loss=0.39303243\n",
      "batch=1271 accuracy=0.7958923 loss=0.6157434\n",
      "batch=1272 accuracy=0.7959299 loss=0.60107696\n",
      "batch=1273 accuracy=0.7959674 loss=0.44206795\n",
      "batch=1274 accuracy=0.79585785 loss=0.69895566\n",
      "batch=1275 accuracy=0.7959933 loss=0.25434387\n",
      "batch=1276 accuracy=0.79600626 loss=0.50892305\n",
      "batch=1277 accuracy=0.79597026 loss=0.5553378\n",
      "batch=1278 accuracy=0.7960321 loss=0.40148497\n",
      "batch=1279 accuracy=0.79604495 loss=0.46006596\n",
      "batch=1280 accuracy=0.79598457 loss=0.64727134\n",
      "batch=1281 accuracy=0.79599744 loss=0.4778569\n",
      "batch=1282 accuracy=0.7961321 loss=0.19684193\n",
      "batch=1283 accuracy=0.7961692 loss=0.5354811\n",
      "batch=1284 accuracy=0.79610896 loss=0.7132437\n",
      "batch=1285 accuracy=0.796146 loss=0.37769246\n",
      "batch=1286 accuracy=0.79615873 loss=0.61749953\n",
      "batch=1287 accuracy=0.7961471 loss=0.6560439\n",
      "batch=1288 accuracy=0.7962083 loss=0.28138885\n",
      "batch=1289 accuracy=0.7961725 loss=0.49627507\n",
      "batch=1290 accuracy=0.79623353 loss=0.32227412\n",
      "batch=1291 accuracy=0.796222 loss=0.5738671\n",
      "batch=1292 accuracy=0.7962587 loss=0.44570774\n",
      "batch=1293 accuracy=0.7962954 loss=0.34005445\n",
      "batch=1294 accuracy=0.7962838 loss=0.51137865\n",
      "batch=1295 accuracy=0.7962481 loss=0.593727\n",
      "batch=1296 accuracy=0.79621243 loss=0.49577427\n",
      "batch=1297 accuracy=0.7962972 loss=0.22165957\n",
      "batch=1298 accuracy=0.79635775 loss=0.32919297\n",
      "batch=1299 accuracy=0.79651445 loss=0.15011442\n",
      "batch=1300 accuracy=0.7965748 loss=0.3770038\n",
      "batch=1301 accuracy=0.796515 loss=0.72471285\n",
      "batch=1302 accuracy=0.7966232 loss=0.3045622\n",
      "batch=1303 accuracy=0.79656345 loss=0.56067806\n",
      "batch=1304 accuracy=0.7966236 loss=0.33040303\n",
      "batch=1305 accuracy=0.79654 loss=0.5851499\n",
      "batch=1306 accuracy=0.79664785 loss=0.36440003\n",
      "batch=1307 accuracy=0.79670775 loss=0.42464083\n",
      "batch=1308 accuracy=0.7967676 loss=0.354991\n",
      "batch=1309 accuracy=0.7968273 loss=0.3542987\n",
      "batch=1310 accuracy=0.7969346 loss=0.290433\n",
      "batch=1311 accuracy=0.7970417 loss=0.22753865\n",
      "batch=1312 accuracy=0.7970535 loss=0.35445246\n",
      "batch=1313 accuracy=0.79708904 loss=0.3625906\n",
      "batch=1314 accuracy=0.7971245 loss=0.38092026\n",
      "batch=1315 accuracy=0.79715997 loss=0.46268868\n",
      "batch=1316 accuracy=0.79721904 loss=0.41379768\n",
      "batch=1317 accuracy=0.79720694 loss=0.5193691\n",
      "batch=1318 accuracy=0.79721856 loss=0.5207491\n",
      "batch=1319 accuracy=0.79727745 loss=0.22753528\n",
      "batch=1320 accuracy=0.7973126 loss=0.5020553\n",
      "batch=1321 accuracy=0.7973714 loss=0.30750233\n",
      "batch=1322 accuracy=0.7973356 loss=0.81007594\n",
      "batch=1323 accuracy=0.79734707 loss=0.47555453\n",
      "batch=1324 accuracy=0.79738206 loss=0.3912999\n",
      "batch=1325 accuracy=0.7974406 loss=0.30512124\n",
      "batch=1326 accuracy=0.7975226 loss=0.2957272\n",
      "batch=1327 accuracy=0.7976045 loss=0.29877093\n",
      "batch=1328 accuracy=0.7975687 loss=0.51902103\n",
      "batch=1329 accuracy=0.7975799 loss=0.6223163\n",
      "batch=1330 accuracy=0.7976146 loss=0.48715502\n",
      "batch=1331 accuracy=0.7976023 loss=0.7661647\n",
      "batch=1332 accuracy=0.79766035 loss=0.25544462\n",
      "batch=1333 accuracy=0.7977418 loss=0.24793331\n",
      "batch=1334 accuracy=0.7977762 loss=0.52885896\n",
      "batch=1335 accuracy=0.7978808 loss=0.25756904\n",
      "batch=1336 accuracy=0.79789174 loss=0.6047607\n",
      "batch=1337 accuracy=0.7979961 loss=0.42442387\n",
      "batch=1338 accuracy=0.7980069 loss=0.49434388\n",
      "batch=1339 accuracy=0.7980877 loss=0.26690632\n",
      "batch=1340 accuracy=0.79812175 loss=0.53772295\n",
      "batch=1341 accuracy=0.79820234 loss=0.2519739\n",
      "batch=1342 accuracy=0.79830605 loss=0.23309529\n",
      "batch=1343 accuracy=0.7983166 loss=0.46524215\n",
      "batch=1344 accuracy=0.7982574 loss=0.7083755\n",
      "batch=1345 accuracy=0.7982216 loss=0.4600305\n",
      "batch=1346 accuracy=0.798209 loss=0.7339053\n",
      "batch=1347 accuracy=0.79826593 loss=0.44858986\n",
      "batch=1348 accuracy=0.7982765 loss=0.41189784\n",
      "batch=1349 accuracy=0.79831016 loss=0.48931664\n",
      "batch=1350 accuracy=0.7982976 loss=0.3737176\n",
      "batch=1351 accuracy=0.7983081 loss=0.5200914\n",
      "batch=1352 accuracy=0.7982724 loss=0.57309884\n",
      "batch=1353 accuracy=0.79828286 loss=0.47087234\n",
      "batch=1354 accuracy=0.7983164 loss=0.30204606\n",
      "batch=1355 accuracy=0.7982808 loss=0.61810803\n",
      "batch=1356 accuracy=0.79836035 loss=0.39075235\n",
      "batch=1357 accuracy=0.7984398 loss=0.25884682\n",
      "batch=1358 accuracy=0.79856515 loss=0.12102435\n",
      "batch=1359 accuracy=0.7986213 loss=0.26218775\n",
      "batch=1360 accuracy=0.79860854 loss=0.5067415\n",
      "batch=1361 accuracy=0.7985958 loss=0.56993365\n",
      "batch=1362 accuracy=0.79860604 loss=0.46775243\n",
      "batch=1363 accuracy=0.79870784 loss=0.292833\n",
      "batch=1364 accuracy=0.79874086 loss=0.5118293\n",
      "batch=1365 accuracy=0.79875094 loss=0.53245676\n",
      "batch=1366 accuracy=0.79882956 loss=0.29159677\n",
      "batch=1367 accuracy=0.79893094 loss=0.23635301\n",
      "batch=1368 accuracy=0.79894084 loss=0.512187\n",
      "batch=1369 accuracy=0.79895073 loss=0.44162938\n",
      "batch=1370 accuracy=0.7989378 loss=0.48966318\n",
      "batch=1371 accuracy=0.79897046 loss=0.2752717\n",
      "batch=1372 accuracy=0.7990031 loss=0.38895166\n",
      "batch=1373 accuracy=0.79903567 loss=0.3687206\n",
      "batch=1374 accuracy=0.79893184 loss=0.58944476\n",
      "batch=1375 accuracy=0.7989871 loss=0.36694986\n",
      "batch=1376 accuracy=0.799065 loss=0.49069795\n",
      "batch=1377 accuracy=0.7990974 loss=0.57932913\n",
      "batch=1378 accuracy=0.7991298 loss=0.4352318\n",
      "batch=1379 accuracy=0.79911685 loss=0.5338078\n",
      "batch=1380 accuracy=0.79908127 loss=0.6618091\n",
      "batch=1381 accuracy=0.79911363 loss=0.3943165\n",
      "batch=1382 accuracy=0.7991459 loss=0.36425585\n",
      "batch=1383 accuracy=0.7992007 loss=0.35017884\n",
      "batch=1384 accuracy=0.79923284 loss=0.2931079\n",
      "batch=1385 accuracy=0.7992875 loss=0.43321633\n",
      "batch=1386 accuracy=0.7993421 loss=0.30270976\n",
      "batch=1387 accuracy=0.79946417 loss=0.2555464\n",
      "batch=1388 accuracy=0.7994285 loss=0.6512058\n",
      "batch=1389 accuracy=0.799393 loss=0.573425\n",
      "batch=1390 accuracy=0.7995147 loss=0.20251402\n",
      "batch=1391 accuracy=0.79954654 loss=0.5813017\n",
      "batch=1392 accuracy=0.79962313 loss=0.35843503\n",
      "batch=1393 accuracy=0.7996772 loss=0.41215277\n",
      "batch=1394 accuracy=0.7997312 loss=0.31364217\n",
      "batch=1395 accuracy=0.79971796 loss=0.3909721\n",
      "batch=1396 accuracy=0.79977185 loss=0.31957045\n",
      "batch=1397 accuracy=0.79982567 loss=0.30899623\n",
      "batch=1398 accuracy=0.799857 loss=0.43096265\n",
      "batch=1399 accuracy=0.7999107 loss=0.35552853\n",
      "batch=1400 accuracy=0.7999643 loss=0.37132874\n",
      "batch=1401 accuracy=0.7998618 loss=0.83947766\n",
      "batch=1402 accuracy=0.7999376 loss=0.24970296\n",
      "batch=1403 accuracy=0.79996884 loss=0.5206025\n",
      "batch=1404 accuracy=0.80002224 loss=0.29389015\n",
      "batch=1405 accuracy=0.8000978 loss=0.26378185\n",
      "batch=1406 accuracy=0.8001066 loss=0.5141645\n",
      "batch=1407 accuracy=0.80009323 loss=0.5926416\n",
      "batch=1408 accuracy=0.8001464 loss=0.3606512\n",
      "batch=1409 accuracy=0.8002438 loss=0.22290878\n",
      "batch=1410 accuracy=0.8002082 loss=0.6056407\n",
      "batch=1411 accuracy=0.800239 loss=0.3695277\n",
      "batch=1412 accuracy=0.80020344 loss=0.70531845\n",
      "batch=1413 accuracy=0.8002785 loss=0.2749765\n",
      "batch=1414 accuracy=0.8003092 loss=0.38467485\n",
      "batch=1415 accuracy=0.800384 loss=0.3443051\n",
      "batch=1416 accuracy=0.8004587 loss=0.26978135\n",
      "batch=1417 accuracy=0.80048925 loss=0.39425278\n",
      "batch=1418 accuracy=0.8004977 loss=0.37491623\n",
      "batch=1419 accuracy=0.80055016 loss=0.3944319\n",
      "batch=1420 accuracy=0.80060256 loss=0.44364136\n",
      "batch=1421 accuracy=0.80061096 loss=0.50742555\n",
      "batch=1422 accuracy=0.8005534 loss=0.53272134\n",
      "batch=1423 accuracy=0.80062765 loss=0.3386042\n",
      "batch=1424 accuracy=0.8005921 loss=0.5897212\n",
      "batch=1425 accuracy=0.8006662 loss=0.2571422\n",
      "batch=1426 accuracy=0.8006307 loss=0.51346284\n",
      "batch=1427 accuracy=0.80063903 loss=0.47827715\n",
      "batch=1428 accuracy=0.8006692 loss=0.38663727\n",
      "batch=1429 accuracy=0.8006556 loss=0.6779152\n",
      "batch=1430 accuracy=0.800642 loss=0.46269453\n",
      "batch=1431 accuracy=0.8007376 loss=0.31166118\n",
      "batch=1432 accuracy=0.80083305 loss=0.19608507\n",
      "batch=1433 accuracy=0.80090654 loss=0.23759688\n",
      "batch=1434 accuracy=0.80089283 loss=0.43486065\n",
      "batch=1435 accuracy=0.8009227 loss=0.40494838\n",
      "batch=1436 accuracy=0.80097425 loss=0.32219583\n",
      "batch=1437 accuracy=0.80104744 loss=0.35082936\n",
      "batch=1438 accuracy=0.8010989 loss=0.38683945\n",
      "batch=1439 accuracy=0.80108505 loss=0.51966727\n",
      "batch=1440 accuracy=0.801093 loss=0.6661702\n",
      "batch=1441 accuracy=0.80112255 loss=0.42342654\n",
      "batch=1442 accuracy=0.8011521 loss=0.47283804\n",
      "batch=1443 accuracy=0.80116 loss=0.5626024\n",
      "batch=1444 accuracy=0.8011894 loss=0.29368967\n",
      "batch=1445 accuracy=0.80121887 loss=0.56135446\n",
      "batch=1446 accuracy=0.80124825 loss=0.37976843\n",
      "batch=1447 accuracy=0.8012992 loss=0.34134048\n",
      "batch=1448 accuracy=0.80135006 loss=0.3586639\n",
      "batch=1449 accuracy=0.8012931 loss=0.6335111\n",
      "batch=1450 accuracy=0.8014085 loss=0.17701805\n",
      "batch=1451 accuracy=0.8014377 loss=0.47093928\n",
      "batch=1452 accuracy=0.8014668 loss=0.5386807\n",
      "batch=1453 accuracy=0.8014529 loss=0.54102224\n",
      "batch=1454 accuracy=0.8015034 loss=0.3537767\n",
      "batch=1455 accuracy=0.80157536 loss=0.28124207\n",
      "batch=1456 accuracy=0.80151856 loss=0.57139784\n",
      "batch=1457 accuracy=0.80152607 loss=0.47489208\n",
      "batch=1458 accuracy=0.8015122 loss=0.5474421\n",
      "batch=1459 accuracy=0.8014983 loss=0.519039\n",
      "batch=1460 accuracy=0.8014844 loss=0.47909337\n",
      "batch=1461 accuracy=0.8015561 loss=0.29159176\n",
      "batch=1462 accuracy=0.8016063 loss=0.4035425\n",
      "batch=1463 accuracy=0.8016564 loss=0.40356243\n",
      "batch=1464 accuracy=0.8017065 loss=0.49255973\n",
      "batch=1465 accuracy=0.8017138 loss=0.43252024\n",
      "batch=1466 accuracy=0.8017425 loss=0.38726646\n",
      "batch=1467 accuracy=0.8018137 loss=0.28066602\n",
      "batch=1468 accuracy=0.8018422 loss=0.35226884\n",
      "batch=1469 accuracy=0.801892 loss=0.32439113\n",
      "batch=1470 accuracy=0.80177176 loss=0.7760887\n",
      "batch=1471 accuracy=0.8017578 loss=0.4804202\n",
      "batch=1472 accuracy=0.8017863 loss=0.6810765\n",
      "batch=1473 accuracy=0.8017936 loss=0.53777385\n",
      "batch=1474 accuracy=0.8018856 loss=0.13349771\n",
      "batch=1475 accuracy=0.80185044 loss=0.8066333\n",
      "batch=1476 accuracy=0.80189997 loss=0.36630934\n",
      "batch=1477 accuracy=0.8018437 loss=0.58273065\n",
      "batch=1478 accuracy=0.80180866 loss=0.5228989\n",
      "batch=1479 accuracy=0.80181587 loss=0.44960502\n",
      "batch=1480 accuracy=0.801802 loss=0.5553563\n",
      "batch=1481 accuracy=0.80176705 loss=0.6518042\n",
      "batch=1482 accuracy=0.8017532 loss=0.45424834\n",
      "batch=1483 accuracy=0.8017815 loss=0.35509086\n",
      "batch=1484 accuracy=0.8018729 loss=0.2570034\n",
      "batch=1485 accuracy=0.8019221 loss=0.45053023\n",
      "batch=1486 accuracy=0.80199224 loss=0.42330202\n",
      "batch=1487 accuracy=0.8020203 loss=0.44745988\n",
      "batch=1488 accuracy=0.8021113 loss=0.26752353\n",
      "batch=1489 accuracy=0.8020973 loss=0.59776986\n",
      "batch=1490 accuracy=0.8022091 loss=0.19552764\n",
      "batch=1491 accuracy=0.80215317 loss=0.71727335\n",
      "batch=1492 accuracy=0.8022438 loss=0.22436698\n",
      "batch=1493 accuracy=0.8022716 loss=0.4458769\n",
      "batch=1494 accuracy=0.80232024 loss=0.41805822\n",
      "batch=1495 accuracy=0.8023688 loss=0.3588885\n",
      "batch=1496 accuracy=0.80245906 loss=0.3208477\n",
      "batch=1497 accuracy=0.8025075 loss=0.25955752\n",
      "batch=1498 accuracy=0.8025767 loss=0.25402957\n",
      "batch=1499 accuracy=0.8025 loss=0.52529085\n",
      "batch=1500 accuracy=0.8025691 loss=0.36189127\n",
      "batch=1501 accuracy=0.8025965 loss=0.2802939\n",
      "batch=1502 accuracy=0.8025408 loss=0.7735044\n",
      "batch=1503 accuracy=0.80258894 loss=0.4128647\n",
      "batch=1504 accuracy=0.80257475 loss=0.6256013\n",
      "batch=1505 accuracy=0.8026436 loss=0.4138196\n",
      "batch=1506 accuracy=0.8026709 loss=0.47726336\n",
      "batch=1507 accuracy=0.80265665 loss=0.55674505\n",
      "batch=1508 accuracy=0.8026632 loss=0.43447393\n",
      "batch=1509 accuracy=0.8026697 loss=0.64251626\n",
      "batch=1510 accuracy=0.8026969 loss=0.37105584\n",
      "batch=1511 accuracy=0.80276537 loss=0.21250072\n",
      "batch=1512 accuracy=0.8027718 loss=0.35879916\n",
      "batch=1513 accuracy=0.8028195 loss=0.2345997\n",
      "batch=1514 accuracy=0.8028672 loss=0.26221234\n",
      "batch=1515 accuracy=0.8028735 loss=0.6541563\n",
      "batch=1516 accuracy=0.80285925 loss=0.64974236\n",
      "batch=1517 accuracy=0.802845 loss=0.38249314\n",
      "batch=1518 accuracy=0.8028925 loss=0.44871333\n",
      "batch=1519 accuracy=0.8028988 loss=0.6398486\n",
      "batch=1520 accuracy=0.80294627 loss=0.56562495\n",
      "batch=1521 accuracy=0.8029525 loss=0.43011442\n",
      "batch=1522 accuracy=0.80299985 loss=0.34617165\n",
      "batch=1523 accuracy=0.80296504 loss=0.7978414\n",
      "batch=1524 accuracy=0.8030123 loss=0.41407663\n",
      "batch=1525 accuracy=0.803039 loss=0.42169583\n",
      "batch=1526 accuracy=0.8031066 loss=0.29329002\n",
      "batch=1527 accuracy=0.8030923 loss=0.53719354\n",
      "batch=1528 accuracy=0.8031393 loss=0.49272668\n",
      "batch=1529 accuracy=0.803125 loss=0.40137607\n",
      "batch=1530 accuracy=0.8031924 loss=0.3596595\n",
      "batch=1531 accuracy=0.80321884 loss=0.3729536\n",
      "batch=1532 accuracy=0.80322486 loss=0.36476207\n",
      "batch=1533 accuracy=0.80327165 loss=0.20363186\n",
      "batch=1534 accuracy=0.8032777 loss=0.5175534\n",
      "batch=1535 accuracy=0.8033244 loss=0.5116295\n",
      "batch=1536 accuracy=0.8032287 loss=0.705918\n",
      "batch=1537 accuracy=0.803255 loss=0.3242625\n",
      "batch=1538 accuracy=0.80328137 loss=0.32188588\n",
      "batch=1539 accuracy=0.80330765 loss=0.33370087\n",
      "batch=1540 accuracy=0.80335414 loss=0.34887618\n",
      "batch=1541 accuracy=0.80333984 loss=0.5077366\n",
      "batch=1542 accuracy=0.80342674 loss=0.310242\n",
      "batch=1543 accuracy=0.8034731 loss=0.44568473\n",
      "batch=1544 accuracy=0.80347896 loss=0.59738445\n",
      "batch=1545 accuracy=0.8034444 loss=0.50228375\n",
      "batch=1546 accuracy=0.80351084 loss=0.2821815\n",
      "batch=1547 accuracy=0.8035166 loss=0.7124151\n",
      "batch=1548 accuracy=0.8035426 loss=0.4817811\n",
      "batch=1549 accuracy=0.8035484 loss=0.33460253\n",
      "batch=1550 accuracy=0.8036146 loss=0.2437153\n",
      "batch=1551 accuracy=0.8036002 loss=0.57906175\n",
      "batch=1552 accuracy=0.80364615 loss=0.4604448\n",
      "batch=1553 accuracy=0.8036921 loss=0.38976833\n",
      "batch=1554 accuracy=0.803758 loss=0.32400542\n",
      "batch=1555 accuracy=0.8037436 loss=0.3905412\n",
      "batch=1556 accuracy=0.8037291 loss=0.5609205\n",
      "batch=1557 accuracy=0.8037548 loss=0.43461138\n",
      "batch=1558 accuracy=0.80370027 loss=0.5653665\n",
      "batch=1559 accuracy=0.80370593 loss=0.42965934\n",
      "batch=1560 accuracy=0.80367154 loss=0.5358418\n",
      "batch=1561 accuracy=0.80369717 loss=0.51113385\n",
      "batch=1562 accuracy=0.8037628 loss=0.2748934\n",
      "batch=1563 accuracy=0.80380833 loss=0.45574087\n",
      "batch=1564 accuracy=0.80389374 loss=0.26489037\n",
      "batch=1565 accuracy=0.8039192 loss=0.33722943\n",
      "batch=1566 accuracy=0.80396456 loss=0.37338287\n",
      "batch=1567 accuracy=0.80398995 loss=0.4713916\n",
      "batch=1568 accuracy=0.8039954 loss=0.54355305\n",
      "batch=1569 accuracy=0.8040406 loss=0.32340068\n",
      "batch=1570 accuracy=0.8040261 loss=0.41702712\n",
      "batch=1571 accuracy=0.8040315 loss=0.5134077\n",
      "batch=1572 accuracy=0.80409646 loss=0.40441024\n",
      "batch=1573 accuracy=0.8041217 loss=0.60218275\n",
      "batch=1574 accuracy=0.8041071 loss=0.40266418\n",
      "batch=1575 accuracy=0.8041323 loss=0.4560743\n",
      "batch=1576 accuracy=0.8041971 loss=0.2767046\n",
      "batch=1577 accuracy=0.8041825 loss=0.658962\n",
      "batch=1578 accuracy=0.8041284 loss=0.63564324\n",
      "batch=1579 accuracy=0.80417323 loss=0.4137593\n",
      "batch=1580 accuracy=0.80417854 loss=0.41825098\n",
      "batch=1581 accuracy=0.8042035 loss=0.471332\n",
      "batch=1582 accuracy=0.80428773 loss=0.16952932\n",
      "batch=1583 accuracy=0.80431265 loss=0.4082929\n",
      "batch=1584 accuracy=0.80435723 loss=0.42389968\n",
      "batch=1585 accuracy=0.8042639 loss=0.89004445\n",
      "batch=1586 accuracy=0.80424935 loss=0.50295347\n",
      "batch=1587 accuracy=0.80429393 loss=0.26737556\n",
      "batch=1588 accuracy=0.8043384 loss=0.75350046\n",
      "batch=1589 accuracy=0.80442214 loss=0.33680356\n",
      "batch=1590 accuracy=0.804388 loss=0.5661522\n",
      "batch=1591 accuracy=0.80447155 loss=0.35641062\n",
      "batch=1592 accuracy=0.8044766 loss=0.5255425\n",
      "batch=1593 accuracy=0.80452085 loss=0.31200537\n",
      "batch=1594 accuracy=0.8045846 loss=0.31463027\n",
      "batch=1595 accuracy=0.80458957 loss=0.4645449\n",
      "batch=1596 accuracy=0.8046337 loss=0.40277147\n",
      "batch=1597 accuracy=0.8046777 loss=0.26700512\n",
      "batch=1598 accuracy=0.8046826 loss=0.5314096\n",
      "batch=1599 accuracy=0.8047461 loss=0.36182386\n",
      "batch=1600 accuracy=0.8047119 loss=0.61668855\n",
      "batch=1601 accuracy=0.80473626 loss=0.4577046\n",
      "batch=1602 accuracy=0.80476063 loss=0.43058103\n",
      "batch=1603 accuracy=0.80484337 loss=0.20151928\n",
      "batch=1604 accuracy=0.8048676 loss=0.4698568\n",
      "batch=1605 accuracy=0.80493075 loss=0.44848847\n",
      "batch=1606 accuracy=0.80493546 loss=0.5241622\n",
      "batch=1607 accuracy=0.8048818 loss=0.5143728\n",
      "batch=1608 accuracy=0.804906 loss=0.3671702\n",
      "batch=1609 accuracy=0.80496895 loss=0.4103333\n",
      "batch=1610 accuracy=0.8050124 loss=0.3887639\n",
      "batch=1611 accuracy=0.80505586 loss=0.2721087\n",
      "batch=1612 accuracy=0.8050217 loss=0.7019521\n",
      "batch=1613 accuracy=0.8050844 loss=0.3775587\n",
      "batch=1614 accuracy=0.8051277 loss=0.34478635\n",
      "batch=1615 accuracy=0.80511296 loss=0.40314478\n",
      "batch=1616 accuracy=0.8051755 loss=0.22887664\n",
      "batch=1617 accuracy=0.8051993 loss=0.38373232\n",
      "batch=1618 accuracy=0.80530035 loss=0.1722044\n",
      "batch=1619 accuracy=0.8053241 loss=0.4268653\n",
      "batch=1620 accuracy=0.80538636 loss=0.35622025\n",
      "batch=1621 accuracy=0.80544853 loss=0.30397767\n",
      "batch=1622 accuracy=0.8054529 loss=0.45742527\n",
      "batch=1623 accuracy=0.8054764 loss=0.311036\n",
      "batch=1624 accuracy=0.8054808 loss=0.41868925\n",
      "batch=1625 accuracy=0.8055043 loss=0.57299656\n",
      "batch=1626 accuracy=0.80560464 loss=0.13506702\n",
      "batch=1627 accuracy=0.8055897 loss=0.56854093\n",
      "batch=1628 accuracy=0.8055364 loss=1.008842\n",
      "batch=1629 accuracy=0.805579 loss=0.35446548\n",
      "batch=1630 accuracy=0.8056024 loss=0.3829894\n",
      "batch=1631 accuracy=0.8056449 loss=0.28477365\n",
      "batch=1632 accuracy=0.8056491 loss=0.38212663\n",
      "batch=1633 accuracy=0.8057107 loss=0.34411526\n",
      "batch=1634 accuracy=0.80579126 loss=0.23315541\n",
      "batch=1635 accuracy=0.8058527 loss=0.22031707\n",
      "batch=1636 accuracy=0.80581856 loss=0.6778501\n",
      "batch=1637 accuracy=0.8058799 loss=0.25684255\n",
      "batch=1638 accuracy=0.8059411 loss=0.29486513\n",
      "batch=1639 accuracy=0.80602133 loss=0.2303446\n",
      "batch=1640 accuracy=0.80610144 loss=0.3368345\n",
      "batch=1641 accuracy=0.8061434 loss=0.47423148\n",
      "batch=1642 accuracy=0.8061473 loss=0.42478412\n",
      "batch=1643 accuracy=0.8062082 loss=0.34604725\n",
      "batch=1644 accuracy=0.806288 loss=0.24176058\n",
      "batch=1645 accuracy=0.80634874 loss=0.25746566\n",
      "batch=1646 accuracy=0.80637145 loss=0.32620823\n",
      "batch=1647 accuracy=0.80643207 loss=0.36612493\n",
      "batch=1648 accuracy=0.8063599 loss=0.9393753\n",
      "batch=1649 accuracy=0.8063447 loss=0.6588547\n",
      "batch=1650 accuracy=0.8063863 loss=0.30549872\n",
      "batch=1651 accuracy=0.8063711 loss=0.7096318\n",
      "batch=1652 accuracy=0.8063937 loss=0.3827759\n",
      "batch=1653 accuracy=0.8063785 loss=0.38432682\n",
      "batch=1654 accuracy=0.8063822 loss=0.517056\n",
      "batch=1655 accuracy=0.8063859 loss=0.4976244\n",
      "batch=1656 accuracy=0.8063707 loss=0.56676376\n",
      "batch=1657 accuracy=0.8064498 loss=0.25534403\n",
      "batch=1658 accuracy=0.80641574 loss=0.4638927\n",
      "batch=1659 accuracy=0.8063629 loss=1.0035583\n",
      "batch=1660 accuracy=0.806329 loss=0.45814562\n",
      "batch=1661 accuracy=0.8062951 loss=0.5587495\n",
      "batch=1662 accuracy=0.806374 loss=0.20644325\n",
      "batch=1663 accuracy=0.8063777 loss=0.32471466\n",
      "batch=1664 accuracy=0.8064377 loss=0.2837045\n",
      "batch=1665 accuracy=0.8064226 loss=0.9255114\n",
      "batch=1666 accuracy=0.80640745 loss=0.54256546\n",
      "batch=1667 accuracy=0.8064486 loss=0.32360816\n",
      "batch=1668 accuracy=0.80652714 loss=0.29450268\n",
      "batch=1669 accuracy=0.8065307 loss=0.4994102\n",
      "batch=1670 accuracy=0.80655295 loss=0.37419027\n",
      "batch=1671 accuracy=0.8066126 loss=0.2564442\n",
      "batch=1672 accuracy=0.8066908 loss=0.24029094\n",
      "batch=1673 accuracy=0.806713 loss=0.5820689\n",
      "batch=1674 accuracy=0.8067351 loss=0.3193867\n",
      "batch=1675 accuracy=0.80675715 loss=0.35699177\n",
      "batch=1676 accuracy=0.80681646 loss=0.25563577\n",
      "batch=1677 accuracy=0.8068385 loss=0.48391485\n",
      "batch=1678 accuracy=0.8068605 loss=0.4334498\n",
      "batch=1679 accuracy=0.80691963 loss=0.3148558\n",
      "batch=1680 accuracy=0.80694157 loss=0.37222472\n",
      "batch=1681 accuracy=0.80694485 loss=0.46873492\n",
      "batch=1682 accuracy=0.8069667 loss=0.33022836\n",
      "batch=1683 accuracy=0.8070257 loss=0.37558603\n",
      "batch=1684 accuracy=0.80702895 loss=0.648129\n",
      "batch=1685 accuracy=0.80703217 loss=0.34333947\n",
      "batch=1686 accuracy=0.80703545 loss=0.4164483\n",
      "batch=1687 accuracy=0.8070757 loss=0.38364333\n",
      "batch=1688 accuracy=0.8070974 loss=0.34286696\n",
      "batch=1689 accuracy=0.8070821 loss=0.61493635\n",
      "batch=1690 accuracy=0.8071038 loss=0.40925732\n",
      "batch=1691 accuracy=0.807107 loss=0.34482265\n",
      "batch=1692 accuracy=0.8071471 loss=0.29263526\n",
      "batch=1693 accuracy=0.80724245 loss=0.23705113\n",
      "batch=1694 accuracy=0.80730087 loss=0.29775652\n",
      "batch=1695 accuracy=0.80737764 loss=0.32648182\n",
      "batch=1696 accuracy=0.8073991 loss=0.5561788\n",
      "batch=1697 accuracy=0.8074941 loss=0.22649166\n",
      "batch=1698 accuracy=0.8074971 loss=0.44643146\n",
      "batch=1699 accuracy=0.80751836 loss=0.49112254\n",
      "batch=1700 accuracy=0.8075764 loss=0.28178886\n",
      "batch=1701 accuracy=0.8075977 loss=0.3531894\n",
      "batch=1702 accuracy=0.80760056 loss=0.72685516\n",
      "batch=1703 accuracy=0.8076034 loss=0.39122522\n",
      "batch=1704 accuracy=0.8076796 loss=0.19638863\n",
      "batch=1705 accuracy=0.80766416 loss=0.504335\n",
      "batch=1706 accuracy=0.80768526 loss=0.52115625\n",
      "batch=1707 accuracy=0.8076698 loss=0.668453\n",
      "batch=1708 accuracy=0.80770916 loss=0.4950592\n",
      "batch=1709 accuracy=0.8077668 loss=0.32250875\n",
      "batch=1710 accuracy=0.80778784 loss=0.40502882\n",
      "batch=1711 accuracy=0.8077906 loss=0.52224743\n",
      "batch=1712 accuracy=0.80781156 loss=0.37519154\n",
      "batch=1713 accuracy=0.8078508 loss=0.40904245\n",
      "batch=1714 accuracy=0.80792636 loss=0.40440032\n",
      "batch=1715 accuracy=0.8079837 loss=0.349571\n",
      "batch=1716 accuracy=0.8079499 loss=0.8507144\n",
      "batch=1717 accuracy=0.80793434 loss=0.7514416\n",
      "batch=1718 accuracy=0.80799156 loss=0.33864534\n",
      "batch=1719 accuracy=0.807976 loss=0.4133599\n",
      "batch=1720 accuracy=0.80797863 loss=0.42065594\n",
      "batch=1721 accuracy=0.80801755 loss=0.4654444\n",
      "batch=1722 accuracy=0.8080927 loss=0.25433424\n",
      "batch=1723 accuracy=0.80807716 loss=0.6380875\n",
      "batch=1724 accuracy=0.8080797 loss=0.44917062\n",
      "batch=1725 accuracy=0.80811846 loss=0.4793769\n",
      "batch=1726 accuracy=0.8081572 loss=0.35758865\n",
      "batch=1727 accuracy=0.80821395 loss=0.26977023\n",
      "batch=1728 accuracy=0.8082345 loss=0.38232082\n",
      "batch=1729 accuracy=0.80821896 loss=0.51271945\n",
      "batch=1730 accuracy=0.8082756 loss=0.24989013\n",
      "batch=1731 accuracy=0.80833215 loss=0.4046412\n",
      "batch=1732 accuracy=0.8083706 loss=0.52180445\n",
      "batch=1733 accuracy=0.8083369 loss=0.54533434\n",
      "batch=1734 accuracy=0.80841136 loss=0.29194224\n",
      "batch=1735 accuracy=0.80844975 loss=0.32718724\n",
      "batch=1736 accuracy=0.8084881 loss=0.29121518\n",
      "batch=1737 accuracy=0.80849034 loss=0.51222026\n",
      "batch=1738 accuracy=0.80851066 loss=0.3656984\n",
      "batch=1739 accuracy=0.8085129 loss=0.49832132\n",
      "batch=1740 accuracy=0.808587 loss=0.25729287\n",
      "batch=1741 accuracy=0.80857134 loss=0.5004479\n",
      "batch=1742 accuracy=0.8085736 loss=0.4818513\n",
      "batch=1743 accuracy=0.8085758 loss=0.5415325\n",
      "batch=1744 accuracy=0.8086139 loss=0.37366116\n",
      "batch=1745 accuracy=0.8086877 loss=0.20699203\n",
      "batch=1746 accuracy=0.8086541 loss=0.569897\n",
      "batch=1747 accuracy=0.80865633 loss=0.4716661\n",
      "batch=1748 accuracy=0.80865854 loss=0.5583353\n",
      "batch=1749 accuracy=0.8086607 loss=0.52772117\n",
      "batch=1750 accuracy=0.8086629 loss=0.49751395\n",
      "batch=1751 accuracy=0.8086651 loss=0.34664774\n",
      "batch=1752 accuracy=0.8086851 loss=0.3581557\n",
      "batch=1753 accuracy=0.8087229 loss=0.31838733\n",
      "batch=1754 accuracy=0.8087963 loss=0.37372863\n",
      "batch=1755 accuracy=0.80879843 loss=0.41715136\n",
      "batch=1756 accuracy=0.8088005 loss=0.7096324\n",
      "batch=1757 accuracy=0.8087493 loss=0.6140415\n",
      "batch=1758 accuracy=0.80876917 loss=0.45074794\n",
      "batch=1759 accuracy=0.80880684 loss=0.34344763\n",
      "batch=1760 accuracy=0.8088267 loss=0.4775043\n",
      "batch=1761 accuracy=0.808811 loss=0.7025852\n",
      "batch=1762 accuracy=0.8088131 loss=0.5343567\n",
      "batch=1763 accuracy=0.8087975 loss=0.57657707\n",
      "batch=1764 accuracy=0.8088527 loss=0.36763668\n",
      "batch=1765 accuracy=0.80883706 loss=0.40133637\n",
      "batch=1766 accuracy=0.8087861 loss=0.4723366\n",
      "batch=1767 accuracy=0.8087705 loss=0.6463327\n",
      "batch=1768 accuracy=0.8087549 loss=0.55973744\n",
      "batch=1769 accuracy=0.8087394 loss=0.4401333\n",
      "batch=1770 accuracy=0.80875915 loss=0.38478136\n",
      "batch=1771 accuracy=0.8087966 loss=0.26018888\n",
      "batch=1772 accuracy=0.80888677 loss=0.22493419\n",
      "batch=1773 accuracy=0.8088712 loss=0.6138053\n",
      "batch=1774 accuracy=0.80887324 loss=0.38335538\n",
      "batch=1775 accuracy=0.8088577 loss=0.57970953\n",
      "batch=1776 accuracy=0.80889493 loss=0.40922755\n",
      "batch=1777 accuracy=0.80894965 loss=0.2911008\n",
      "batch=1778 accuracy=0.8089341 loss=0.4242016\n",
      "batch=1779 accuracy=0.8089712 loss=0.41799843\n",
      "batch=1780 accuracy=0.8089907 loss=0.31642476\n",
      "batch=1781 accuracy=0.8090278 loss=0.50418013\n",
      "batch=1782 accuracy=0.8090472 loss=0.44631076\n",
      "batch=1783 accuracy=0.80908424 loss=0.33602843\n",
      "batch=1784 accuracy=0.80908614 loss=0.38684544\n",
      "batch=1785 accuracy=0.80917555 loss=0.26986727\n",
      "batch=1786 accuracy=0.80922985 loss=0.35065922\n",
      "batch=1787 accuracy=0.80921423 loss=0.369452\n",
      "batch=1788 accuracy=0.8091811 loss=0.4909822\n",
      "batch=1789 accuracy=0.8092179 loss=0.35281938\n",
      "batch=1790 accuracy=0.8092197 loss=0.39703694\n",
      "batch=1791 accuracy=0.80922157 loss=0.5458877\n",
      "batch=1792 accuracy=0.8092408 loss=0.39895126\n",
      "batch=1793 accuracy=0.8092775 loss=0.32060868\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-1-8dc24ed91e05>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     81\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     82\u001b[0m                                    )\n\u001b[0;32m---> 83\u001b[0;31m         \u001b[0mfairing\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     84\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     85\u001b[0m         \u001b[0mremote_train\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mMyFashionMnist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/kubeflow/fairing/config.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    129\u001b[0m         \u001b[0mbuilder\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    130\u001b[0m         \u001b[0mpod_spec\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbuilder\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgenerate_pod_spec\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 131\u001b[0;31m         \u001b[0mdeployer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdeploy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpod_spec\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    133\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mpreprocessor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuilder\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdeployer\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/kubeflow/fairing/deployers/job/job.py\u001b[0m in \u001b[0;36mdeploy\u001b[0;34m(self, pod_spec)\u001b[0m\n\u001b[1;32m     91\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     92\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstream_log\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 93\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_logs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     94\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     95\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/kubeflow/fairing/deployers/job/job.py\u001b[0m in \u001b[0;36mget_logs\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    153\u001b[0m                          \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_created_job\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnamespace\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m                          \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlabels\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m                          container=\"fairing-job\")\n\u001b[0m\u001b[1;32m    156\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcleanup\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/kubeflow/fairing/kubernetes/manager.py\u001b[0m in \u001b[0;36mlog\u001b[0;34m(self, name, namespace, selectors, container, follow)\u001b[0m\n\u001b[1;32m    259\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mtail\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    260\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 261\u001b[0;31m                 \u001b[0;32mfor\u001b[0m \u001b[0mchunk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtail\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstream\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mMAX_STREAM_BYTES\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    262\u001b[0m                     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchunk\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrstrip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'utf8'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    263\u001b[0m             \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/urllib3/response.py\u001b[0m in \u001b[0;36mstream\u001b[0;34m(self, amt, decode_content)\u001b[0m\n\u001b[1;32m    488\u001b[0m         \"\"\"\n\u001b[1;32m    489\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchunked\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msupports_chunked_reads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 490\u001b[0;31m             \u001b[0;32mfor\u001b[0m \u001b[0mline\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_chunked\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mamt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdecode_content\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdecode_content\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    491\u001b[0m                 \u001b[0;32myield\u001b[0m \u001b[0mline\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    492\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/urllib3/response.py\u001b[0m in \u001b[0;36mread_chunked\u001b[0;34m(self, amt, decode_content)\u001b[0m\n\u001b[1;32m    664\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    665\u001b[0m             \u001b[0;32mwhile\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 666\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_update_chunk_length\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    667\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchunk_left\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    668\u001b[0m                     \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/urllib3/response.py\u001b[0m in \u001b[0;36m_update_chunk_length\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    596\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchunk_left\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    597\u001b[0m             \u001b[0;32mreturn\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 598\u001b[0;31m         \u001b[0mline\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreadline\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    599\u001b[0m         \u001b[0mline\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mline\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msplit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mb';'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    600\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/lib/python3.6/socket.py\u001b[0m in \u001b[0;36mreadinto\u001b[0;34m(self, b)\u001b[0m\n\u001b[1;32m    584\u001b[0m         \u001b[0;32mwhile\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    585\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 586\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv_into\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    587\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mtimeout\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    588\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_timeout_occurred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/lib/python3.6/ssl.py\u001b[0m in \u001b[0;36mrecv_into\u001b[0;34m(self, buffer, nbytes, flags)\u001b[0m\n\u001b[1;32m   1010\u001b[0m                   \u001b[0;34m\"non-zero flags not allowed in calls to recv_into() on %s\"\u001b[0m \u001b[0;34m%\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1011\u001b[0m                   self.__class__)\n\u001b[0;32m-> 1012\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnbytes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1013\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1014\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0msocket\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv_into\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnbytes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mflags\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/lib/python3.6/ssl.py\u001b[0m in \u001b[0;36mread\u001b[0;34m(self, len, buffer)\u001b[0m\n\u001b[1;32m    872\u001b[0m             \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Read on closed or unwrapped SSL socket.\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    873\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 874\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sslobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    875\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mSSLError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    876\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mSSL_ERROR_EOF\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msuppress_ragged_eofs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/lib/python3.6/ssl.py\u001b[0m in \u001b[0;36mread\u001b[0;34m(self, len, buffer)\u001b[0m\n\u001b[1;32m    629\u001b[0m         \"\"\"\n\u001b[1;32m    630\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mbuffer\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 631\u001b[0;31m             \u001b[0mv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sslobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    632\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    633\u001b[0m             \u001b[0mv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sslobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import os\n",
    "import argparse\n",
    "from tensorflow.python.keras.callbacks import Callback\n",
    "\n",
    "\n",
    "\n",
    "class MyFashionMnist(object):\n",
    "  def train(self):\n",
    "    \n",
    "    # 입력 값을 받게 추가합니다.\n",
    "    parser = argparse.ArgumentParser()\n",
    "    parser.add_argument('--learning_rate', required=False, type=float, default=0.001)\n",
    "    parser.add_argument('--dropout_rate', required=False, type=float, default=0.2)\n",
    "    # epoch 5 ~ 15\n",
    "    parser.add_argument('--epoch', required=False, type=int, default=5)    \n",
    "    # relu, sigmoid, softmax, tanh\n",
    "    parser.add_argument('--act', required=False, type=str, default='relu')        \n",
    "    # layer 1 ~ 5\n",
    "    parser.add_argument('--layer', required=False, type=int, default=1)        \n",
    "    \n",
    "    \n",
    "    args = parser.parse_args()    \n",
    "    \n",
    "    (x_train, y_train), (x_test, y_test) = tf.keras.datasets.fashion_mnist.load_data()\n",
    "    x_train, x_test = x_train / 255.0, x_test / 255.0\n",
    "\n",
    "    model = tf.keras.models.Sequential()\n",
    "    model.add(tf.keras.layers.Flatten(input_shape=(28, 28)))\n",
    "    \n",
    "    for i in range(int(args.layer)):    \n",
    "        model.add(tf.keras.layers.Dense(128, activation=args.act))\n",
    "        model.add(tf.keras.layers.Dropout(args.dropout_rate))\n",
    "        \n",
    "    model.add(tf.keras.layers.Dense(10, activation='softmax'))\n",
    "    model.summary()\n",
    "    \n",
    "    model.compile(optimizer=tf.keras.optimizers.Adam(lr=args.learning_rate),\n",
    "                  loss='sparse_categorical_crossentropy',\n",
    "                  metrics=['acc'])\n",
    "\n",
    "    model.fit(x_train, y_train,\n",
    "              verbose=0,\n",
    "              validation_data=(x_test, y_test),\n",
    "              epochs=args.epoch,\n",
    "              callbacks=[KatibMetricLog()])\n",
    "\n",
    "    model.evaluate(x_test,  y_test, verbose=0)\n",
    "\n",
    "class KatibMetricLog(Callback):\n",
    "    def on_batch_end(self, batch, logs={}):\n",
    "        print(\"batch=\" + str(batch),\n",
    "              \"accuracy=\" + str(logs.get('acc')),\n",
    "              \"loss=\" + str(logs.get('loss')))\n",
    "    def on_epoch_begin(self, epoch, logs={}):\n",
    "        print(\"epoch \" + str(epoch) + \":\")\n",
    "    \n",
    "    def on_epoch_end(self, epoch, logs={}):\n",
    "        print(\"Validation-accuracy=\" + str(logs.get('val_acc')),\n",
    "              \"Validation-loss=\" + str(logs.get('val_loss')))\n",
    "        return\n",
    "\n",
    "if __name__ == '__main__':\n",
    "    if os.getenv('FAIRING_RUNTIME', None) is None:\n",
    "        from kubeflow import fairing\n",
    "        from kubeflow.fairing.kubernetes import utils as k8s_utils\n",
    "\n",
    "        DOCKER_REGISTRY = 'kubeflow-registry.default.svc.cluster.local:30000'\n",
    "        fairing.config.set_builder(\n",
    "            'append',\n",
    "            image_name='fairing-job',\n",
    "            base_image='brightfly/kubeflow-jupyter-lab:tf2.0-cpu',\n",
    "            registry=DOCKER_REGISTRY, \n",
    "            push=True)\n",
    "        # cpu 2, memory 5GiB\n",
    "        fairing.config.set_deployer('job',\n",
    "                                    namespace='dudaji',\n",
    "                                    pod_spec_mutators=[\n",
    "                                        k8s_utils.get_resource_mutator(cpu=1,\n",
    "                                                                       memory=2)]\n",
    "         \n",
    "                                   )\n",
    "        fairing.config.run()\n",
    "    else:\n",
    "        remote_train = MyFashionMnist()\n",
    "        remote_train.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}