{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train Mini Jasper"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Paper related https://arxiv.org/abs/1904.03288\n",
    "\n",
    "Small factor of JasperNet, defined at https://github.com/NVIDIA/OpenSeq2Seq/blob/master/example_configs/speech2text/jasper-Mini-for-Jetson.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-info\">\n",
    "\n",
    "This tutorial is available as an IPython notebook at [malaya-speech/example/train-asr](https://github.com/huseinzol05/malaya-speech/tree/master/example/train-asr).\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-warning\">\n",
    "\n",
    "This example trained on a very small dataset, do not use it for production.\n",
    "    \n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Interface\n",
    "\n",
    "To initiate Mini Jasper,\n",
    "\n",
    "```python\n",
    "import malaya_speech.train.model.mini_jasper as mini_jasper\n",
    "\n",
    "x = tf.placeholder(tf.float32, [None, None, num_features])\n",
    "x_len = tf.placeholder(tf.int32, [None])\n",
    "mini_jasper.Model(x, x_len)\n",
    "```\n",
    "\n",
    "Required input lengths, so the model able to calculate output lengths.\n",
    "\n",
    "Also included actual size of Jasper,\n",
    "\n",
    "```python\n",
    "import malaya_speech.train.model.jasper as jasper\n",
    "\n",
    "x = tf.placeholder(tf.float32, [None, None, num_features])\n",
    "x_len = tf.placeholder(tf.int32, [None])\n",
    "jasper.Model(x, x_len)\n",
    "```\n",
    "\n",
    "But in this example, we just use Mini Jasper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "os.environ['CUDA_VISIBLE_DEVICES'] = '0'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/quartznet/layer.py:6: The name tf.layers.Conv1D is deprecated. Please use tf.compat.v1.layers.Conv1D instead.\n",
      "\n",
      "WARNING:tensorflow:\n",
      "The TensorFlow contrib module will not be included in TensorFlow 2.0.\n",
      "For more information, please see:\n",
      "  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md\n",
      "  * https://github.com/tensorflow/addons\n",
      "  * https://github.com/tensorflow/io (for I/O related ops)\n",
      "If you depend on functionality not listed there, please file an issue.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from glob import glob\n",
    "import malaya_speech.config\n",
    "import malaya_speech.train.model.mini_jasper as mini_jasper\n",
    "import malaya_speech.train.model.ctc as ctc\n",
    "import malaya_speech.train as train\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Featurizer\n",
    "\n",
    "We prefer use `malaya_speech.tf_featurization.STTFeaturizer` for ASR."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = malaya_speech.config.transducer_featurizer_config\n",
    "featurizer = malaya_speech.tf_featurization.STTFeaturizer(**config)\n",
    "n_mels = config['num_feature_bins']\n",
    "initial_learning_rate = 1e-3\n",
    "max_gradient_norm = 5.0\n",
    "train_steps = 1000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define Data Pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_inputs(example):\n",
    "    s = featurizer.vectorize(example['waveforms'])\n",
    "    mel_fbanks = tf.reshape(s, (-1, n_mels))\n",
    "    length = tf.cast(tf.shape(mel_fbanks)[0], tf.int32)\n",
    "    length = tf.expand_dims(length, 0)\n",
    "    example['inputs'] = mel_fbanks\n",
    "    example['inputs_length'] = length\n",
    "\n",
    "    return example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse(serialized_example):\n",
    "\n",
    "    data_fields = {\n",
    "        'waveforms': tf.VarLenFeature(tf.float32),\n",
    "        'targets': tf.VarLenFeature(tf.int64),\n",
    "    }\n",
    "    features = tf.parse_single_example(\n",
    "        serialized_example, features = data_fields\n",
    "    )\n",
    "    for k in features.keys():\n",
    "        features[k] = features[k].values\n",
    "\n",
    "    features = preprocess_inputs(features)\n",
    "\n",
    "    keys = list(features.keys())\n",
    "    for k in keys:\n",
    "        if k not in ['inputs', 'inputs_length', 'targets']:\n",
    "            features.pop(k, None)\n",
    "\n",
    "    return features\n",
    "\n",
    "\n",
    "def get_dataset(path, batch_size = 32, shuffle_size = 32, thread_count = 24):\n",
    "    def get():\n",
    "        files = glob(path)\n",
    "        dataset = tf.data.TFRecordDataset(files)\n",
    "        dataset = dataset.shuffle(shuffle_size)\n",
    "        dataset = dataset.map(parse, num_parallel_calls = thread_count)\n",
    "        dataset = dataset.padded_batch(\n",
    "            batch_size,\n",
    "            padded_shapes = {\n",
    "                'inputs': tf.TensorShape([None, n_mels]),\n",
    "                'inputs_length': tf.TensorShape([None]),\n",
    "                'targets': tf.TensorShape([None]),\n",
    "            },\n",
    "            padding_values = {\n",
    "                'inputs': tf.constant(0, dtype = tf.float32),\n",
    "                'inputs_length': tf.constant(0, dtype = tf.int32),\n",
    "                'targets': tf.constant(0, dtype = tf.int64),\n",
    "            },\n",
    "        )\n",
    "        dataset = dataset.repeat()\n",
    "        return dataset\n",
    "\n",
    "    return get"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define Model definition\n",
    "\n",
    "1. Use malaya_speech CTC loss,\n",
    "\n",
    "```python\n",
    "mean_error, sum_error, sum_weight = ctc.loss.ctc_loss(\n",
    "    logits, targets_int32, seq_lens\n",
    ")\n",
    "```\n",
    "\n",
    "2. Use malaya_speech CTC sequence accuracy,\n",
    "\n",
    "```python\n",
    "accuracy = ctc.metrics.ctc_sequence_accuracy(\n",
    "    logits, targets_int32, seq_lens\n",
    ")\n",
    "```\n",
    "\n",
    "This will automatically recorded in Tensorboard.\n",
    "\n",
    "3. Use malaya_speech metrics for Evaluation session,\n",
    "\n",
    "```python\n",
    "elif mode == tf.estimator.ModeKeys.EVAL:\n",
    "\n",
    "    estimator_spec = tf.estimator.EstimatorSpec(\n",
    "        mode = tf.estimator.ModeKeys.EVAL,\n",
    "        loss = loss,\n",
    "        eval_metric_ops = {\n",
    "            'accuracy': ctc.metrics.ctc_sequence_accuracy_estimator(\n",
    "                logits, targets_int32, seq_lens\n",
    "            ),\n",
    "            'WER': ctc.metrics.word_error_rate_estimator(\n",
    "                logits, targets_int32\n",
    "            ),\n",
    "        },\n",
    "    )\n",
    "```\n",
    "\n",
    "If we passed directly tensor values to `eval_metric_ops`, Tensorflow will throw error, required update operations. So just stick with that implementation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_fn(features, labels, mode, params):\n",
    "\n",
    "    model = mini_jasper.Model(\n",
    "        features['inputs'], features['inputs_length'][:, 0], mode = 'train'\n",
    "    )\n",
    "    logits = tf.layers.dense(\n",
    "        model.logits['outputs'], malaya_speech.char.VOCAB_SIZE\n",
    "    )\n",
    "    seq_lens = model.logits['src_length']\n",
    "\n",
    "    targets_int32 = tf.cast(features['targets'], tf.int32)\n",
    "\n",
    "    mean_error, sum_error, sum_weight = ctc.loss.ctc_loss(\n",
    "        logits, targets_int32, seq_lens\n",
    "    )\n",
    "\n",
    "    loss = mean_error\n",
    "    accuracy = ctc.metrics.ctc_sequence_accuracy(\n",
    "        logits, targets_int32, seq_lens\n",
    "    )\n",
    "\n",
    "    tf.identity(loss, 'train_loss')\n",
    "    tf.identity(accuracy, name = 'train_accuracy')\n",
    "\n",
    "    if mode == tf.estimator.ModeKeys.TRAIN:\n",
    "        global_step = tf.train.get_or_create_global_step()\n",
    "        learning_rate = tf.train.polynomial_decay(\n",
    "            initial_learning_rate,\n",
    "            global_step,\n",
    "            train_steps,\n",
    "            end_learning_rate = 1e-4,\n",
    "            power = 1.0,\n",
    "            cycle = False,\n",
    "        )\n",
    "        optimizer = tf.train.AdamOptimizer(learning_rate = learning_rate)\n",
    "        train_op = optimizer.minimize(loss, global_step = global_step)\n",
    "        estimator_spec = tf.estimator.EstimatorSpec(\n",
    "            mode = mode, loss = loss, train_op = train_op\n",
    "        )\n",
    "\n",
    "    elif mode == tf.estimator.ModeKeys.EVAL:\n",
    "\n",
    "        estimator_spec = tf.estimator.EstimatorSpec(\n",
    "            mode = tf.estimator.ModeKeys.EVAL,\n",
    "            loss = loss,\n",
    "            eval_metric_ops = {\n",
    "                'accuracy': ctc.metrics.ctc_sequence_accuracy_estimator(\n",
    "                    logits, targets_int32, seq_lens\n",
    "                ),\n",
    "                'WER': ctc.metrics.word_error_rate_estimator(\n",
    "                    logits, targets_int32\n",
    "                ),\n",
    "            },\n",
    "        )\n",
    "\n",
    "    return estimator_spec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_hooks = [\n",
    "    tf.train.LoggingTensorHook(\n",
    "        ['train_accuracy', 'train_loss'], every_n_iter = 1\n",
    "    )\n",
    "]\n",
    "train_dataset = get_dataset('tolong-sebut/data/tolong-sebut-train*')\n",
    "dev_dataset = get_dataset('tolong-sebut/data/tolong-sebut-dev*')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "!rm -rf asr-minijasper"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initate training session"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/__init__.py:38: The name tf.logging.set_verbosity is deprecated. Please use tf.compat.v1.logging.set_verbosity instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/__init__.py:38: The name tf.logging.INFO is deprecated. Please use tf.compat.v1.logging.INFO instead.\n",
      "\n",
      "INFO:tensorflow:Using config: {'_model_dir': 'asr-minijasper', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': 200, '_save_checkpoints_secs': None, '_session_config': allow_soft_placement: true\n",
      "graph_options {\n",
      "  rewrite_options {\n",
      "    meta_optimizer_iterations: ONE\n",
      "  }\n",
      "}\n",
      ", '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 1, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x7f44d1b5a4a8>, '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}\n",
      "INFO:tensorflow:Not using Distribute Coordinator.\n",
      "INFO:tensorflow:Running training and evaluation locally (non-distributed).\n",
      "INFO:tensorflow:Start train and evaluate loop. The evaluate will happen after every checkpoint. Checkpoint frequency is determined based on RunConfig arguments: save_checkpoints_steps 200 or save_checkpoints_secs None.\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/training/training_util.py:236: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/autograph/converters/directives.py:119: The name tf.VarLenFeature is deprecated. Please use tf.io.VarLenFeature instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/autograph/converters/directives.py:119: The name tf.parse_single_example is deprecated. Please use tf.io.parse_single_example instead.\n",
      "\n",
      "INFO:tensorflow:Calling model_fn.\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/quartznet/abstract.py:141: The name tf.variable_scope is deprecated. Please use tf.compat.v1.variable_scope instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/quartznet/layer.py:327: separable_conv1d (from tensorflow.python.layers.convolutional) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use `tf.keras.layers.SeparableConv1D` instead.\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/layers/convolutional.py:971: Layer.apply (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `layer.__call__` method instead.\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/quartznet/layer.py:358: batch_normalization (from tensorflow.python.layers.normalization) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use keras.layers.BatchNormalization instead.  In particular, `tf.control_dependencies(tf.GraphKeys.UPDATE_OPS)` should not be used (consult the `tf.keras.layers.batch_normalization` documentation).\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/quartznet/abstract.py:369: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "WARNING:tensorflow:From <ipython-input-7-9570020f60e8>:7: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use keras.layers.Dense instead.\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/ctc/loss.py:14: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use `tf.cast` instead.\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/ops/array_ops.py:1475: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/ctc/loss.py:20: The name tf.nn.ctc_loss is deprecated. Please use tf.compat.v1.nn.ctc_loss instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/util/deprecation.py:507: calling count_nonzero (from tensorflow.python.ops.math_ops) with axis is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "reduction_indices is deprecated, use axis instead\n",
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Create CheckpointSaverHook.\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n",
      "INFO:tensorflow:Saving checkpoints for 0 into asr-minijasper/model.ckpt.\n",
      "INFO:tensorflow:train_accuracy = 0.0025773195, train_loss = 724.7258\n",
      "INFO:tensorflow:loss = 724.7258, step = 0\n",
      "INFO:tensorflow:global_step/sec: 0.399545\n",
      "INFO:tensorflow:train_accuracy = 0.058744993, train_loss = 506.95657 (2.503 sec)\n",
      "INFO:tensorflow:loss = 506.95657, step = 1 (2.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.6977\n",
      "INFO:tensorflow:train_accuracy = 0.053254437, train_loss = 326.39246 (0.590 sec)\n",
      "INFO:tensorflow:loss = 326.39246, step = 2 (0.590 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9287\n",
      "INFO:tensorflow:train_accuracy = 0.035, train_loss = 243.18027 (0.518 sec)\n",
      "INFO:tensorflow:loss = 243.18027, step = 3 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.72866\n",
      "INFO:tensorflow:train_accuracy = 0.005, train_loss = 178.23248 (0.579 sec)\n",
      "INFO:tensorflow:loss = 178.23248, step = 4 (0.579 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84795\n",
      "INFO:tensorflow:train_accuracy = 0.025606468, train_loss = 134.63281 (0.541 sec)\n",
      "INFO:tensorflow:loss = 134.63281, step = 5 (0.541 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04673\n",
      "INFO:tensorflow:train_accuracy = 0.005319149, train_loss = 114.20859 (0.489 sec)\n",
      "INFO:tensorflow:loss = 114.20859, step = 6 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98297\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 100.00012 (0.504 sec)\n",
      "INFO:tensorflow:loss = 100.00012, step = 7 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00969\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 91.21152 (0.498 sec)\n",
      "INFO:tensorflow:loss = 91.21152, step = 8 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83032\n",
      "INFO:tensorflow:train_accuracy = 0.006476684, train_loss = 92.35603 (0.546 sec)\n",
      "INFO:tensorflow:loss = 92.35603, step = 9 (0.546 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0557\n",
      "INFO:tensorflow:train_accuracy = 0.001312336, train_loss = 81.915085 (0.487 sec)\n",
      "INFO:tensorflow:loss = 81.915085, step = 10 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09034\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 65.02089 (0.478 sec)\n",
      "INFO:tensorflow:loss = 65.02089, step = 11 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84006\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 64.36902 (0.544 sec)\n",
      "INFO:tensorflow:loss = 64.36902, step = 12 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09803\n",
      "INFO:tensorflow:train_accuracy = 0.0012886598, train_loss = 68.24688 (0.477 sec)\n",
      "INFO:tensorflow:loss = 68.24688, step = 13 (0.477 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03696\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 58.303776 (0.491 sec)\n",
      "INFO:tensorflow:loss = 58.303776, step = 14 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02052\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 57.889862 (0.495 sec)\n",
      "INFO:tensorflow:loss = 57.889862, step = 15 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21574\n",
      "INFO:tensorflow:train_accuracy = 0.0039011703, train_loss = 57.588512 (0.451 sec)\n",
      "INFO:tensorflow:loss = 57.588512, step = 16 (0.451 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.4793\n",
      "INFO:tensorflow:train_accuracy = 0.0050188205, train_loss = 60.14511 (0.403 sec)\n",
      "INFO:tensorflow:loss = 60.14511, step = 17 (0.403 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11518\n",
      "INFO:tensorflow:train_accuracy = 0.0027137042, train_loss = 49.564636 (0.473 sec)\n",
      "INFO:tensorflow:loss = 49.564636, step = 18 (0.472 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.54407\n",
      "INFO:tensorflow:train_accuracy = 0.0013262599, train_loss = 58.36293 (0.393 sec)\n",
      "INFO:tensorflow:loss = 58.36293, step = 19 (0.394 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09752\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 64.0806 (0.477 sec)\n",
      "INFO:tensorflow:loss = 64.0806, step = 20 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98654\n",
      "INFO:tensorflow:train_accuracy = 0.0025940337, train_loss = 51.485535 (0.504 sec)\n",
      "INFO:tensorflow:loss = 51.485535, step = 21 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.43972\n",
      "INFO:tensorflow:train_accuracy = 0.0025542784, train_loss = 54.174507 (0.410 sec)\n",
      "INFO:tensorflow:loss = 54.174507, step = 22 (0.410 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08941\n",
      "INFO:tensorflow:train_accuracy = 0.009433962, train_loss = 48.97183 (0.479 sec)\n",
      "INFO:tensorflow:loss = 48.97183, step = 23 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05669\n",
      "INFO:tensorflow:train_accuracy = 0.007905139, train_loss = 65.10916 (0.486 sec)\n",
      "INFO:tensorflow:loss = 65.10916, step = 24 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9432\n",
      "INFO:tensorflow:train_accuracy = 0.0038860103, train_loss = 48.82689 (0.515 sec)\n",
      "INFO:tensorflow:loss = 48.82689, step = 25 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96494\n",
      "INFO:tensorflow:train_accuracy = 0.0026845639, train_loss = 49.525047 (0.509 sec)\n",
      "INFO:tensorflow:loss = 49.525047, step = 26 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04559\n",
      "INFO:tensorflow:train_accuracy = 0.0037641155, train_loss = 56.719467 (0.488 sec)\n",
      "INFO:tensorflow:loss = 56.719467, step = 27 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03516\n",
      "INFO:tensorflow:train_accuracy = 0.0013003901, train_loss = 47.173588 (0.492 sec)\n",
      "INFO:tensorflow:loss = 47.173588, step = 28 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.6741\n",
      "INFO:tensorflow:train_accuracy = 0.0038363172, train_loss = 52.478592 (0.374 sec)\n",
      "INFO:tensorflow:loss = 52.478592, step = 29 (0.374 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20966\n",
      "INFO:tensorflow:train_accuracy = 0.0012594458, train_loss = 48.47357 (0.452 sec)\n",
      "INFO:tensorflow:loss = 48.47357, step = 30 (0.452 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11213\n",
      "INFO:tensorflow:train_accuracy = 0.0038461538, train_loss = 48.13115 (0.474 sec)\n",
      "INFO:tensorflow:loss = 48.13115, step = 31 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18734\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 51.70265 (0.457 sec)\n",
      "INFO:tensorflow:loss = 51.70265, step = 32 (0.457 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14538\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 48.473965 (0.466 sec)\n",
      "INFO:tensorflow:loss = 48.473965, step = 33 (0.467 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21515\n",
      "INFO:tensorflow:train_accuracy = 0.007853403, train_loss = 53.605797 (0.451 sec)\n",
      "INFO:tensorflow:loss = 53.605797, step = 34 (0.451 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92417\n",
      "INFO:tensorflow:train_accuracy = 0.007290401, train_loss = 54.757706 (0.520 sec)\n",
      "INFO:tensorflow:loss = 54.757706, step = 35 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.25521\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 47.362015 (0.443 sec)\n",
      "INFO:tensorflow:loss = 47.362015, step = 36 (0.444 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04552\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 45.36531 (0.489 sec)\n",
      "INFO:tensorflow:loss = 45.36531, step = 37 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.31664\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 47.204563 (0.431 sec)\n",
      "INFO:tensorflow:loss = 47.204563, step = 38 (0.431 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.47275\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 42.515358 (0.405 sec)\n",
      "INFO:tensorflow:loss = 42.515358, step = 39 (0.406 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.60102\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 41.9398 (0.384 sec)\n",
      "INFO:tensorflow:loss = 41.9398, step = 40 (0.384 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89298\n",
      "INFO:tensorflow:train_accuracy = 0.032869786, train_loss = 47.192486 (0.528 sec)\n",
      "INFO:tensorflow:loss = 47.192486, step = 41 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20811\n",
      "INFO:tensorflow:train_accuracy = 0.018064516, train_loss = 40.293747 (0.453 sec)\n",
      "INFO:tensorflow:loss = 40.293747, step = 42 (0.453 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01551\n",
      "INFO:tensorflow:train_accuracy = 0.026845638, train_loss = 37.10353 (0.496 sec)\n",
      "INFO:tensorflow:loss = 37.10353, step = 43 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.45263\n",
      "INFO:tensorflow:train_accuracy = 0.010230179, train_loss = 41.98634 (0.408 sec)\n",
      "INFO:tensorflow:loss = 41.98634, step = 44 (0.408 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.46642\n",
      "INFO:tensorflow:train_accuracy = 0.018779343, train_loss = 42.38569 (0.155 sec)\n",
      "INFO:tensorflow:loss = 42.38569, step = 45 (0.155 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94026\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 40.01487 (0.515 sec)\n",
      "INFO:tensorflow:loss = 40.01487, step = 46 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.47992\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 39.459282 (0.403 sec)\n",
      "INFO:tensorflow:loss = 39.459282, step = 47 (0.403 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94708\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 43.16781 (0.514 sec)\n",
      "INFO:tensorflow:loss = 43.16781, step = 48 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08557\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 39.5 (0.479 sec)\n",
      "INFO:tensorflow:loss = 39.5, step = 49 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11056\n",
      "INFO:tensorflow:train_accuracy = 0.0, train_loss = 40.54854 (0.474 sec)\n",
      "INFO:tensorflow:loss = 40.54854, step = 50 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17297\n",
      "INFO:tensorflow:train_accuracy = 0.006684492, train_loss = 35.02738 (0.460 sec)\n",
      "INFO:tensorflow:loss = 35.02738, step = 51 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1344\n",
      "INFO:tensorflow:train_accuracy = 0.038512617, train_loss = 35.401386 (0.469 sec)\n",
      "INFO:tensorflow:loss = 35.401386, step = 52 (0.469 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10948\n",
      "INFO:tensorflow:train_accuracy = 0.038107753, train_loss = 34.86833 (0.474 sec)\n",
      "INFO:tensorflow:loss = 34.86833, step = 53 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07587\n",
      "INFO:tensorflow:train_accuracy = 0.03548796, train_loss = 38.445595 (0.482 sec)\n",
      "INFO:tensorflow:loss = 38.445595, step = 54 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.26243\n",
      "INFO:tensorflow:train_accuracy = 0.015936255, train_loss = 31.80059 (0.442 sec)\n",
      "INFO:tensorflow:loss = 31.80059, step = 55 (0.442 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01261\n",
      "INFO:tensorflow:train_accuracy = 0.0053547523, train_loss = 34.857674 (0.497 sec)\n",
      "INFO:tensorflow:loss = 34.857674, step = 56 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12276\n",
      "INFO:tensorflow:train_accuracy = 0.007905139, train_loss = 35.96846 (0.470 sec)\n",
      "INFO:tensorflow:loss = 35.96846, step = 57 (0.469 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18673\n",
      "INFO:tensorflow:train_accuracy = 0.013020833, train_loss = 35.46472 (0.457 sec)\n",
      "INFO:tensorflow:loss = 35.46472, step = 58 (0.458 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14039\n",
      "INFO:tensorflow:train_accuracy = 0.02607562, train_loss = 36.45217 (0.467 sec)\n",
      "INFO:tensorflow:loss = 36.45217, step = 59 (0.467 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24082\n",
      "INFO:tensorflow:train_accuracy = 0.028608583, train_loss = 34.635532 (0.447 sec)\n",
      "INFO:tensorflow:loss = 34.635532, step = 60 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06473\n",
      "INFO:tensorflow:train_accuracy = 0.036410924, train_loss = 34.757175 (0.484 sec)\n",
      "INFO:tensorflow:loss = 34.757175, step = 61 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10201\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.040920716, train_loss = 36.700653 (0.476 sec)\n",
      "INFO:tensorflow:loss = 36.700653, step = 62 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.35393\n",
      "INFO:tensorflow:train_accuracy = 0.040897097, train_loss = 31.582521 (0.425 sec)\n",
      "INFO:tensorflow:loss = 31.582521, step = 63 (0.426 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05959\n",
      "INFO:tensorflow:train_accuracy = 0.043137256, train_loss = 34.118744 (0.485 sec)\n",
      "INFO:tensorflow:loss = 34.118744, step = 64 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94399\n",
      "INFO:tensorflow:train_accuracy = 0.062834226, train_loss = 49.79178 (0.515 sec)\n",
      "INFO:tensorflow:loss = 49.79178, step = 65 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15142\n",
      "INFO:tensorflow:train_accuracy = 0.017971758, train_loss = 34.08287 (0.465 sec)\n",
      "INFO:tensorflow:loss = 34.08287, step = 66 (0.465 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07075\n",
      "INFO:tensorflow:train_accuracy = 0.016169155, train_loss = 35.98874 (0.483 sec)\n",
      "INFO:tensorflow:loss = 35.98874, step = 67 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95405\n",
      "INFO:tensorflow:train_accuracy = 0.03698811, train_loss = 37.25161 (0.512 sec)\n",
      "INFO:tensorflow:loss = 37.25161, step = 68 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.44538\n",
      "INFO:tensorflow:train_accuracy = 0.025401069, train_loss = 34.060402 (0.409 sec)\n",
      "INFO:tensorflow:loss = 34.060402, step = 69 (0.409 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.64635\n",
      "INFO:tensorflow:train_accuracy = 0.02987013, train_loss = 34.145233 (0.378 sec)\n",
      "INFO:tensorflow:loss = 34.145233, step = 70 (0.378 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10505\n",
      "INFO:tensorflow:train_accuracy = 0.04447268, train_loss = 34.89531 (0.475 sec)\n",
      "INFO:tensorflow:loss = 34.89531, step = 71 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99525\n",
      "INFO:tensorflow:train_accuracy = 0.04890605, train_loss = 34.78028 (0.501 sec)\n",
      "INFO:tensorflow:loss = 34.78028, step = 72 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.28521\n",
      "INFO:tensorflow:train_accuracy = 0.0462585, train_loss = 31.57812 (0.437 sec)\n",
      "INFO:tensorflow:loss = 31.57812, step = 73 (0.437 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0122\n",
      "INFO:tensorflow:train_accuracy = 0.06641604, train_loss = 36.673447 (0.497 sec)\n",
      "INFO:tensorflow:loss = 36.673447, step = 74 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88794\n",
      "INFO:tensorflow:train_accuracy = 0.08974359, train_loss = 34.906475 (0.530 sec)\n",
      "INFO:tensorflow:loss = 34.906475, step = 75 (0.531 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90993\n",
      "INFO:tensorflow:train_accuracy = 0.056166057, train_loss = 36.9394 (0.523 sec)\n",
      "INFO:tensorflow:loss = 36.9394, step = 76 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9924\n",
      "INFO:tensorflow:train_accuracy = 0.045033112, train_loss = 29.71182 (0.502 sec)\n",
      "INFO:tensorflow:loss = 29.71182, step = 77 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0097\n",
      "INFO:tensorflow:train_accuracy = 0.051378448, train_loss = 36.559658 (0.498 sec)\n",
      "INFO:tensorflow:loss = 36.559658, step = 78 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0929\n",
      "INFO:tensorflow:train_accuracy = 0.04076433, train_loss = 34.416306 (0.478 sec)\n",
      "INFO:tensorflow:loss = 34.416306, step = 79 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.4213\n",
      "INFO:tensorflow:train_accuracy = 0.044155844, train_loss = 30.881483 (0.413 sec)\n",
      "INFO:tensorflow:loss = 30.881483, step = 80 (0.412 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.82311\n",
      "INFO:tensorflow:train_accuracy = 0.046936113, train_loss = 31.950916 (0.549 sec)\n",
      "INFO:tensorflow:loss = 31.950916, step = 81 (0.549 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95334\n",
      "INFO:tensorflow:train_accuracy = 0.06723716, train_loss = 35.696182 (0.512 sec)\n",
      "INFO:tensorflow:loss = 35.696182, step = 82 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14558\n",
      "INFO:tensorflow:train_accuracy = 0.11375, train_loss = 37.477093 (0.466 sec)\n",
      "INFO:tensorflow:loss = 37.477093, step = 83 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.38028\n",
      "INFO:tensorflow:train_accuracy = 0.055482168, train_loss = 30.922962 (0.420 sec)\n",
      "INFO:tensorflow:loss = 30.922962, step = 84 (0.420 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.27033\n",
      "INFO:tensorflow:train_accuracy = 0.043312103, train_loss = 32.309807 (0.441 sec)\n",
      "INFO:tensorflow:loss = 32.309807, step = 85 (0.441 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.38281\n",
      "INFO:tensorflow:train_accuracy = 0.046666667, train_loss = 28.924862 (0.420 sec)\n",
      "INFO:tensorflow:loss = 28.924862, step = 86 (0.420 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07647\n",
      "INFO:tensorflow:train_accuracy = 0.052311435, train_loss = 35.131157 (0.481 sec)\n",
      "INFO:tensorflow:loss = 35.131157, step = 87 (0.481 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10093\n",
      "INFO:tensorflow:train_accuracy = 0.10546875, train_loss = 30.466839 (0.476 sec)\n",
      "INFO:tensorflow:loss = 30.466839, step = 88 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98433\n",
      "INFO:tensorflow:train_accuracy = 0.07209613, train_loss = 29.009989 (0.504 sec)\n",
      "INFO:tensorflow:loss = 29.009989, step = 89 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97054\n",
      "INFO:tensorflow:train_accuracy = 0.110526316, train_loss = 30.885231 (0.507 sec)\n",
      "INFO:tensorflow:loss = 30.885231, step = 90 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.25422\n",
      "INFO:tensorflow:train_accuracy = 0.08675799, train_loss = 28.56562 (0.160 sec)\n",
      "INFO:tensorflow:loss = 28.56562, step = 91 (0.160 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99113\n",
      "INFO:tensorflow:train_accuracy = 0.11488251, train_loss = 29.25378 (0.503 sec)\n",
      "INFO:tensorflow:loss = 29.25378, step = 92 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8044\n",
      "INFO:tensorflow:train_accuracy = 0.07958922, train_loss = 29.485928 (0.554 sec)\n",
      "INFO:tensorflow:loss = 29.485928, step = 93 (0.553 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98299\n",
      "INFO:tensorflow:train_accuracy = 0.100728154, train_loss = 34.63394 (0.505 sec)\n",
      "INFO:tensorflow:loss = 34.63394, step = 94 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99635\n",
      "INFO:tensorflow:train_accuracy = 0.19148937, train_loss = 33.48763 (0.501 sec)\n",
      "INFO:tensorflow:loss = 33.48763, step = 95 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90464\n",
      "INFO:tensorflow:train_accuracy = 0.2972259, train_loss = 28.07332 (0.525 sec)\n",
      "INFO:tensorflow:loss = 28.07332, step = 96 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08804\n",
      "INFO:tensorflow:train_accuracy = 0.16883117, train_loss = 29.517097 (0.479 sec)\n",
      "INFO:tensorflow:loss = 29.517097, step = 97 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24691\n",
      "INFO:tensorflow:train_accuracy = 0.21419518, train_loss = 29.93724 (0.445 sec)\n",
      "INFO:tensorflow:loss = 29.93724, step = 98 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87493\n",
      "INFO:tensorflow:train_accuracy = 0.21343873, train_loss = 27.116478 (0.533 sec)\n",
      "INFO:tensorflow:loss = 27.116478, step = 99 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1014\n",
      "INFO:tensorflow:train_accuracy = 0.2160414, train_loss = 28.053684 (0.476 sec)\n",
      "INFO:tensorflow:loss = 28.053684, step = 100 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.47835\n",
      "INFO:tensorflow:train_accuracy = 0.34270415, train_loss = 27.16365 (0.403 sec)\n",
      "INFO:tensorflow:loss = 27.16365, step = 101 (0.403 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95545\n",
      "INFO:tensorflow:train_accuracy = 0.44042838, train_loss = 29.179304 (0.512 sec)\n",
      "INFO:tensorflow:loss = 29.179304, step = 102 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07081\n",
      "INFO:tensorflow:train_accuracy = 0.48645163, train_loss = 27.68938 (0.483 sec)\n",
      "INFO:tensorflow:loss = 27.68938, step = 103 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01986\n",
      "INFO:tensorflow:train_accuracy = 0.48966408, train_loss = 28.077806 (0.495 sec)\n",
      "INFO:tensorflow:loss = 28.077806, step = 104 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94777\n",
      "INFO:tensorflow:train_accuracy = 0.4189189, train_loss = 27.40237 (0.514 sec)\n",
      "INFO:tensorflow:loss = 27.40237, step = 105 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96198\n",
      "INFO:tensorflow:train_accuracy = 0.38113207, train_loss = 32.162262 (0.509 sec)\n",
      "INFO:tensorflow:loss = 32.162262, step = 106 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95655\n",
      "INFO:tensorflow:train_accuracy = 0.36657682, train_loss = 24.93024 (0.511 sec)\n",
      "INFO:tensorflow:loss = 24.93024, step = 107 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99928\n",
      "INFO:tensorflow:train_accuracy = 0.33039093, train_loss = 29.263489 (0.500 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 29.263489, step = 108 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01096\n",
      "INFO:tensorflow:train_accuracy = 0.3606341, train_loss = 27.04092 (0.497 sec)\n",
      "INFO:tensorflow:loss = 27.04092, step = 109 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87763\n",
      "INFO:tensorflow:train_accuracy = 0.449935, train_loss = 27.962086 (0.533 sec)\n",
      "INFO:tensorflow:loss = 27.962086, step = 110 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93189\n",
      "INFO:tensorflow:train_accuracy = 0.49132177, train_loss = 27.491446 (0.518 sec)\n",
      "INFO:tensorflow:loss = 27.491446, step = 111 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14362\n",
      "INFO:tensorflow:train_accuracy = 0.43960148, train_loss = 31.080612 (0.466 sec)\n",
      "INFO:tensorflow:loss = 31.080612, step = 112 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18577\n",
      "INFO:tensorflow:train_accuracy = 0.5111402, train_loss = 25.162666 (0.458 sec)\n",
      "INFO:tensorflow:loss = 25.162666, step = 113 (0.458 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02662\n",
      "INFO:tensorflow:train_accuracy = 0.4837873, train_loss = 27.419075 (0.494 sec)\n",
      "INFO:tensorflow:loss = 27.419075, step = 114 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93306\n",
      "INFO:tensorflow:train_accuracy = 0.47272727, train_loss = 26.696947 (0.517 sec)\n",
      "INFO:tensorflow:loss = 26.696947, step = 115 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80934\n",
      "INFO:tensorflow:train_accuracy = 0.50199205, train_loss = 24.791067 (0.553 sec)\n",
      "INFO:tensorflow:loss = 24.791067, step = 116 (0.552 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.36036\n",
      "INFO:tensorflow:train_accuracy = 0.4713715, train_loss = 24.58071 (0.424 sec)\n",
      "INFO:tensorflow:loss = 24.58071, step = 117 (0.424 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86257\n",
      "INFO:tensorflow:train_accuracy = 0.49809402, train_loss = 28.257198 (0.537 sec)\n",
      "INFO:tensorflow:loss = 28.257198, step = 118 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91622\n",
      "INFO:tensorflow:train_accuracy = 0.53188604, train_loss = 24.704346 (0.522 sec)\n",
      "INFO:tensorflow:loss = 24.704346, step = 119 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92202\n",
      "INFO:tensorflow:train_accuracy = 0.47544912, train_loss = 33.50606 (0.520 sec)\n",
      "INFO:tensorflow:loss = 33.50606, step = 120 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87561\n",
      "INFO:tensorflow:train_accuracy = 0.47058824, train_loss = 27.279325 (0.534 sec)\n",
      "INFO:tensorflow:loss = 27.279325, step = 121 (0.534 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84688\n",
      "INFO:tensorflow:train_accuracy = 0.4947644, train_loss = 25.670364 (0.541 sec)\n",
      "INFO:tensorflow:loss = 25.670364, step = 122 (0.541 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.7286\n",
      "INFO:tensorflow:train_accuracy = 0.55333334, train_loss = 24.217957 (0.579 sec)\n",
      "INFO:tensorflow:loss = 24.217957, step = 123 (0.579 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.73391\n",
      "INFO:tensorflow:train_accuracy = 0.48030496, train_loss = 28.493114 (0.576 sec)\n",
      "INFO:tensorflow:loss = 28.493114, step = 124 (0.575 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90204\n",
      "INFO:tensorflow:train_accuracy = 0.48192772, train_loss = 30.09999 (0.526 sec)\n",
      "INFO:tensorflow:loss = 30.09999, step = 125 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04138\n",
      "INFO:tensorflow:train_accuracy = 0.50583655, train_loss = 32.836838 (0.490 sec)\n",
      "INFO:tensorflow:loss = 32.836838, step = 126 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86627\n",
      "INFO:tensorflow:train_accuracy = 0.5042424, train_loss = 31.832924 (0.536 sec)\n",
      "INFO:tensorflow:loss = 31.832924, step = 127 (0.536 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86463\n",
      "INFO:tensorflow:train_accuracy = 0.50626564, train_loss = 27.889744 (0.536 sec)\n",
      "INFO:tensorflow:loss = 27.889744, step = 128 (0.536 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9711\n",
      "INFO:tensorflow:train_accuracy = 0.54847276, train_loss = 26.196657 (0.507 sec)\n",
      "INFO:tensorflow:loss = 26.196657, step = 129 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92263\n",
      "INFO:tensorflow:train_accuracy = 0.48320413, train_loss = 26.193642 (0.520 sec)\n",
      "INFO:tensorflow:loss = 26.193642, step = 130 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.37492\n",
      "INFO:tensorflow:train_accuracy = 0.496063, train_loss = 25.493832 (0.421 sec)\n",
      "INFO:tensorflow:loss = 25.493832, step = 131 (0.421 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87574\n",
      "INFO:tensorflow:train_accuracy = 0.5075567, train_loss = 28.95036 (0.533 sec)\n",
      "INFO:tensorflow:loss = 28.95036, step = 132 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87485\n",
      "INFO:tensorflow:train_accuracy = 0.4993598, train_loss = 26.78411 (0.534 sec)\n",
      "INFO:tensorflow:loss = 26.78411, step = 133 (0.534 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96415\n",
      "INFO:tensorflow:train_accuracy = 0.50502515, train_loss = 27.82849 (0.509 sec)\n",
      "INFO:tensorflow:loss = 27.82849, step = 134 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95117\n",
      "INFO:tensorflow:train_accuracy = 0.55826557, train_loss = 22.21254 (0.512 sec)\n",
      "INFO:tensorflow:loss = 22.21254, step = 135 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97046\n",
      "INFO:tensorflow:train_accuracy = 0.5235532, train_loss = 24.65918 (0.508 sec)\n",
      "INFO:tensorflow:loss = 24.65918, step = 136 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.91216\n",
      "INFO:tensorflow:train_accuracy = 0.38723403, train_loss = 30.405416 (0.169 sec)\n",
      "INFO:tensorflow:loss = 30.405416, step = 137 (0.169 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.71373\n",
      "INFO:tensorflow:train_accuracy = 0.5330688, train_loss = 24.462086 (0.584 sec)\n",
      "INFO:tensorflow:loss = 24.462086, step = 138 (0.584 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14601\n",
      "INFO:tensorflow:train_accuracy = 0.46878824, train_loss = 29.177868 (0.466 sec)\n",
      "INFO:tensorflow:loss = 29.177868, step = 139 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86087\n",
      "INFO:tensorflow:train_accuracy = 0.5031211, train_loss = 31.087126 (0.537 sec)\n",
      "INFO:tensorflow:loss = 31.087126, step = 140 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.81796\n",
      "INFO:tensorflow:train_accuracy = 0.4942821, train_loss = 26.300816 (0.550 sec)\n",
      "INFO:tensorflow:loss = 26.300816, step = 141 (0.550 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83375\n",
      "INFO:tensorflow:train_accuracy = 0.52217996, train_loss = 28.420208 (0.545 sec)\n",
      "INFO:tensorflow:loss = 28.420208, step = 142 (0.544 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01703\n",
      "INFO:tensorflow:train_accuracy = 0.51184213, train_loss = 23.961441 (0.496 sec)\n",
      "INFO:tensorflow:loss = 23.961441, step = 143 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90524\n",
      "INFO:tensorflow:train_accuracy = 0.5123537, train_loss = 25.691536 (0.525 sec)\n",
      "INFO:tensorflow:loss = 25.691536, step = 144 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13918\n",
      "INFO:tensorflow:train_accuracy = 0.4641944, train_loss = 26.967665 (0.468 sec)\n",
      "INFO:tensorflow:loss = 26.967665, step = 145 (0.468 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21197\n",
      "INFO:tensorflow:train_accuracy = 0.4855643, train_loss = 31.171885 (0.452 sec)\n",
      "INFO:tensorflow:loss = 31.171885, step = 146 (0.452 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84852\n",
      "INFO:tensorflow:train_accuracy = 0.51216215, train_loss = 23.508577 (0.541 sec)\n",
      "INFO:tensorflow:loss = 23.508577, step = 147 (0.541 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98163\n",
      "INFO:tensorflow:train_accuracy = 0.49066666, train_loss = 25.919252 (0.504 sec)\n",
      "INFO:tensorflow:loss = 25.919252, step = 148 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07053\n",
      "INFO:tensorflow:train_accuracy = 0.46640316, train_loss = 24.305145 (0.483 sec)\n",
      "INFO:tensorflow:loss = 24.305145, step = 149 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95264\n",
      "INFO:tensorflow:train_accuracy = 0.5045045, train_loss = 26.837296 (0.512 sec)\n",
      "INFO:tensorflow:loss = 26.837296, step = 150 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97454\n",
      "INFO:tensorflow:train_accuracy = 0.4717477, train_loss = 28.298721 (0.506 sec)\n",
      "INFO:tensorflow:loss = 28.298721, step = 151 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12168\n",
      "INFO:tensorflow:train_accuracy = 0.48262033, train_loss = 26.303406 (0.471 sec)\n",
      "INFO:tensorflow:loss = 26.303406, step = 152 (0.471 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.27143\n",
      "INFO:tensorflow:train_accuracy = 0.49811795, train_loss = 29.36441 (0.440 sec)\n",
      "INFO:tensorflow:loss = 29.36441, step = 153 (0.440 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13407\n",
      "INFO:tensorflow:train_accuracy = 0.50775194, train_loss = 27.0293 (0.468 sec)\n",
      "INFO:tensorflow:loss = 27.0293, step = 154 (0.468 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:global_step/sec: 1.84593\n",
      "INFO:tensorflow:train_accuracy = 0.5012854, train_loss = 25.363762 (0.542 sec)\n",
      "INFO:tensorflow:loss = 25.363762, step = 155 (0.542 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.77019\n",
      "INFO:tensorflow:train_accuracy = 0.5313352, train_loss = 21.51256 (0.565 sec)\n",
      "INFO:tensorflow:loss = 21.51256, step = 156 (0.564 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.22995\n",
      "INFO:tensorflow:train_accuracy = 0.48995984, train_loss = 25.39513 (0.448 sec)\n",
      "INFO:tensorflow:loss = 25.39513, step = 157 (0.449 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04456\n",
      "INFO:tensorflow:train_accuracy = 0.49210206, train_loss = 40.815117 (0.489 sec)\n",
      "INFO:tensorflow:loss = 40.815117, step = 158 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.78673\n",
      "INFO:tensorflow:train_accuracy = 0.529563, train_loss = 27.555763 (0.560 sec)\n",
      "INFO:tensorflow:loss = 27.555763, step = 159 (0.560 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93738\n",
      "INFO:tensorflow:train_accuracy = 0.50530505, train_loss = 23.255363 (0.516 sec)\n",
      "INFO:tensorflow:loss = 23.255363, step = 160 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.28569\n",
      "INFO:tensorflow:train_accuracy = 0.49935317, train_loss = 26.965736 (0.438 sec)\n",
      "INFO:tensorflow:loss = 26.965736, step = 161 (0.438 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06997\n",
      "INFO:tensorflow:train_accuracy = 0.5465425, train_loss = 24.236313 (0.483 sec)\n",
      "INFO:tensorflow:loss = 24.236313, step = 162 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97974\n",
      "INFO:tensorflow:train_accuracy = 0.52988046, train_loss = 23.63542 (0.505 sec)\n",
      "INFO:tensorflow:loss = 23.63542, step = 163 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23226\n",
      "INFO:tensorflow:train_accuracy = 0.51937985, train_loss = 23.810226 (0.448 sec)\n",
      "INFO:tensorflow:loss = 23.810226, step = 164 (0.449 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9063\n",
      "INFO:tensorflow:train_accuracy = 0.5272497, train_loss = 27.152328 (0.524 sec)\n",
      "INFO:tensorflow:loss = 27.152328, step = 165 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85344\n",
      "INFO:tensorflow:train_accuracy = 0.50753766, train_loss = 29.08688 (0.540 sec)\n",
      "INFO:tensorflow:loss = 29.08688, step = 166 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09809\n",
      "INFO:tensorflow:train_accuracy = 0.54593176, train_loss = 23.805458 (0.477 sec)\n",
      "INFO:tensorflow:loss = 23.805458, step = 167 (0.477 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24667\n",
      "INFO:tensorflow:train_accuracy = 0.538262, train_loss = 26.465647 (0.445 sec)\n",
      "INFO:tensorflow:loss = 26.465647, step = 168 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97929\n",
      "INFO:tensorflow:train_accuracy = 0.5372751, train_loss = 25.534643 (0.505 sec)\n",
      "INFO:tensorflow:loss = 25.534643, step = 169 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13776\n",
      "INFO:tensorflow:train_accuracy = 0.53128994, train_loss = 27.314032 (0.468 sec)\n",
      "INFO:tensorflow:loss = 27.314032, step = 170 (0.468 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90162\n",
      "INFO:tensorflow:train_accuracy = 0.5193026, train_loss = 29.763172 (0.526 sec)\n",
      "INFO:tensorflow:loss = 29.763172, step = 171 (0.526 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0181\n",
      "INFO:tensorflow:train_accuracy = 0.534018, train_loss = 23.728981 (0.495 sec)\n",
      "INFO:tensorflow:loss = 23.728981, step = 172 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.70358\n",
      "INFO:tensorflow:train_accuracy = 0.5, train_loss = 29.59895 (0.587 sec)\n",
      "INFO:tensorflow:loss = 29.59895, step = 173 (0.587 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15246\n",
      "INFO:tensorflow:train_accuracy = 0.52125, train_loss = 30.968512 (0.465 sec)\n",
      "INFO:tensorflow:loss = 30.968512, step = 174 (0.465 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10544\n",
      "INFO:tensorflow:train_accuracy = 0.53737116, train_loss = 24.129997 (0.475 sec)\n",
      "INFO:tensorflow:loss = 24.129997, step = 175 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.28454\n",
      "INFO:tensorflow:train_accuracy = 0.5437909, train_loss = 25.600739 (0.438 sec)\n",
      "INFO:tensorflow:loss = 25.600739, step = 176 (0.438 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03996\n",
      "INFO:tensorflow:train_accuracy = 0.5178117, train_loss = 25.875225 (0.490 sec)\n",
      "INFO:tensorflow:loss = 25.875225, step = 177 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94343\n",
      "INFO:tensorflow:train_accuracy = 0.5265151, train_loss = 26.57745 (0.515 sec)\n",
      "INFO:tensorflow:loss = 26.57745, step = 178 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.34516\n",
      "INFO:tensorflow:train_accuracy = 0.5225225, train_loss = 26.252274 (0.426 sec)\n",
      "INFO:tensorflow:loss = 26.252274, step = 179 (0.426 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.81486\n",
      "INFO:tensorflow:train_accuracy = 0.49679896, train_loss = 28.482304 (0.551 sec)\n",
      "INFO:tensorflow:loss = 28.482304, step = 180 (0.551 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09771\n",
      "INFO:tensorflow:train_accuracy = 0.55390835, train_loss = 21.586687 (0.477 sec)\n",
      "INFO:tensorflow:loss = 21.586687, step = 181 (0.477 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01087\n",
      "INFO:tensorflow:train_accuracy = 0.53571427, train_loss = 24.080921 (0.497 sec)\n",
      "INFO:tensorflow:loss = 24.080921, step = 182 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.71959\n",
      "INFO:tensorflow:train_accuracy = 0.5625, train_loss = 20.091438 (0.149 sec)\n",
      "INFO:tensorflow:loss = 20.091438, step = 183 (0.149 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.76557\n",
      "INFO:tensorflow:train_accuracy = 0.53945667, train_loss = 23.672928 (0.567 sec)\n",
      "INFO:tensorflow:loss = 23.672928, step = 184 (0.567 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95921\n",
      "INFO:tensorflow:train_accuracy = 0.52130324, train_loss = 28.100956 (0.510 sec)\n",
      "INFO:tensorflow:loss = 28.100956, step = 185 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.69409\n",
      "INFO:tensorflow:train_accuracy = 0.5148515, train_loss = 27.868935 (0.590 sec)\n",
      "INFO:tensorflow:loss = 27.868935, step = 186 (0.591 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.22327\n",
      "INFO:tensorflow:train_accuracy = 0.52926207, train_loss = 24.98473 (0.449 sec)\n",
      "INFO:tensorflow:loss = 24.98473, step = 187 (0.449 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23386\n",
      "INFO:tensorflow:train_accuracy = 0.53290325, train_loss = 25.433632 (0.448 sec)\n",
      "INFO:tensorflow:loss = 25.433632, step = 188 (0.449 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88586\n",
      "INFO:tensorflow:train_accuracy = 0.5479632, train_loss = 23.81685 (0.530 sec)\n",
      "INFO:tensorflow:loss = 23.81685, step = 189 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94819\n",
      "INFO:tensorflow:train_accuracy = 0.5313316, train_loss = 25.550385 (0.514 sec)\n",
      "INFO:tensorflow:loss = 25.550385, step = 190 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.30779\n",
      "INFO:tensorflow:train_accuracy = 0.54652685, train_loss = 23.796959 (0.433 sec)\n",
      "INFO:tensorflow:loss = 23.796959, step = 191 (0.433 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90278\n",
      "INFO:tensorflow:train_accuracy = 0.534018, train_loss = 26.272541 (0.525 sec)\n",
      "INFO:tensorflow:loss = 26.272541, step = 192 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.82395\n",
      "INFO:tensorflow:train_accuracy = 0.5437909, train_loss = 23.95362 (0.548 sec)\n",
      "INFO:tensorflow:loss = 23.95362, step = 193 (0.549 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98896\n",
      "INFO:tensorflow:train_accuracy = 0.5329815, train_loss = 22.864567 (0.503 sec)\n",
      "INFO:tensorflow:loss = 22.864567, step = 194 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06466\n",
      "INFO:tensorflow:train_accuracy = 0.5274869, train_loss = 23.380917 (0.484 sec)\n",
      "INFO:tensorflow:loss = 23.380917, step = 195 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2554\n",
      "INFO:tensorflow:train_accuracy = 0.51846963, train_loss = 23.641136 (0.444 sec)\n",
      "INFO:tensorflow:loss = 23.641136, step = 196 (0.444 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06261\n",
      "INFO:tensorflow:train_accuracy = 0.54521275, train_loss = 21.943083 (0.485 sec)\n",
      "INFO:tensorflow:loss = 21.943083, step = 197 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1923\n",
      "INFO:tensorflow:train_accuracy = 0.49331552, train_loss = 23.141731 (0.456 sec)\n",
      "INFO:tensorflow:loss = 23.141731, step = 198 (0.456 sec)\n",
      "INFO:tensorflow:Saving checkpoints for 200 into asr-minijasper/model.ckpt.\n",
      "INFO:tensorflow:Calling model_fn.\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/ctc/metrics.py:58: The name tf.metrics.mean is deprecated. Please use tf.compat.v1.metrics.mean instead.\n",
      "\n",
      "WARNING:tensorflow:From /home/husein/malaya-speech/malaya_speech/train/model/ctc/metrics.py:106: The name tf.regex_replace is deprecated. Please use tf.strings.regex_replace instead.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Starting evaluation at 2020-10-25T19:10:02Z\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Restoring parameters from asr-minijasper/model.ckpt-200\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n",
      "INFO:tensorflow:Evaluation [1/10]\n",
      "INFO:tensorflow:Evaluation [2/10]\n",
      "INFO:tensorflow:Evaluation [3/10]\n",
      "INFO:tensorflow:Evaluation [4/10]\n",
      "INFO:tensorflow:Evaluation [5/10]\n",
      "INFO:tensorflow:Evaluation [6/10]\n",
      "INFO:tensorflow:Evaluation [7/10]\n",
      "INFO:tensorflow:Evaluation [8/10]\n",
      "INFO:tensorflow:Evaluation [9/10]\n",
      "INFO:tensorflow:Evaluation [10/10]\n",
      "INFO:tensorflow:Finished evaluation at 2020-10-25-19:10:07\n",
      "INFO:tensorflow:Saving dict for global step 200: WER = 1.0, accuracy = 0.55828226, global_step = 200, loss = 23.848358\n",
      "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 200: asr-minijasper/model.ckpt-200\n",
      "INFO:tensorflow:global_step/sec: 0.0933012\n",
      "INFO:tensorflow:train_accuracy = 0.5146871, train_loss = 24.983624 (10.717 sec)\n",
      "INFO:tensorflow:loss = 24.983624, step = 199 (10.717 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10122\n",
      "INFO:tensorflow:train_accuracy = 0.5347044, train_loss = 24.560173 (0.477 sec)\n",
      "INFO:tensorflow:loss = 24.560173, step = 200 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16301\n",
      "INFO:tensorflow:train_accuracy = 0.534018, train_loss = 25.162354 (0.462 sec)\n",
      "INFO:tensorflow:loss = 25.162354, step = 201 (0.461 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09149\n",
      "INFO:tensorflow:train_accuracy = 0.48339975, train_loss = 20.886059 (0.478 sec)\n",
      "INFO:tensorflow:loss = 20.886059, step = 202 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04314\n",
      "INFO:tensorflow:train_accuracy = 0.5267974, train_loss = 26.104502 (0.490 sec)\n",
      "INFO:tensorflow:loss = 26.104502, step = 203 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00787\n",
      "INFO:tensorflow:train_accuracy = 0.53806454, train_loss = 24.820435 (0.498 sec)\n",
      "INFO:tensorflow:loss = 24.820435, step = 204 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89408\n",
      "INFO:tensorflow:train_accuracy = 0.53667957, train_loss = 24.369816 (0.528 sec)\n",
      "INFO:tensorflow:loss = 24.369816, step = 205 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16912\n",
      "INFO:tensorflow:train_accuracy = 0.5359897, train_loss = 24.948566 (0.461 sec)\n",
      "INFO:tensorflow:loss = 24.948566, step = 206 (0.461 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95472\n",
      "INFO:tensorflow:train_accuracy = 0.5561497, train_loss = 22.480593 (0.511 sec)\n",
      "INFO:tensorflow:loss = 22.480593, step = 207 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98942\n",
      "INFO:tensorflow:train_accuracy = 0.52791876, train_loss = 24.317387 (0.503 sec)\n",
      "INFO:tensorflow:loss = 24.317387, step = 208 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9706\n",
      "INFO:tensorflow:train_accuracy = 0.57734805, train_loss = 21.348259 (0.508 sec)\n",
      "INFO:tensorflow:loss = 21.348259, step = 209 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03005\n",
      "INFO:tensorflow:train_accuracy = 0.54808956, train_loss = 21.740454 (0.493 sec)\n",
      "INFO:tensorflow:loss = 21.740454, step = 210 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00218\n",
      "INFO:tensorflow:train_accuracy = 0.50375, train_loss = 25.077614 (0.499 sec)\n",
      "INFO:tensorflow:loss = 25.077614, step = 211 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01705\n",
      "INFO:tensorflow:train_accuracy = 0.5299363, train_loss = 24.148499 (0.496 sec)\n",
      "INFO:tensorflow:loss = 24.148499, step = 212 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87632\n",
      "INFO:tensorflow:train_accuracy = 0.5148515, train_loss = 28.527988 (0.533 sec)\n",
      "INFO:tensorflow:loss = 28.527988, step = 213 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03864\n",
      "INFO:tensorflow:train_accuracy = 0.5388601, train_loss = 25.074417 (0.490 sec)\n",
      "INFO:tensorflow:loss = 25.074417, step = 214 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86276\n",
      "INFO:tensorflow:train_accuracy = 0.5401554, train_loss = 23.26139 (0.537 sec)\n",
      "INFO:tensorflow:loss = 23.26139, step = 215 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89648\n",
      "INFO:tensorflow:train_accuracy = 0.5265151, train_loss = 25.37247 (0.527 sec)\n",
      "INFO:tensorflow:loss = 25.37247, step = 216 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07484\n",
      "INFO:tensorflow:train_accuracy = 0.5232704, train_loss = 26.182932 (0.482 sec)\n",
      "INFO:tensorflow:loss = 26.182932, step = 217 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8249\n",
      "INFO:tensorflow:train_accuracy = 0.52261305, train_loss = 26.362236 (0.548 sec)\n",
      "INFO:tensorflow:loss = 26.362236, step = 218 (0.549 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95416\n",
      "INFO:tensorflow:train_accuracy = 0.55172414, train_loss = 21.262903 (0.512 sec)\n",
      "INFO:tensorflow:loss = 21.262903, step = 219 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10177\n",
      "INFO:tensorflow:train_accuracy = 0.50793654, train_loss = 28.768776 (0.475 sec)\n",
      "INFO:tensorflow:loss = 28.768776, step = 220 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98595\n",
      "INFO:tensorflow:train_accuracy = 0.5193508, train_loss = 25.426113 (0.504 sec)\n",
      "INFO:tensorflow:loss = 25.426113, step = 221 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94378\n",
      "INFO:tensorflow:train_accuracy = 0.5530504, train_loss = 22.451971 (0.514 sec)\n",
      "INFO:tensorflow:loss = 22.451971, step = 222 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02277\n",
      "INFO:tensorflow:train_accuracy = 0.5299363, train_loss = 25.293451 (0.495 sec)\n",
      "INFO:tensorflow:loss = 25.293451, step = 223 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99591\n",
      "INFO:tensorflow:train_accuracy = 0.54545456, train_loss = 23.194794 (0.501 sec)\n",
      "INFO:tensorflow:loss = 23.194794, step = 224 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00733\n",
      "INFO:tensorflow:train_accuracy = 0.521197, train_loss = 25.128677 (0.498 sec)\n",
      "INFO:tensorflow:loss = 25.128677, step = 225 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96902\n",
      "INFO:tensorflow:train_accuracy = 0.52926207, train_loss = 24.211334 (0.508 sec)\n",
      "INFO:tensorflow:loss = 24.211334, step = 226 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9376\n",
      "INFO:tensorflow:train_accuracy = 0.54593176, train_loss = 23.717175 (0.516 sec)\n",
      "INFO:tensorflow:loss = 23.717175, step = 227 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8394\n",
      "INFO:tensorflow:train_accuracy = 0.5416116, train_loss = 23.134804 (0.544 sec)\n",
      "INFO:tensorflow:loss = 23.134804, step = 228 (0.545 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.81845\n",
      "INFO:tensorflow:train_accuracy = 0.5625, train_loss = 20.101585 (0.172 sec)\n",
      "INFO:tensorflow:loss = 20.101585, step = 229 (0.171 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97684\n",
      "INFO:tensorflow:train_accuracy = 0.55245686, train_loss = 18.932888 (0.506 sec)\n",
      "INFO:tensorflow:loss = 18.932888, step = 230 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07726\n",
      "INFO:tensorflow:train_accuracy = 0.49026763, train_loss = 29.806269 (0.482 sec)\n",
      "INFO:tensorflow:loss = 29.806269, step = 231 (0.481 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.823\n",
      "INFO:tensorflow:train_accuracy = 0.51358026, train_loss = 28.41264 (0.548 sec)\n",
      "INFO:tensorflow:loss = 28.41264, step = 232 (0.548 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95222\n",
      "INFO:tensorflow:train_accuracy = 0.53128994, train_loss = 23.96789 (0.512 sec)\n",
      "INFO:tensorflow:loss = 23.96789, step = 233 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86937\n",
      "INFO:tensorflow:train_accuracy = 0.53938586, train_loss = 21.19197 (0.535 sec)\n",
      "INFO:tensorflow:loss = 21.19197, step = 234 (0.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03005\n",
      "INFO:tensorflow:train_accuracy = 0.54817706, train_loss = 23.28476 (0.493 sec)\n",
      "INFO:tensorflow:loss = 23.28476, step = 235 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96766\n",
      "INFO:tensorflow:train_accuracy = 0.5423729, train_loss = 23.13093 (0.508 sec)\n",
      "INFO:tensorflow:loss = 23.13093, step = 236 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09073\n",
      "INFO:tensorflow:train_accuracy = 0.54155844, train_loss = 24.369368 (0.478 sec)\n",
      "INFO:tensorflow:loss = 24.369368, step = 237 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89652\n",
      "INFO:tensorflow:train_accuracy = 0.52125, train_loss = 27.402975 (0.527 sec)\n",
      "INFO:tensorflow:loss = 27.402975, step = 238 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23948\n",
      "INFO:tensorflow:train_accuracy = 0.55452126, train_loss = 21.61861 (0.447 sec)\n",
      "INFO:tensorflow:loss = 21.61861, step = 239 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90711\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.5595716, train_loss = 21.238106 (0.524 sec)\n",
      "INFO:tensorflow:loss = 21.238106, step = 240 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.26807\n",
      "INFO:tensorflow:train_accuracy = 0.5488127, train_loss = 21.097755 (0.441 sec)\n",
      "INFO:tensorflow:loss = 21.097755, step = 241 (0.441 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94952\n",
      "INFO:tensorflow:train_accuracy = 0.5472441, train_loss = 22.3786 (0.513 sec)\n",
      "INFO:tensorflow:loss = 22.3786, step = 242 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96304\n",
      "INFO:tensorflow:train_accuracy = 0.526658, train_loss = 23.868479 (0.509 sec)\n",
      "INFO:tensorflow:loss = 23.868479, step = 243 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88072\n",
      "INFO:tensorflow:train_accuracy = 0.523212, train_loss = 26.576149 (0.532 sec)\n",
      "INFO:tensorflow:loss = 26.576149, step = 244 (0.532 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93794\n",
      "INFO:tensorflow:train_accuracy = 0.562753, train_loss = 19.914238 (0.516 sec)\n",
      "INFO:tensorflow:loss = 19.914238, step = 245 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00588\n",
      "INFO:tensorflow:train_accuracy = 0.5423729, train_loss = 22.933659 (0.499 sec)\n",
      "INFO:tensorflow:loss = 22.933659, step = 246 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92139\n",
      "INFO:tensorflow:train_accuracy = 0.5423729, train_loss = 21.91372 (0.520 sec)\n",
      "INFO:tensorflow:loss = 21.91372, step = 247 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0236\n",
      "INFO:tensorflow:train_accuracy = 0.5285714, train_loss = 22.906141 (0.494 sec)\n",
      "INFO:tensorflow:loss = 22.906141, step = 248 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91953\n",
      "INFO:tensorflow:train_accuracy = 0.5246753, train_loss = 26.67718 (0.521 sec)\n",
      "INFO:tensorflow:loss = 26.67718, step = 249 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90278\n",
      "INFO:tensorflow:train_accuracy = 0.5436767, train_loss = 23.415367 (0.526 sec)\n",
      "INFO:tensorflow:loss = 23.415367, step = 250 (0.526 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90821\n",
      "INFO:tensorflow:train_accuracy = 0.539559, train_loss = 22.626877 (0.524 sec)\n",
      "INFO:tensorflow:loss = 22.626877, step = 251 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.30512\n",
      "INFO:tensorflow:train_accuracy = 0.55291003, train_loss = 22.108795 (0.434 sec)\n",
      "INFO:tensorflow:loss = 22.108795, step = 252 (0.434 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83858\n",
      "INFO:tensorflow:train_accuracy = 0.5423729, train_loss = 23.238556 (0.544 sec)\n",
      "INFO:tensorflow:loss = 23.238556, step = 253 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92288\n",
      "INFO:tensorflow:train_accuracy = 0.5347044, train_loss = 23.054352 (0.520 sec)\n",
      "INFO:tensorflow:loss = 23.054352, step = 254 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95929\n",
      "INFO:tensorflow:train_accuracy = 0.5107731, train_loss = 22.976673 (0.510 sec)\n",
      "INFO:tensorflow:loss = 22.976673, step = 255 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99023\n",
      "INFO:tensorflow:train_accuracy = 0.55245686, train_loss = 22.422703 (0.503 sec)\n",
      "INFO:tensorflow:loss = 22.422703, step = 256 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01114\n",
      "INFO:tensorflow:train_accuracy = 0.54438645, train_loss = 22.262274 (0.497 sec)\n",
      "INFO:tensorflow:loss = 22.262274, step = 257 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84128\n",
      "INFO:tensorflow:train_accuracy = 0.5180124, train_loss = 26.32161 (0.543 sec)\n",
      "INFO:tensorflow:loss = 26.32161, step = 258 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98463\n",
      "INFO:tensorflow:train_accuracy = 0.54438645, train_loss = 22.781101 (0.504 sec)\n",
      "INFO:tensorflow:loss = 22.781101, step = 259 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.82478\n",
      "INFO:tensorflow:train_accuracy = 0.5206508, train_loss = 25.960897 (0.548 sec)\n",
      "INFO:tensorflow:loss = 25.960897, step = 260 (0.548 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86327\n",
      "INFO:tensorflow:train_accuracy = 0.55217963, train_loss = 21.010235 (0.537 sec)\n",
      "INFO:tensorflow:loss = 21.010235, step = 261 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97578\n",
      "INFO:tensorflow:train_accuracy = 0.5353925, train_loss = 22.591011 (0.506 sec)\n",
      "INFO:tensorflow:loss = 22.591011, step = 262 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91414\n",
      "INFO:tensorflow:train_accuracy = 0.5167702, train_loss = 24.933334 (0.523 sec)\n",
      "INFO:tensorflow:loss = 24.933334, step = 263 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98275\n",
      "INFO:tensorflow:train_accuracy = 0.53324807, train_loss = 23.863075 (0.504 sec)\n",
      "INFO:tensorflow:loss = 23.863075, step = 264 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83232\n",
      "INFO:tensorflow:train_accuracy = 0.5312102, train_loss = 26.166813 (0.546 sec)\n",
      "INFO:tensorflow:loss = 26.166813, step = 265 (0.545 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87292\n",
      "INFO:tensorflow:train_accuracy = 0.51865673, train_loss = 26.864492 (0.534 sec)\n",
      "INFO:tensorflow:loss = 26.864492, step = 266 (0.534 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23681\n",
      "INFO:tensorflow:train_accuracy = 0.53248405, train_loss = 23.107563 (0.447 sec)\n",
      "INFO:tensorflow:loss = 23.107563, step = 267 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06559\n",
      "INFO:tensorflow:train_accuracy = 0.539559, train_loss = 23.872486 (0.484 sec)\n",
      "INFO:tensorflow:loss = 23.872486, step = 268 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00572\n",
      "INFO:tensorflow:train_accuracy = 0.5374677, train_loss = 23.002785 (0.498 sec)\n",
      "INFO:tensorflow:loss = 23.002785, step = 269 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83269\n",
      "INFO:tensorflow:train_accuracy = 0.5422627, train_loss = 23.89658 (0.546 sec)\n",
      "INFO:tensorflow:loss = 23.89658, step = 270 (0.546 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10869\n",
      "INFO:tensorflow:train_accuracy = 0.5290404, train_loss = 25.648514 (0.474 sec)\n",
      "INFO:tensorflow:loss = 25.648514, step = 271 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95112\n",
      "INFO:tensorflow:train_accuracy = 0.5278481, train_loss = 23.672064 (0.513 sec)\n",
      "INFO:tensorflow:loss = 23.672064, step = 272 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97213\n",
      "INFO:tensorflow:train_accuracy = 0.5302632, train_loss = 22.115698 (0.507 sec)\n",
      "INFO:tensorflow:loss = 22.115698, step = 273 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01557\n",
      "INFO:tensorflow:train_accuracy = 0.54664916, train_loss = 21.736559 (0.496 sec)\n",
      "INFO:tensorflow:loss = 21.736559, step = 274 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.69003\n",
      "INFO:tensorflow:train_accuracy = 0.56796116, train_loss = 20.66366 (0.150 sec)\n",
      "INFO:tensorflow:loss = 20.66366, step = 275 (0.150 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85876\n",
      "INFO:tensorflow:train_accuracy = 0.5488127, train_loss = 20.1858 (0.538 sec)\n",
      "INFO:tensorflow:loss = 20.1858, step = 276 (0.538 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99\n",
      "INFO:tensorflow:train_accuracy = 0.5135468, train_loss = 27.618649 (0.502 sec)\n",
      "INFO:tensorflow:loss = 27.618649, step = 277 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80995\n",
      "INFO:tensorflow:train_accuracy = 0.52791876, train_loss = 25.448622 (0.553 sec)\n",
      "INFO:tensorflow:loss = 25.448622, step = 278 (0.553 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96661\n",
      "INFO:tensorflow:train_accuracy = 0.51421505, train_loss = 26.39189 (0.508 sec)\n",
      "INFO:tensorflow:loss = 26.39189, step = 279 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02151\n",
      "INFO:tensorflow:train_accuracy = 0.5345268, train_loss = 24.332024 (0.494 sec)\n",
      "INFO:tensorflow:loss = 24.332024, step = 280 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91887\n",
      "INFO:tensorflow:train_accuracy = 0.55764073, train_loss = 20.57592 (0.521 sec)\n",
      "INFO:tensorflow:loss = 20.57592, step = 281 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98815\n",
      "INFO:tensorflow:train_accuracy = 0.5471204, train_loss = 22.4365 (0.503 sec)\n",
      "INFO:tensorflow:loss = 22.4365, step = 282 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03382\n",
      "INFO:tensorflow:train_accuracy = 0.5271808, train_loss = 23.439089 (0.492 sec)\n",
      "INFO:tensorflow:loss = 23.439089, step = 283 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17243\n",
      "INFO:tensorflow:train_accuracy = 0.540856, train_loss = 22.398922 (0.461 sec)\n",
      "INFO:tensorflow:loss = 22.398922, step = 284 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91855\n",
      "INFO:tensorflow:train_accuracy = 0.5344828, train_loss = 21.946547 (0.521 sec)\n",
      "INFO:tensorflow:loss = 21.946547, step = 285 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01725\n",
      "INFO:tensorflow:train_accuracy = 0.55748665, train_loss = 19.805904 (0.496 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 19.805904, step = 286 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2797\n",
      "INFO:tensorflow:train_accuracy = 0.5488127, train_loss = 21.106628 (0.439 sec)\n",
      "INFO:tensorflow:loss = 21.106628, step = 287 (0.439 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02273\n",
      "INFO:tensorflow:train_accuracy = 0.5409623, train_loss = 23.450085 (0.494 sec)\n",
      "INFO:tensorflow:loss = 23.450085, step = 288 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86417\n",
      "INFO:tensorflow:train_accuracy = 0.53856385, train_loss = 20.975697 (0.537 sec)\n",
      "INFO:tensorflow:loss = 20.975697, step = 289 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2346\n",
      "INFO:tensorflow:train_accuracy = 0.532567, train_loss = 24.022818 (0.447 sec)\n",
      "INFO:tensorflow:loss = 24.022818, step = 290 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.27204\n",
      "INFO:tensorflow:train_accuracy = 0.54438645, train_loss = 22.653017 (0.440 sec)\n",
      "INFO:tensorflow:loss = 22.653017, step = 291 (0.441 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07016\n",
      "INFO:tensorflow:train_accuracy = 0.532567, train_loss = 24.694881 (0.483 sec)\n",
      "INFO:tensorflow:loss = 24.694881, step = 292 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94673\n",
      "INFO:tensorflow:train_accuracy = 0.5374677, train_loss = 21.324133 (0.514 sec)\n",
      "INFO:tensorflow:loss = 21.324133, step = 293 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84537\n",
      "INFO:tensorflow:train_accuracy = 0.54965985, train_loss = 22.291103 (0.542 sec)\n",
      "INFO:tensorflow:loss = 22.291103, step = 294 (0.542 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11698\n",
      "INFO:tensorflow:train_accuracy = 0.5473684, train_loss = 24.006884 (0.473 sec)\n",
      "INFO:tensorflow:loss = 24.006884, step = 295 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85423\n",
      "INFO:tensorflow:train_accuracy = 0.5291878, train_loss = 22.945522 (0.539 sec)\n",
      "INFO:tensorflow:loss = 22.945522, step = 296 (0.539 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97362\n",
      "INFO:tensorflow:train_accuracy = 0.50948167, train_loss = 23.475906 (0.507 sec)\n",
      "INFO:tensorflow:loss = 23.475906, step = 297 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98865\n",
      "INFO:tensorflow:train_accuracy = 0.55291003, train_loss = 20.220757 (0.503 sec)\n",
      "INFO:tensorflow:loss = 20.220757, step = 298 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91012\n",
      "INFO:tensorflow:train_accuracy = 0.5494071, train_loss = 21.952446 (0.524 sec)\n",
      "INFO:tensorflow:loss = 21.952446, step = 299 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01317\n",
      "INFO:tensorflow:train_accuracy = 0.5409623, train_loss = 21.444267 (0.498 sec)\n",
      "INFO:tensorflow:loss = 21.444267, step = 300 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95845\n",
      "INFO:tensorflow:train_accuracy = 0.5565912, train_loss = 20.549994 (0.510 sec)\n",
      "INFO:tensorflow:loss = 20.549994, step = 301 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97913\n",
      "INFO:tensorflow:train_accuracy = 0.5402597, train_loss = 23.226234 (0.505 sec)\n",
      "INFO:tensorflow:loss = 23.226234, step = 302 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20165\n",
      "INFO:tensorflow:train_accuracy = 0.51541305, train_loss = 24.956228 (0.455 sec)\n",
      "INFO:tensorflow:loss = 24.956228, step = 303 (0.455 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96886\n",
      "INFO:tensorflow:train_accuracy = 0.5225, train_loss = 24.690657 (0.508 sec)\n",
      "INFO:tensorflow:loss = 24.690657, step = 304 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06996\n",
      "INFO:tensorflow:train_accuracy = 0.54155844, train_loss = 22.238176 (0.483 sec)\n",
      "INFO:tensorflow:loss = 22.238176, step = 305 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05194\n",
      "INFO:tensorflow:train_accuracy = 0.5326505, train_loss = 23.78048 (0.487 sec)\n",
      "INFO:tensorflow:loss = 23.78048, step = 306 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04385\n",
      "INFO:tensorflow:train_accuracy = 0.54953766, train_loss = 21.275253 (0.489 sec)\n",
      "INFO:tensorflow:loss = 21.275253, step = 307 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03586\n",
      "INFO:tensorflow:train_accuracy = 0.52130324, train_loss = 25.662323 (0.491 sec)\n",
      "INFO:tensorflow:loss = 25.662323, step = 308 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02085\n",
      "INFO:tensorflow:train_accuracy = 0.53333336, train_loss = 22.346825 (0.495 sec)\n",
      "INFO:tensorflow:loss = 22.346825, step = 309 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91567\n",
      "INFO:tensorflow:train_accuracy = 0.5312102, train_loss = 24.509443 (0.522 sec)\n",
      "INFO:tensorflow:loss = 24.509443, step = 310 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86185\n",
      "INFO:tensorflow:train_accuracy = 0.5265823, train_loss = 24.602962 (0.537 sec)\n",
      "INFO:tensorflow:loss = 24.602962, step = 311 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09181\n",
      "INFO:tensorflow:train_accuracy = 0.52572143, train_loss = 23.79536 (0.478 sec)\n",
      "INFO:tensorflow:loss = 23.79536, step = 312 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80061\n",
      "INFO:tensorflow:train_accuracy = 0.54486346, train_loss = 22.081993 (0.555 sec)\n",
      "INFO:tensorflow:loss = 22.081993, step = 313 (0.555 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99284\n",
      "INFO:tensorflow:train_accuracy = 0.53589743, train_loss = 23.172455 (0.502 sec)\n",
      "INFO:tensorflow:loss = 23.172455, step = 314 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86516\n",
      "INFO:tensorflow:train_accuracy = 0.5054678, train_loss = 28.661213 (0.536 sec)\n",
      "INFO:tensorflow:loss = 28.661213, step = 315 (0.536 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.7077\n",
      "INFO:tensorflow:train_accuracy = 0.5346154, train_loss = 23.155056 (0.586 sec)\n",
      "INFO:tensorflow:loss = 23.155056, step = 316 (0.586 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95793\n",
      "INFO:tensorflow:train_accuracy = 0.5536424, train_loss = 21.550247 (0.511 sec)\n",
      "INFO:tensorflow:loss = 21.550247, step = 317 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91323\n",
      "INFO:tensorflow:train_accuracy = 0.51672864, train_loss = 24.908377 (0.523 sec)\n",
      "INFO:tensorflow:loss = 24.908377, step = 318 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84437\n",
      "INFO:tensorflow:train_accuracy = 0.552381, train_loss = 20.58707 (0.542 sec)\n",
      "INFO:tensorflow:loss = 20.58707, step = 319 (0.542 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.33869\n",
      "INFO:tensorflow:train_accuracy = 0.5591398, train_loss = 19.261257 (0.428 sec)\n",
      "INFO:tensorflow:loss = 19.261257, step = 320 (0.428 sec)\n",
      "INFO:tensorflow:global_step/sec: 7.93919\n",
      "INFO:tensorflow:train_accuracy = 0.5625, train_loss = 18.190605 (0.126 sec)\n",
      "INFO:tensorflow:loss = 18.190605, step = 321 (0.126 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80628\n",
      "INFO:tensorflow:train_accuracy = 0.5561497, train_loss = 19.972704 (0.554 sec)\n",
      "INFO:tensorflow:loss = 19.972704, step = 322 (0.553 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91978\n",
      "INFO:tensorflow:train_accuracy = 0.52059925, train_loss = 26.491688 (0.521 sec)\n",
      "INFO:tensorflow:loss = 26.491688, step = 323 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94602\n",
      "INFO:tensorflow:train_accuracy = 0.50606793, train_loss = 26.843414 (0.514 sec)\n",
      "INFO:tensorflow:loss = 26.843414, step = 324 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.19484\n",
      "INFO:tensorflow:train_accuracy = 0.5281782, train_loss = 21.440182 (0.455 sec)\n",
      "INFO:tensorflow:loss = 21.440182, step = 325 (0.455 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83424\n",
      "INFO:tensorflow:train_accuracy = 0.5316456, train_loss = 25.608822 (0.545 sec)\n",
      "INFO:tensorflow:loss = 25.608822, step = 326 (0.544 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87472\n",
      "INFO:tensorflow:train_accuracy = 0.53562343, train_loss = 23.17941 (0.534 sec)\n",
      "INFO:tensorflow:loss = 23.17941, step = 327 (0.534 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.74247\n",
      "INFO:tensorflow:train_accuracy = 0.5533597, train_loss = 22.59892 (0.574 sec)\n",
      "INFO:tensorflow:loss = 22.59892, step = 328 (0.573 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0047\n",
      "INFO:tensorflow:train_accuracy = 0.5318066, train_loss = 23.23203 (0.499 sec)\n",
      "INFO:tensorflow:loss = 23.23203, step = 329 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13184\n",
      "INFO:tensorflow:train_accuracy = 0.51861364, train_loss = 24.02952 (0.469 sec)\n",
      "INFO:tensorflow:loss = 24.02952, step = 330 (0.469 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8198\n",
      "INFO:tensorflow:train_accuracy = 0.5625841, train_loss = 19.667305 (0.549 sec)\n",
      "INFO:tensorflow:loss = 19.667305, step = 331 (0.549 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06192\n",
      "INFO:tensorflow:train_accuracy = 0.55466664, train_loss = 20.352573 (0.485 sec)\n",
      "INFO:tensorflow:loss = 20.352573, step = 332 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2942\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.56580734, train_loss = 19.240814 (0.436 sec)\n",
      "INFO:tensorflow:loss = 19.240814, step = 333 (0.436 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08017\n",
      "INFO:tensorflow:train_accuracy = 0.52791876, train_loss = 24.70099 (0.480 sec)\n",
      "INFO:tensorflow:loss = 24.70099, step = 334 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11908\n",
      "INFO:tensorflow:train_accuracy = 0.5402597, train_loss = 21.233028 (0.472 sec)\n",
      "INFO:tensorflow:loss = 21.233028, step = 335 (0.472 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05984\n",
      "INFO:tensorflow:train_accuracy = 0.5561497, train_loss = 20.779266 (0.485 sec)\n",
      "INFO:tensorflow:loss = 20.779266, step = 336 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.19727\n",
      "INFO:tensorflow:train_accuracy = 0.5387597, train_loss = 22.426971 (0.455 sec)\n",
      "INFO:tensorflow:loss = 22.426971, step = 337 (0.455 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83037\n",
      "INFO:tensorflow:train_accuracy = 0.5387597, train_loss = 24.663685 (0.546 sec)\n",
      "INFO:tensorflow:loss = 24.663685, step = 338 (0.545 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16218\n",
      "INFO:tensorflow:train_accuracy = 0.53061223, train_loss = 22.140572 (0.463 sec)\n",
      "INFO:tensorflow:loss = 22.140572, step = 339 (0.463 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87777\n",
      "INFO:tensorflow:train_accuracy = 0.55733335, train_loss = 20.423313 (0.533 sec)\n",
      "INFO:tensorflow:loss = 20.423313, step = 340 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92705\n",
      "INFO:tensorflow:train_accuracy = 0.52610964, train_loss = 23.581009 (0.519 sec)\n",
      "INFO:tensorflow:loss = 23.581009, step = 341 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.30981\n",
      "INFO:tensorflow:train_accuracy = 0.5402597, train_loss = 21.60571 (0.433 sec)\n",
      "INFO:tensorflow:loss = 21.60571, step = 342 (0.433 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2374\n",
      "INFO:tensorflow:train_accuracy = 0.53521127, train_loss = 22.296955 (0.447 sec)\n",
      "INFO:tensorflow:loss = 22.296955, step = 343 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87097\n",
      "INFO:tensorflow:train_accuracy = 0.53777206, train_loss = 25.035557 (0.534 sec)\n",
      "INFO:tensorflow:loss = 25.035557, step = 344 (0.534 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0654\n",
      "INFO:tensorflow:train_accuracy = 0.5607094, train_loss = 16.708258 (0.484 sec)\n",
      "INFO:tensorflow:loss = 16.708258, step = 345 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89378\n",
      "INFO:tensorflow:train_accuracy = 0.5606469, train_loss = 20.631256 (0.528 sec)\n",
      "INFO:tensorflow:loss = 20.631256, step = 346 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96911\n",
      "INFO:tensorflow:train_accuracy = 0.5179949, train_loss = 22.41425 (0.508 sec)\n",
      "INFO:tensorflow:loss = 22.41425, step = 347 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88401\n",
      "INFO:tensorflow:train_accuracy = 0.51796776, train_loss = 24.891224 (0.531 sec)\n",
      "INFO:tensorflow:loss = 24.891224, step = 348 (0.531 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8142\n",
      "INFO:tensorflow:train_accuracy = 0.54486346, train_loss = 22.535679 (0.551 sec)\n",
      "INFO:tensorflow:loss = 22.535679, step = 349 (0.551 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94912\n",
      "INFO:tensorflow:train_accuracy = 0.540856, train_loss = 21.082241 (0.513 sec)\n",
      "INFO:tensorflow:loss = 21.082241, step = 350 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87186\n",
      "INFO:tensorflow:train_accuracy = 0.5265823, train_loss = 25.824083 (0.534 sec)\n",
      "INFO:tensorflow:loss = 25.824083, step = 351 (0.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1029\n",
      "INFO:tensorflow:train_accuracy = 0.53188777, train_loss = 23.629204 (0.475 sec)\n",
      "INFO:tensorflow:loss = 23.629204, step = 352 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9855\n",
      "INFO:tensorflow:train_accuracy = 0.512285, train_loss = 23.89436 (0.504 sec)\n",
      "INFO:tensorflow:loss = 23.89436, step = 353 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96253\n",
      "INFO:tensorflow:train_accuracy = 0.55, train_loss = 21.420177 (0.509 sec)\n",
      "INFO:tensorflow:loss = 21.420177, step = 354 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96285\n",
      "INFO:tensorflow:train_accuracy = 0.5219024, train_loss = 23.55933 (0.510 sec)\n",
      "INFO:tensorflow:loss = 23.55933, step = 355 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04781\n",
      "INFO:tensorflow:train_accuracy = 0.5488127, train_loss = 21.699303 (0.488 sec)\n",
      "INFO:tensorflow:loss = 21.699303, step = 356 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80216\n",
      "INFO:tensorflow:train_accuracy = 0.51541305, train_loss = 23.751726 (0.555 sec)\n",
      "INFO:tensorflow:loss = 23.751726, step = 357 (0.555 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98939\n",
      "INFO:tensorflow:train_accuracy = 0.52644837, train_loss = 23.826664 (0.503 sec)\n",
      "INFO:tensorflow:loss = 23.826664, step = 358 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99102\n",
      "INFO:tensorflow:train_accuracy = 0.52386934, train_loss = 23.611252 (0.502 sec)\n",
      "INFO:tensorflow:loss = 23.611252, step = 359 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91647\n",
      "INFO:tensorflow:train_accuracy = 0.5606469, train_loss = 20.580166 (0.522 sec)\n",
      "INFO:tensorflow:loss = 20.580166, step = 360 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16688\n",
      "INFO:tensorflow:train_accuracy = 0.53945667, train_loss = 22.255821 (0.462 sec)\n",
      "INFO:tensorflow:loss = 22.255821, step = 361 (0.462 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8612\n",
      "INFO:tensorflow:train_accuracy = 0.519802, train_loss = 24.42416 (0.537 sec)\n",
      "INFO:tensorflow:loss = 24.42416, step = 362 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02365\n",
      "INFO:tensorflow:train_accuracy = 0.5536424, train_loss = 21.215042 (0.494 sec)\n",
      "INFO:tensorflow:loss = 21.215042, step = 363 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95086\n",
      "INFO:tensorflow:train_accuracy = 0.53806454, train_loss = 22.123684 (0.512 sec)\n",
      "INFO:tensorflow:loss = 22.123684, step = 364 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1177\n",
      "INFO:tensorflow:train_accuracy = 0.5436767, train_loss = 20.076 (0.473 sec)\n",
      "INFO:tensorflow:loss = 20.076, step = 365 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98193\n",
      "INFO:tensorflow:train_accuracy = 0.53667957, train_loss = 22.270775 (0.504 sec)\n",
      "INFO:tensorflow:loss = 22.270775, step = 366 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.62708\n",
      "INFO:tensorflow:train_accuracy = 0.5891089, train_loss = 15.7023115 (0.151 sec)\n",
      "INFO:tensorflow:loss = 15.7023115, step = 367 (0.151 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.78172\n",
      "INFO:tensorflow:train_accuracy = 0.53935486, train_loss = 21.71092 (0.561 sec)\n",
      "INFO:tensorflow:loss = 21.71092, step = 368 (0.561 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05214\n",
      "INFO:tensorflow:train_accuracy = 0.53188777, train_loss = 21.66859 (0.488 sec)\n",
      "INFO:tensorflow:loss = 21.66859, step = 369 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.77683\n",
      "INFO:tensorflow:train_accuracy = 0.48736462, train_loss = 27.057297 (0.563 sec)\n",
      "INFO:tensorflow:loss = 27.057297, step = 370 (0.563 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94216\n",
      "INFO:tensorflow:train_accuracy = 0.5386598, train_loss = 22.009024 (0.515 sec)\n",
      "INFO:tensorflow:loss = 22.009024, step = 371 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.26828\n",
      "INFO:tensorflow:train_accuracy = 0.54855645, train_loss = 21.64001 (0.441 sec)\n",
      "INFO:tensorflow:loss = 21.64001, step = 372 (0.441 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.27871\n",
      "INFO:tensorflow:train_accuracy = 0.52195734, train_loss = 21.641146 (0.438 sec)\n",
      "INFO:tensorflow:loss = 21.641146, step = 373 (0.438 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96669\n",
      "INFO:tensorflow:train_accuracy = 0.54415584, train_loss = 21.973766 (0.509 sec)\n",
      "INFO:tensorflow:loss = 21.973766, step = 374 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85954\n",
      "INFO:tensorflow:train_accuracy = 0.56469005, train_loss = 20.935225 (0.537 sec)\n",
      "INFO:tensorflow:loss = 20.935225, step = 375 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08737\n",
      "INFO:tensorflow:train_accuracy = 0.5456919, train_loss = 19.465961 (0.480 sec)\n",
      "INFO:tensorflow:loss = 19.465961, step = 376 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07826\n",
      "INFO:tensorflow:train_accuracy = 0.5376486, train_loss = 19.83078 (0.481 sec)\n",
      "INFO:tensorflow:loss = 19.83078, step = 377 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18586\n",
      "INFO:tensorflow:train_accuracy = 0.55350065, train_loss = 20.355059 (0.458 sec)\n",
      "INFO:tensorflow:loss = 20.355059, step = 378 (0.458 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02219\n",
      "INFO:tensorflow:train_accuracy = 0.5507246, train_loss = 21.028152 (0.494 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 21.028152, step = 379 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24757\n",
      "INFO:tensorflow:train_accuracy = 0.5318066, train_loss = 22.64 (0.445 sec)\n",
      "INFO:tensorflow:loss = 22.64, step = 380 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.25997\n",
      "INFO:tensorflow:train_accuracy = 0.54, train_loss = 19.866201 (0.442 sec)\n",
      "INFO:tensorflow:loss = 19.866201, step = 381 (0.442 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09572\n",
      "INFO:tensorflow:train_accuracy = 0.5614035, train_loss = 19.535679 (0.478 sec)\n",
      "INFO:tensorflow:loss = 19.535679, step = 382 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.75883\n",
      "INFO:tensorflow:train_accuracy = 0.5318066, train_loss = 23.607548 (0.569 sec)\n",
      "INFO:tensorflow:loss = 23.607548, step = 383 (0.569 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.22912\n",
      "INFO:tensorflow:train_accuracy = 0.5226978, train_loss = 20.068905 (0.448 sec)\n",
      "INFO:tensorflow:loss = 20.068905, step = 384 (0.449 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94685\n",
      "INFO:tensorflow:train_accuracy = 0.55437666, train_loss = 19.821594 (0.513 sec)\n",
      "INFO:tensorflow:loss = 19.821594, step = 385 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.3026\n",
      "INFO:tensorflow:train_accuracy = 0.5436767, train_loss = 20.434778 (0.434 sec)\n",
      "INFO:tensorflow:loss = 20.434778, step = 386 (0.435 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15163\n",
      "INFO:tensorflow:train_accuracy = 0.5442708, train_loss = 23.437775 (0.465 sec)\n",
      "INFO:tensorflow:loss = 23.437775, step = 387 (0.464 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95753\n",
      "INFO:tensorflow:train_accuracy = 0.51625, train_loss = 23.33173 (0.511 sec)\n",
      "INFO:tensorflow:loss = 23.33173, step = 388 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01979\n",
      "INFO:tensorflow:train_accuracy = 0.54652685, train_loss = 20.971436 (0.495 sec)\n",
      "INFO:tensorflow:loss = 20.971436, step = 389 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90015\n",
      "INFO:tensorflow:train_accuracy = 0.5239295, train_loss = 23.502117 (0.526 sec)\n",
      "INFO:tensorflow:loss = 23.502117, step = 390 (0.526 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06263\n",
      "INFO:tensorflow:train_accuracy = 0.5619946, train_loss = 19.112 (0.485 sec)\n",
      "INFO:tensorflow:loss = 19.112, step = 391 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90871\n",
      "INFO:tensorflow:train_accuracy = 0.54746425, train_loss = 22.906612 (0.524 sec)\n",
      "INFO:tensorflow:loss = 22.906612, step = 392 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20461\n",
      "INFO:tensorflow:train_accuracy = 0.542044, train_loss = 21.030972 (0.454 sec)\n",
      "INFO:tensorflow:loss = 21.030972, step = 393 (0.454 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86753\n",
      "INFO:tensorflow:train_accuracy = 0.54652685, train_loss = 21.298126 (0.535 sec)\n",
      "INFO:tensorflow:loss = 21.298126, step = 394 (0.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97873\n",
      "INFO:tensorflow:train_accuracy = 0.53426397, train_loss = 21.459484 (0.506 sec)\n",
      "INFO:tensorflow:loss = 21.459484, step = 395 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18871\n",
      "INFO:tensorflow:train_accuracy = 0.55378485, train_loss = 19.568169 (0.457 sec)\n",
      "INFO:tensorflow:loss = 19.568169, step = 396 (0.457 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.28486\n",
      "INFO:tensorflow:train_accuracy = 0.5193026, train_loss = 23.278412 (0.437 sec)\n",
      "INFO:tensorflow:loss = 23.278412, step = 397 (0.437 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.29701\n",
      "INFO:tensorflow:train_accuracy = 0.5397436, train_loss = 22.017982 (0.435 sec)\n",
      "INFO:tensorflow:loss = 22.017982, step = 398 (0.435 sec)\n",
      "INFO:tensorflow:Saving checkpoints for 400 into asr-minijasper/model.ckpt.\n",
      "INFO:tensorflow:Skip the current checkpoint eval due to throttle secs (120 secs).\n",
      "INFO:tensorflow:global_step/sec: 0.650873\n",
      "INFO:tensorflow:train_accuracy = 0.5318066, train_loss = 23.789211 (1.536 sec)\n",
      "INFO:tensorflow:loss = 23.789211, step = 399 (1.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20925\n",
      "INFO:tensorflow:train_accuracy = 0.5353925, train_loss = 22.14156 (0.454 sec)\n",
      "INFO:tensorflow:loss = 22.14156, step = 400 (0.454 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98764\n",
      "INFO:tensorflow:train_accuracy = 0.52059925, train_loss = 23.128513 (0.503 sec)\n",
      "INFO:tensorflow:loss = 23.128513, step = 401 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8396\n",
      "INFO:tensorflow:train_accuracy = 0.5359897, train_loss = 22.848978 (0.543 sec)\n",
      "INFO:tensorflow:loss = 22.848978, step = 402 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85448\n",
      "INFO:tensorflow:train_accuracy = 0.5128834, train_loss = 23.12448 (0.539 sec)\n",
      "INFO:tensorflow:loss = 23.12448, step = 403 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00811\n",
      "INFO:tensorflow:train_accuracy = 0.53796655, train_loss = 22.702892 (0.498 sec)\n",
      "INFO:tensorflow:loss = 22.702892, step = 404 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92023\n",
      "INFO:tensorflow:train_accuracy = 0.5285171, train_loss = 21.908585 (0.521 sec)\n",
      "INFO:tensorflow:loss = 21.908585, step = 405 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11685\n",
      "INFO:tensorflow:train_accuracy = 0.5397436, train_loss = 22.262875 (0.472 sec)\n",
      "INFO:tensorflow:loss = 22.262875, step = 406 (0.472 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90523\n",
      "INFO:tensorflow:train_accuracy = 0.5374677, train_loss = 21.622656 (0.525 sec)\n",
      "INFO:tensorflow:loss = 21.622656, step = 407 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90305\n",
      "INFO:tensorflow:train_accuracy = 0.54486346, train_loss = 20.388287 (0.525 sec)\n",
      "INFO:tensorflow:loss = 20.388287, step = 408 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91126\n",
      "INFO:tensorflow:train_accuracy = 0.53945667, train_loss = 21.919037 (0.524 sec)\n",
      "INFO:tensorflow:loss = 21.919037, step = 409 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96027\n",
      "INFO:tensorflow:train_accuracy = 0.5224439, train_loss = 22.304428 (0.510 sec)\n",
      "INFO:tensorflow:loss = 22.304428, step = 410 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2117\n",
      "INFO:tensorflow:train_accuracy = 0.5612382, train_loss = 16.907074 (0.452 sec)\n",
      "INFO:tensorflow:loss = 16.907074, step = 411 (0.452 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.25404\n",
      "INFO:tensorflow:train_accuracy = 0.55026454, train_loss = 20.341158 (0.444 sec)\n",
      "INFO:tensorflow:loss = 20.341158, step = 412 (0.444 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.21476\n",
      "INFO:tensorflow:train_accuracy = 0.5619048, train_loss = 21.011845 (0.161 sec)\n",
      "INFO:tensorflow:loss = 21.011845, step = 413 (0.161 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.76152\n",
      "INFO:tensorflow:train_accuracy = 0.53188777, train_loss = 21.87595 (0.568 sec)\n",
      "INFO:tensorflow:loss = 21.87595, step = 414 (0.568 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04702\n",
      "INFO:tensorflow:train_accuracy = 0.5258512, train_loss = 21.577827 (0.488 sec)\n",
      "INFO:tensorflow:loss = 21.577827, step = 415 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89332\n",
      "INFO:tensorflow:train_accuracy = 0.52578616, train_loss = 24.401983 (0.528 sec)\n",
      "INFO:tensorflow:loss = 24.401983, step = 416 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8227\n",
      "INFO:tensorflow:train_accuracy = 0.5204461, train_loss = 23.443314 (0.548 sec)\n",
      "INFO:tensorflow:loss = 23.443314, step = 417 (0.548 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91242\n",
      "INFO:tensorflow:train_accuracy = 0.55172414, train_loss = 18.808058 (0.523 sec)\n",
      "INFO:tensorflow:loss = 18.808058, step = 418 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86759\n",
      "INFO:tensorflow:train_accuracy = 0.55104715, train_loss = 19.658163 (0.535 sec)\n",
      "INFO:tensorflow:loss = 19.658163, step = 419 (0.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94803\n",
      "INFO:tensorflow:train_accuracy = 0.5446313, train_loss = 21.060349 (0.513 sec)\n",
      "INFO:tensorflow:loss = 21.060349, step = 420 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87746\n",
      "INFO:tensorflow:train_accuracy = 0.53915274, train_loss = 20.679642 (0.532 sec)\n",
      "INFO:tensorflow:loss = 20.679642, step = 421 (0.532 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03625\n",
      "INFO:tensorflow:train_accuracy = 0.5412371, train_loss = 19.7669 (0.491 sec)\n",
      "INFO:tensorflow:loss = 19.7669, step = 422 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99528\n",
      "INFO:tensorflow:train_accuracy = 0.53642386, train_loss = 19.970499 (0.502 sec)\n",
      "INFO:tensorflow:loss = 19.970499, step = 423 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.25803\n",
      "INFO:tensorflow:train_accuracy = 0.5555556, train_loss = 19.402746 (0.443 sec)\n",
      "INFO:tensorflow:loss = 19.402746, step = 424 (0.443 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16527\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.5509934, train_loss = 18.509548 (0.462 sec)\n",
      "INFO:tensorflow:loss = 18.509548, step = 425 (0.461 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91652\n",
      "INFO:tensorflow:train_accuracy = 0.54486346, train_loss = 20.26668 (0.522 sec)\n",
      "INFO:tensorflow:loss = 20.26668, step = 426 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01187\n",
      "INFO:tensorflow:train_accuracy = 0.5406125, train_loss = 19.277946 (0.497 sec)\n",
      "INFO:tensorflow:loss = 19.277946, step = 427 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89718\n",
      "INFO:tensorflow:train_accuracy = 0.54664916, train_loss = 20.90541 (0.527 sec)\n",
      "INFO:tensorflow:loss = 20.90541, step = 428 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89401\n",
      "INFO:tensorflow:train_accuracy = 0.53846157, train_loss = 21.939167 (0.528 sec)\n",
      "INFO:tensorflow:loss = 21.939167, step = 429 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11602\n",
      "INFO:tensorflow:train_accuracy = 0.55585104, train_loss = 18.578194 (0.473 sec)\n",
      "INFO:tensorflow:loss = 18.578194, step = 430 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97423\n",
      "INFO:tensorflow:train_accuracy = 0.54581153, train_loss = 19.794907 (0.506 sec)\n",
      "INFO:tensorflow:loss = 19.794907, step = 431 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1889\n",
      "INFO:tensorflow:train_accuracy = 0.53367215, train_loss = 22.24279 (0.457 sec)\n",
      "INFO:tensorflow:loss = 22.24279, step = 432 (0.457 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85468\n",
      "INFO:tensorflow:train_accuracy = 0.5508587, train_loss = 19.939796 (0.539 sec)\n",
      "INFO:tensorflow:loss = 19.939796, step = 433 (0.539 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91293\n",
      "INFO:tensorflow:train_accuracy = 0.52516127, train_loss = 24.353247 (0.523 sec)\n",
      "INFO:tensorflow:loss = 24.353247, step = 434 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98605\n",
      "INFO:tensorflow:train_accuracy = 0.53426397, train_loss = 20.056252 (0.504 sec)\n",
      "INFO:tensorflow:loss = 20.056252, step = 435 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.81422\n",
      "INFO:tensorflow:train_accuracy = 0.56891894, train_loss = 19.038744 (0.551 sec)\n",
      "INFO:tensorflow:loss = 19.038744, step = 436 (0.552 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.33327\n",
      "INFO:tensorflow:train_accuracy = 0.5364917, train_loss = 19.919533 (0.429 sec)\n",
      "INFO:tensorflow:loss = 19.919533, step = 437 (0.428 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98742\n",
      "INFO:tensorflow:train_accuracy = 0.54215306, train_loss = 20.532394 (0.503 sec)\n",
      "INFO:tensorflow:loss = 20.532394, step = 438 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14999\n",
      "INFO:tensorflow:train_accuracy = 0.56469005, train_loss = 18.608425 (0.465 sec)\n",
      "INFO:tensorflow:loss = 18.608425, step = 439 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86708\n",
      "INFO:tensorflow:train_accuracy = 0.54986876, train_loss = 22.18822 (0.536 sec)\n",
      "INFO:tensorflow:loss = 22.18822, step = 440 (0.536 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17734\n",
      "INFO:tensorflow:train_accuracy = 0.5280899, train_loss = 22.135502 (0.459 sec)\n",
      "INFO:tensorflow:loss = 22.135502, step = 441 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95918\n",
      "INFO:tensorflow:train_accuracy = 0.54498714, train_loss = 22.71225 (0.510 sec)\n",
      "INFO:tensorflow:loss = 22.71225, step = 442 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00689\n",
      "INFO:tensorflow:train_accuracy = 0.53324807, train_loss = 20.424582 (0.498 sec)\n",
      "INFO:tensorflow:loss = 20.424582, step = 443 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88221\n",
      "INFO:tensorflow:train_accuracy = 0.52837324, train_loss = 22.09518 (0.531 sec)\n",
      "INFO:tensorflow:loss = 22.09518, step = 444 (0.532 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95742\n",
      "INFO:tensorflow:train_accuracy = 0.5419355, train_loss = 20.177479 (0.511 sec)\n",
      "INFO:tensorflow:loss = 20.177479, step = 445 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96287\n",
      "INFO:tensorflow:train_accuracy = 0.5397436, train_loss = 20.621763 (0.509 sec)\n",
      "INFO:tensorflow:loss = 20.621763, step = 446 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99162\n",
      "INFO:tensorflow:train_accuracy = 0.525593, train_loss = 22.55097 (0.502 sec)\n",
      "INFO:tensorflow:loss = 22.55097, step = 447 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93813\n",
      "INFO:tensorflow:train_accuracy = 0.5277078, train_loss = 22.393368 (0.516 sec)\n",
      "INFO:tensorflow:loss = 22.393368, step = 448 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87323\n",
      "INFO:tensorflow:train_accuracy = 0.53562343, train_loss = 20.994196 (0.534 sec)\n",
      "INFO:tensorflow:loss = 20.994196, step = 449 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94757\n",
      "INFO:tensorflow:train_accuracy = 0.5327456, train_loss = 23.307804 (0.513 sec)\n",
      "INFO:tensorflow:loss = 23.307804, step = 450 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06594\n",
      "INFO:tensorflow:train_accuracy = 0.5395408, train_loss = 20.442913 (0.484 sec)\n",
      "INFO:tensorflow:loss = 20.442913, step = 451 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01263\n",
      "INFO:tensorflow:train_accuracy = 0.5479632, train_loss = 19.583874 (0.497 sec)\n",
      "INFO:tensorflow:loss = 19.583874, step = 452 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96872\n",
      "INFO:tensorflow:train_accuracy = 0.53816795, train_loss = 21.367037 (0.508 sec)\n",
      "INFO:tensorflow:loss = 21.367037, step = 453 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02872\n",
      "INFO:tensorflow:train_accuracy = 0.54415584, train_loss = 21.110802 (0.493 sec)\n",
      "INFO:tensorflow:loss = 21.110802, step = 454 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83259\n",
      "INFO:tensorflow:train_accuracy = 0.55585104, train_loss = 17.642113 (0.545 sec)\n",
      "INFO:tensorflow:loss = 17.642113, step = 455 (0.545 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87364\n",
      "INFO:tensorflow:train_accuracy = 0.51537514, train_loss = 24.035698 (0.534 sec)\n",
      "INFO:tensorflow:loss = 24.035698, step = 456 (0.534 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96777\n",
      "INFO:tensorflow:train_accuracy = 0.5369898, train_loss = 20.188538 (0.508 sec)\n",
      "INFO:tensorflow:loss = 20.188538, step = 457 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94039\n",
      "INFO:tensorflow:train_accuracy = 0.57260275, train_loss = 16.033606 (0.515 sec)\n",
      "INFO:tensorflow:loss = 16.033606, step = 458 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.16024\n",
      "INFO:tensorflow:train_accuracy = 0.5315315, train_loss = 23.145021 (0.163 sec)\n",
      "INFO:tensorflow:loss = 23.145021, step = 459 (0.163 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.70111\n",
      "INFO:tensorflow:train_accuracy = 0.54521966, train_loss = 19.924656 (0.588 sec)\n",
      "INFO:tensorflow:loss = 19.924656, step = 460 (0.588 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99021\n",
      "INFO:tensorflow:train_accuracy = 0.54333764, train_loss = 21.139683 (0.502 sec)\n",
      "INFO:tensorflow:loss = 21.139683, step = 461 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.73572\n",
      "INFO:tensorflow:train_accuracy = 0.5241038, train_loss = 24.258884 (0.576 sec)\n",
      "INFO:tensorflow:loss = 24.258884, step = 462 (0.576 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96949\n",
      "INFO:tensorflow:train_accuracy = 0.49632353, train_loss = 22.859818 (0.508 sec)\n",
      "INFO:tensorflow:loss = 22.859818, step = 463 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93897\n",
      "INFO:tensorflow:train_accuracy = 0.53603035, train_loss = 21.0904 (0.516 sec)\n",
      "INFO:tensorflow:loss = 21.0904, step = 464 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98907\n",
      "INFO:tensorflow:train_accuracy = 0.5582329, train_loss = 17.042295 (0.503 sec)\n",
      "INFO:tensorflow:loss = 17.042295, step = 465 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00095\n",
      "INFO:tensorflow:train_accuracy = 0.5372751, train_loss = 20.65278 (0.499 sec)\n",
      "INFO:tensorflow:loss = 20.65278, step = 466 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98752\n",
      "INFO:tensorflow:train_accuracy = 0.55925435, train_loss = 17.547153 (0.503 sec)\n",
      "INFO:tensorflow:loss = 17.547153, step = 467 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91966\n",
      "INFO:tensorflow:train_accuracy = 0.5568783, train_loss = 18.720947 (0.521 sec)\n",
      "INFO:tensorflow:loss = 18.720947, step = 468 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89298\n",
      "INFO:tensorflow:train_accuracy = 0.52699226, train_loss = 20.254364 (0.528 sec)\n",
      "INFO:tensorflow:loss = 20.254364, step = 469 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91979\n",
      "INFO:tensorflow:train_accuracy = 0.55599475, train_loss = 20.068573 (0.521 sec)\n",
      "INFO:tensorflow:loss = 20.068573, step = 470 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09407\n",
      "INFO:tensorflow:train_accuracy = 0.5519054, train_loss = 19.319252 (0.478 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 19.319252, step = 471 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.26451\n",
      "INFO:tensorflow:train_accuracy = 0.5359897, train_loss = 20.459549 (0.442 sec)\n",
      "INFO:tensorflow:loss = 20.459549, step = 472 (0.442 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12527\n",
      "INFO:tensorflow:train_accuracy = 0.55496687, train_loss = 19.011646 (0.471 sec)\n",
      "INFO:tensorflow:loss = 19.011646, step = 473 (0.471 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04623\n",
      "INFO:tensorflow:train_accuracy = 0.55866665, train_loss = 17.760443 (0.489 sec)\n",
      "INFO:tensorflow:loss = 17.760443, step = 474 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92641\n",
      "INFO:tensorflow:train_accuracy = 0.53231937, train_loss = 22.80167 (0.519 sec)\n",
      "INFO:tensorflow:loss = 22.80167, step = 475 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89877\n",
      "INFO:tensorflow:train_accuracy = 0.54438645, train_loss = 20.045355 (0.527 sec)\n",
      "INFO:tensorflow:loss = 20.045355, step = 476 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99717\n",
      "INFO:tensorflow:train_accuracy = 0.54593176, train_loss = 18.048159 (0.500 sec)\n",
      "INFO:tensorflow:loss = 18.048159, step = 477 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.68155\n",
      "INFO:tensorflow:train_accuracy = 0.56580734, train_loss = 18.607553 (0.595 sec)\n",
      "INFO:tensorflow:loss = 18.607553, step = 478 (0.595 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00023\n",
      "INFO:tensorflow:train_accuracy = 0.55496687, train_loss = 20.306776 (0.500 sec)\n",
      "INFO:tensorflow:loss = 20.306776, step = 479 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14973\n",
      "INFO:tensorflow:train_accuracy = 0.5225, train_loss = 21.678867 (0.465 sec)\n",
      "INFO:tensorflow:loss = 21.678867, step = 480 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.30091\n",
      "INFO:tensorflow:train_accuracy = 0.5503268, train_loss = 17.514095 (0.435 sec)\n",
      "INFO:tensorflow:loss = 17.514095, step = 481 (0.435 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93856\n",
      "INFO:tensorflow:train_accuracy = 0.5540897, train_loss = 20.051476 (0.517 sec)\n",
      "INFO:tensorflow:loss = 20.051476, step = 482 (0.517 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.32243\n",
      "INFO:tensorflow:train_accuracy = 0.5601604, train_loss = 17.586313 (0.430 sec)\n",
      "INFO:tensorflow:loss = 17.586313, step = 483 (0.430 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95649\n",
      "INFO:tensorflow:train_accuracy = 0.5382653, train_loss = 20.718431 (0.511 sec)\n",
      "INFO:tensorflow:loss = 20.718431, step = 484 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08651\n",
      "INFO:tensorflow:train_accuracy = 0.55277044, train_loss = 20.249866 (0.479 sec)\n",
      "INFO:tensorflow:loss = 20.249866, step = 485 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13246\n",
      "INFO:tensorflow:train_accuracy = 0.53717947, train_loss = 19.222666 (0.469 sec)\n",
      "INFO:tensorflow:loss = 19.222666, step = 486 (0.468 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.75075\n",
      "INFO:tensorflow:train_accuracy = 0.5190652, train_loss = 21.970016 (0.571 sec)\n",
      "INFO:tensorflow:loss = 21.970016, step = 487 (0.572 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04841\n",
      "INFO:tensorflow:train_accuracy = 0.5501956, train_loss = 20.11476 (0.488 sec)\n",
      "INFO:tensorflow:loss = 20.11476, step = 488 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06141\n",
      "INFO:tensorflow:train_accuracy = 0.53915274, train_loss = 20.570763 (0.485 sec)\n",
      "INFO:tensorflow:loss = 20.570763, step = 489 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01314\n",
      "INFO:tensorflow:train_accuracy = 0.519802, train_loss = 20.909647 (0.497 sec)\n",
      "INFO:tensorflow:loss = 20.909647, step = 490 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92294\n",
      "INFO:tensorflow:train_accuracy = 0.55026454, train_loss = 20.829536 (0.520 sec)\n",
      "INFO:tensorflow:loss = 20.829536, step = 491 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8713\n",
      "INFO:tensorflow:train_accuracy = 0.54311454, train_loss = 19.979042 (0.534 sec)\n",
      "INFO:tensorflow:loss = 19.979042, step = 492 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.31546\n",
      "INFO:tensorflow:train_accuracy = 0.52697617, train_loss = 20.526361 (0.432 sec)\n",
      "INFO:tensorflow:loss = 20.526361, step = 493 (0.432 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96242\n",
      "INFO:tensorflow:train_accuracy = 0.546875, train_loss = 20.188343 (0.510 sec)\n",
      "INFO:tensorflow:loss = 20.188343, step = 494 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.77577\n",
      "INFO:tensorflow:train_accuracy = 0.50358, train_loss = 23.765747 (0.563 sec)\n",
      "INFO:tensorflow:loss = 23.765747, step = 495 (0.563 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87799\n",
      "INFO:tensorflow:train_accuracy = 0.53417724, train_loss = 20.852993 (0.532 sec)\n",
      "INFO:tensorflow:loss = 20.852993, step = 496 (0.532 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.7763\n",
      "INFO:tensorflow:train_accuracy = 0.5467532, train_loss = 19.638718 (0.563 sec)\n",
      "INFO:tensorflow:loss = 19.638718, step = 497 (0.564 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08315\n",
      "INFO:tensorflow:train_accuracy = 0.53435117, train_loss = 21.219467 (0.480 sec)\n",
      "INFO:tensorflow:loss = 21.219467, step = 498 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95889\n",
      "INFO:tensorflow:train_accuracy = 0.5539474, train_loss = 20.322731 (0.511 sec)\n",
      "INFO:tensorflow:loss = 20.322731, step = 499 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9086\n",
      "INFO:tensorflow:train_accuracy = 0.5230386, train_loss = 22.109852 (0.525 sec)\n",
      "INFO:tensorflow:loss = 22.109852, step = 500 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80517\n",
      "INFO:tensorflow:train_accuracy = 0.5432258, train_loss = 19.318096 (0.553 sec)\n",
      "INFO:tensorflow:loss = 19.318096, step = 501 (0.553 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17514\n",
      "INFO:tensorflow:train_accuracy = 0.53589743, train_loss = 19.41978 (0.460 sec)\n",
      "INFO:tensorflow:loss = 19.41978, step = 502 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95082\n",
      "INFO:tensorflow:train_accuracy = 0.5675676, train_loss = 16.709156 (0.512 sec)\n",
      "INFO:tensorflow:loss = 16.709156, step = 503 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91795\n",
      "INFO:tensorflow:train_accuracy = 0.552356, train_loss = 17.901367 (0.522 sec)\n",
      "INFO:tensorflow:loss = 17.901367, step = 504 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.47019\n",
      "INFO:tensorflow:train_accuracy = 0.5625, train_loss = 19.00425 (0.154 sec)\n",
      "INFO:tensorflow:loss = 19.00425, step = 505 (0.154 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.70047\n",
      "INFO:tensorflow:train_accuracy = 0.552356, train_loss = 18.684895 (0.589 sec)\n",
      "INFO:tensorflow:loss = 18.684895, step = 506 (0.589 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10901\n",
      "INFO:tensorflow:train_accuracy = 0.51724136, train_loss = 20.313595 (0.474 sec)\n",
      "INFO:tensorflow:loss = 20.313595, step = 507 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.32374\n",
      "INFO:tensorflow:train_accuracy = 0.5302267, train_loss = 20.757694 (0.431 sec)\n",
      "INFO:tensorflow:loss = 20.757694, step = 508 (0.431 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86171\n",
      "INFO:tensorflow:train_accuracy = 0.52416354, train_loss = 23.296875 (0.537 sec)\n",
      "INFO:tensorflow:loss = 23.296875, step = 509 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06507\n",
      "INFO:tensorflow:train_accuracy = 0.55249345, train_loss = 18.328049 (0.484 sec)\n",
      "INFO:tensorflow:loss = 18.328049, step = 510 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96225\n",
      "INFO:tensorflow:train_accuracy = 0.5349428, train_loss = 19.53911 (0.509 sec)\n",
      "INFO:tensorflow:loss = 19.53911, step = 511 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83977\n",
      "INFO:tensorflow:train_accuracy = 0.5460358, train_loss = 20.511906 (0.543 sec)\n",
      "INFO:tensorflow:loss = 20.511906, step = 512 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88161\n",
      "INFO:tensorflow:train_accuracy = 0.5570118, train_loss = 18.991028 (0.531 sec)\n",
      "INFO:tensorflow:loss = 18.991028, step = 513 (0.531 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84361\n",
      "INFO:tensorflow:train_accuracy = 0.53875476, train_loss = 22.91716 (0.543 sec)\n",
      "INFO:tensorflow:loss = 22.91716, step = 514 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07575\n",
      "INFO:tensorflow:train_accuracy = 0.56334233, train_loss = 17.410011 (0.482 sec)\n",
      "INFO:tensorflow:loss = 17.410011, step = 515 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.22364\n",
      "INFO:tensorflow:train_accuracy = 0.5471204, train_loss = 18.880306 (0.450 sec)\n",
      "INFO:tensorflow:loss = 18.880306, step = 516 (0.450 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04995\n",
      "INFO:tensorflow:train_accuracy = 0.5606469, train_loss = 18.89855 (0.488 sec)\n",
      "INFO:tensorflow:loss = 18.89855, step = 517 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96065\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.52697617, train_loss = 20.358944 (0.510 sec)\n",
      "INFO:tensorflow:loss = 20.358944, step = 518 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00831\n",
      "INFO:tensorflow:train_accuracy = 0.56562924, train_loss = 16.82433 (0.498 sec)\n",
      "INFO:tensorflow:loss = 16.82433, step = 519 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14992\n",
      "INFO:tensorflow:train_accuracy = 0.55510205, train_loss = 17.132923 (0.465 sec)\n",
      "INFO:tensorflow:loss = 17.132923, step = 520 (0.465 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94076\n",
      "INFO:tensorflow:train_accuracy = 0.525593, train_loss = 20.509502 (0.515 sec)\n",
      "INFO:tensorflow:loss = 20.509502, step = 521 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05374\n",
      "INFO:tensorflow:train_accuracy = 0.52693826, train_loss = 19.886673 (0.486 sec)\n",
      "INFO:tensorflow:loss = 19.886673, step = 522 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04303\n",
      "INFO:tensorflow:train_accuracy = 0.52837324, train_loss = 20.910532 (0.490 sec)\n",
      "INFO:tensorflow:loss = 20.910532, step = 523 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89333\n",
      "INFO:tensorflow:train_accuracy = 0.55496687, train_loss = 18.933023 (0.528 sec)\n",
      "INFO:tensorflow:loss = 18.933023, step = 524 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86762\n",
      "INFO:tensorflow:train_accuracy = 0.55376345, train_loss = 19.479372 (0.535 sec)\n",
      "INFO:tensorflow:loss = 19.479372, step = 525 (0.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04387\n",
      "INFO:tensorflow:train_accuracy = 0.5368957, train_loss = 19.894112 (0.489 sec)\n",
      "INFO:tensorflow:loss = 19.894112, step = 526 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95155\n",
      "INFO:tensorflow:train_accuracy = 0.5435898, train_loss = 19.982298 (0.512 sec)\n",
      "INFO:tensorflow:loss = 19.982298, step = 527 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95466\n",
      "INFO:tensorflow:train_accuracy = 0.5401274, train_loss = 19.695343 (0.512 sec)\n",
      "INFO:tensorflow:loss = 19.695343, step = 528 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.35511\n",
      "INFO:tensorflow:train_accuracy = 0.54901963, train_loss = 18.194378 (0.424 sec)\n",
      "INFO:tensorflow:loss = 18.194378, step = 529 (0.424 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12392\n",
      "INFO:tensorflow:train_accuracy = 0.57277626, train_loss = 18.579857 (0.471 sec)\n",
      "INFO:tensorflow:loss = 18.579857, step = 530 (0.471 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.25652\n",
      "INFO:tensorflow:train_accuracy = 0.54993516, train_loss = 19.509533 (0.444 sec)\n",
      "INFO:tensorflow:loss = 19.509533, step = 531 (0.444 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.276\n",
      "INFO:tensorflow:train_accuracy = 0.5539474, train_loss = 18.282904 (0.439 sec)\n",
      "INFO:tensorflow:loss = 18.282904, step = 532 (0.439 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21811\n",
      "INFO:tensorflow:train_accuracy = 0.5473411, train_loss = 19.227774 (0.451 sec)\n",
      "INFO:tensorflow:loss = 19.227774, step = 533 (0.451 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02266\n",
      "INFO:tensorflow:train_accuracy = 0.55249345, train_loss = 17.461918 (0.495 sec)\n",
      "INFO:tensorflow:loss = 17.461918, step = 534 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93411\n",
      "INFO:tensorflow:train_accuracy = 0.50530034, train_loss = 25.688496 (0.517 sec)\n",
      "INFO:tensorflow:loss = 25.688496, step = 535 (0.517 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94166\n",
      "INFO:tensorflow:train_accuracy = 0.5438144, train_loss = 20.149887 (0.515 sec)\n",
      "INFO:tensorflow:loss = 20.149887, step = 536 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97037\n",
      "INFO:tensorflow:train_accuracy = 0.5570118, train_loss = 20.073536 (0.508 sec)\n",
      "INFO:tensorflow:loss = 20.073536, step = 537 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06178\n",
      "INFO:tensorflow:train_accuracy = 0.55006504, train_loss = 18.319881 (0.485 sec)\n",
      "INFO:tensorflow:loss = 18.319881, step = 538 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97268\n",
      "INFO:tensorflow:train_accuracy = 0.53375, train_loss = 20.531145 (0.507 sec)\n",
      "INFO:tensorflow:loss = 20.531145, step = 539 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18803\n",
      "INFO:tensorflow:train_accuracy = 0.54451615, train_loss = 19.559044 (0.457 sec)\n",
      "INFO:tensorflow:loss = 19.559044, step = 540 (0.456 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8605\n",
      "INFO:tensorflow:train_accuracy = 0.52625, train_loss = 21.561512 (0.538 sec)\n",
      "INFO:tensorflow:loss = 21.561512, step = 541 (0.538 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85051\n",
      "INFO:tensorflow:train_accuracy = 0.52919257, train_loss = 20.089968 (0.540 sec)\n",
      "INFO:tensorflow:loss = 20.089968, step = 542 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09263\n",
      "INFO:tensorflow:train_accuracy = 0.55051816, train_loss = 17.837063 (0.478 sec)\n",
      "INFO:tensorflow:loss = 17.837063, step = 543 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98276\n",
      "INFO:tensorflow:train_accuracy = 0.53089535, train_loss = 21.604061 (0.505 sec)\n",
      "INFO:tensorflow:loss = 21.604061, step = 544 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96058\n",
      "INFO:tensorflow:train_accuracy = 0.5472186, train_loss = 19.62613 (0.510 sec)\n",
      "INFO:tensorflow:loss = 19.62613, step = 545 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0661\n",
      "INFO:tensorflow:train_accuracy = 0.5419355, train_loss = 20.345652 (0.484 sec)\n",
      "INFO:tensorflow:loss = 20.345652, step = 546 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0133\n",
      "INFO:tensorflow:train_accuracy = 0.55321944, train_loss = 16.692146 (0.497 sec)\n",
      "INFO:tensorflow:loss = 16.692146, step = 547 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.22491\n",
      "INFO:tensorflow:train_accuracy = 0.5438144, train_loss = 18.47528 (0.449 sec)\n",
      "INFO:tensorflow:loss = 18.47528, step = 548 (0.449 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17691\n",
      "INFO:tensorflow:train_accuracy = 0.5488918, train_loss = 18.02267 (0.459 sec)\n",
      "INFO:tensorflow:loss = 18.02267, step = 549 (0.459 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97223\n",
      "INFO:tensorflow:train_accuracy = 0.5602649, train_loss = 18.828592 (0.507 sec)\n",
      "INFO:tensorflow:loss = 18.828592, step = 550 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 7.78124\n",
      "INFO:tensorflow:train_accuracy = 0.52678573, train_loss = 20.17854 (0.128 sec)\n",
      "INFO:tensorflow:loss = 20.17854, step = 551 (0.128 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.82318\n",
      "INFO:tensorflow:train_accuracy = 0.55283505, train_loss = 19.847902 (0.549 sec)\n",
      "INFO:tensorflow:loss = 19.847902, step = 552 (0.549 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04572\n",
      "INFO:tensorflow:train_accuracy = 0.5423729, train_loss = 19.287708 (0.489 sec)\n",
      "INFO:tensorflow:loss = 19.287708, step = 553 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88184\n",
      "INFO:tensorflow:train_accuracy = 0.5247525, train_loss = 19.850588 (0.531 sec)\n",
      "INFO:tensorflow:loss = 19.850588, step = 554 (0.531 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8211\n",
      "INFO:tensorflow:train_accuracy = 0.4987745, train_loss = 22.323269 (0.549 sec)\n",
      "INFO:tensorflow:loss = 22.323269, step = 555 (0.549 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1034\n",
      "INFO:tensorflow:train_accuracy = 0.55599475, train_loss = 17.764435 (0.475 sec)\n",
      "INFO:tensorflow:loss = 17.764435, step = 556 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03556\n",
      "INFO:tensorflow:train_accuracy = 0.5394402, train_loss = 18.823647 (0.491 sec)\n",
      "INFO:tensorflow:loss = 18.823647, step = 557 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90053\n",
      "INFO:tensorflow:train_accuracy = 0.5599473, train_loss = 15.857203 (0.526 sec)\n",
      "INFO:tensorflow:loss = 15.857203, step = 558 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05844\n",
      "INFO:tensorflow:train_accuracy = 0.5423077, train_loss = 19.850119 (0.486 sec)\n",
      "INFO:tensorflow:loss = 19.850119, step = 559 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04332\n",
      "INFO:tensorflow:train_accuracy = 0.54404145, train_loss = 18.861237 (0.489 sec)\n",
      "INFO:tensorflow:loss = 18.861237, step = 560 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15287\n",
      "INFO:tensorflow:train_accuracy = 0.56241614, train_loss = 16.671545 (0.464 sec)\n",
      "INFO:tensorflow:loss = 16.671545, step = 561 (0.465 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04281\n",
      "INFO:tensorflow:train_accuracy = 0.54860187, train_loss = 16.782202 (0.489 sec)\n",
      "INFO:tensorflow:loss = 16.782202, step = 562 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89489\n",
      "INFO:tensorflow:train_accuracy = 0.55807745, train_loss = 18.485186 (0.528 sec)\n",
      "INFO:tensorflow:loss = 18.485186, step = 563 (0.528 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88904\n",
      "INFO:tensorflow:train_accuracy = 0.5421995, train_loss = 20.354519 (0.529 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 20.354519, step = 564 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03248\n",
      "INFO:tensorflow:train_accuracy = 0.5387597, train_loss = 18.306065 (0.492 sec)\n",
      "INFO:tensorflow:loss = 18.306065, step = 565 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.28494\n",
      "INFO:tensorflow:train_accuracy = 0.56, train_loss = 17.556515 (0.437 sec)\n",
      "INFO:tensorflow:loss = 17.556515, step = 566 (0.437 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04458\n",
      "INFO:tensorflow:train_accuracy = 0.53324807, train_loss = 19.373947 (0.489 sec)\n",
      "INFO:tensorflow:loss = 19.373947, step = 567 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05509\n",
      "INFO:tensorflow:train_accuracy = 0.54699737, train_loss = 20.745226 (0.487 sec)\n",
      "INFO:tensorflow:loss = 20.745226, step = 568 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03045\n",
      "INFO:tensorflow:train_accuracy = 0.53737116, train_loss = 18.778954 (0.493 sec)\n",
      "INFO:tensorflow:loss = 18.778954, step = 569 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09828\n",
      "INFO:tensorflow:train_accuracy = 0.55291003, train_loss = 17.797728 (0.477 sec)\n",
      "INFO:tensorflow:loss = 17.797728, step = 570 (0.477 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04702\n",
      "INFO:tensorflow:train_accuracy = 0.54182756, train_loss = 19.870327 (0.488 sec)\n",
      "INFO:tensorflow:loss = 19.870327, step = 571 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21546\n",
      "INFO:tensorflow:train_accuracy = 0.5596817, train_loss = 17.155134 (0.452 sec)\n",
      "INFO:tensorflow:loss = 17.155134, step = 572 (0.453 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84625\n",
      "INFO:tensorflow:train_accuracy = 0.54356307, train_loss = 19.004797 (0.541 sec)\n",
      "INFO:tensorflow:loss = 19.004797, step = 573 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.26191\n",
      "INFO:tensorflow:train_accuracy = 0.54826254, train_loss = 17.876743 (0.442 sec)\n",
      "INFO:tensorflow:loss = 17.876743, step = 574 (0.443 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90099\n",
      "INFO:tensorflow:train_accuracy = 0.5472186, train_loss = 19.329348 (0.526 sec)\n",
      "INFO:tensorflow:loss = 19.329348, step = 575 (0.526 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2851\n",
      "INFO:tensorflow:train_accuracy = 0.55249345, train_loss = 16.86176 (0.437 sec)\n",
      "INFO:tensorflow:loss = 16.86176, step = 576 (0.437 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17345\n",
      "INFO:tensorflow:train_accuracy = 0.53935486, train_loss = 20.689993 (0.460 sec)\n",
      "INFO:tensorflow:loss = 20.689993, step = 577 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17809\n",
      "INFO:tensorflow:train_accuracy = 0.5638298, train_loss = 16.76937 (0.459 sec)\n",
      "INFO:tensorflow:loss = 16.76937, step = 578 (0.459 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.52041\n",
      "INFO:tensorflow:train_accuracy = 0.5519054, train_loss = 16.722399 (0.397 sec)\n",
      "INFO:tensorflow:loss = 16.722399, step = 579 (0.396 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10939\n",
      "INFO:tensorflow:train_accuracy = 0.54545456, train_loss = 18.30367 (0.474 sec)\n",
      "INFO:tensorflow:loss = 18.30367, step = 580 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01858\n",
      "INFO:tensorflow:train_accuracy = 0.5200486, train_loss = 22.696983 (0.496 sec)\n",
      "INFO:tensorflow:loss = 22.696983, step = 581 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0532\n",
      "INFO:tensorflow:train_accuracy = 0.5584757, train_loss = 18.919016 (0.487 sec)\n",
      "INFO:tensorflow:loss = 18.919016, step = 582 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.27632\n",
      "INFO:tensorflow:train_accuracy = 0.548718, train_loss = 17.859695 (0.439 sec)\n",
      "INFO:tensorflow:loss = 17.859695, step = 583 (0.439 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23834\n",
      "INFO:tensorflow:train_accuracy = 0.5393401, train_loss = 19.194855 (0.447 sec)\n",
      "INFO:tensorflow:loss = 19.194855, step = 584 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15213\n",
      "INFO:tensorflow:train_accuracy = 0.53895277, train_loss = 18.299162 (0.465 sec)\n",
      "INFO:tensorflow:loss = 18.299162, step = 585 (0.464 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21295\n",
      "INFO:tensorflow:train_accuracy = 0.5471204, train_loss = 17.304317 (0.452 sec)\n",
      "INFO:tensorflow:loss = 17.304317, step = 586 (0.452 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.78339\n",
      "INFO:tensorflow:train_accuracy = 0.5225, train_loss = 19.557175 (0.560 sec)\n",
      "INFO:tensorflow:loss = 19.557175, step = 587 (0.560 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94246\n",
      "INFO:tensorflow:train_accuracy = 0.52238804, train_loss = 19.66106 (0.515 sec)\n",
      "INFO:tensorflow:loss = 19.66106, step = 588 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.27452\n",
      "INFO:tensorflow:train_accuracy = 0.5631579, train_loss = 18.79627 (0.440 sec)\n",
      "INFO:tensorflow:loss = 18.79627, step = 589 (0.440 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94772\n",
      "INFO:tensorflow:train_accuracy = 0.52398527, train_loss = 21.698334 (0.514 sec)\n",
      "INFO:tensorflow:loss = 21.698334, step = 590 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1055\n",
      "INFO:tensorflow:train_accuracy = 0.551634, train_loss = 15.718913 (0.475 sec)\n",
      "INFO:tensorflow:loss = 15.718913, step = 591 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05529\n",
      "INFO:tensorflow:train_accuracy = 0.5489691, train_loss = 20.679928 (0.487 sec)\n",
      "INFO:tensorflow:loss = 20.679928, step = 592 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.32056\n",
      "INFO:tensorflow:train_accuracy = 0.55217963, train_loss = 16.007835 (0.430 sec)\n",
      "INFO:tensorflow:loss = 16.007835, step = 593 (0.430 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01009\n",
      "INFO:tensorflow:train_accuracy = 0.5322581, train_loss = 21.857246 (0.498 sec)\n",
      "INFO:tensorflow:loss = 21.857246, step = 594 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05403\n",
      "INFO:tensorflow:train_accuracy = 0.5317259, train_loss = 17.791203 (0.487 sec)\n",
      "INFO:tensorflow:loss = 17.791203, step = 595 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06123\n",
      "INFO:tensorflow:train_accuracy = 0.56738544, train_loss = 14.483719 (0.485 sec)\n",
      "INFO:tensorflow:loss = 14.483719, step = 596 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.79194\n",
      "INFO:tensorflow:train_accuracy = 0.5870647, train_loss = 15.238626 (0.147 sec)\n",
      "INFO:tensorflow:loss = 15.238626, step = 597 (0.147 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83617\n",
      "INFO:tensorflow:train_accuracy = 0.5638999, train_loss = 17.610683 (0.545 sec)\n",
      "INFO:tensorflow:loss = 17.610683, step = 598 (0.545 sec)\n",
      "INFO:tensorflow:Saving checkpoints for 600 into asr-minijasper/model.ckpt.\n",
      "INFO:tensorflow:Calling model_fn.\n",
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Starting evaluation at 2020-10-25T19:13:30Z\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Restoring parameters from asr-minijasper/model.ckpt-600\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n",
      "INFO:tensorflow:Evaluation [1/10]\n",
      "INFO:tensorflow:Evaluation [2/10]\n",
      "INFO:tensorflow:Evaluation [3/10]\n",
      "INFO:tensorflow:Evaluation [4/10]\n",
      "INFO:tensorflow:Evaluation [5/10]\n",
      "INFO:tensorflow:Evaluation [6/10]\n",
      "INFO:tensorflow:Evaluation [7/10]\n",
      "INFO:tensorflow:Evaluation [8/10]\n",
      "INFO:tensorflow:Evaluation [9/10]\n",
      "INFO:tensorflow:Evaluation [10/10]\n",
      "INFO:tensorflow:Finished evaluation at 2020-10-25-19:13:35\n",
      "INFO:tensorflow:Saving dict for global step 600: WER = 1.0, accuracy = 0.56134975, global_step = 600, loss = 20.254488\n",
      "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 600: asr-minijasper/model.ckpt-600\n",
      "INFO:tensorflow:global_step/sec: 0.104338\n",
      "INFO:tensorflow:train_accuracy = 0.53231937, train_loss = 18.902767 (9.584 sec)\n",
      "INFO:tensorflow:loss = 18.902767, step = 599 (9.583 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85645\n",
      "INFO:tensorflow:train_accuracy = 0.4976247, train_loss = 23.89381 (0.540 sec)\n",
      "INFO:tensorflow:loss = 23.89381, step = 600 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15801\n",
      "INFO:tensorflow:train_accuracy = 0.53786904, train_loss = 15.744118 (0.463 sec)\n",
      "INFO:tensorflow:loss = 15.744118, step = 601 (0.463 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05446\n",
      "INFO:tensorflow:train_accuracy = 0.56233424, train_loss = 17.122803 (0.487 sec)\n",
      "INFO:tensorflow:loss = 17.122803, step = 602 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.19152\n",
      "INFO:tensorflow:train_accuracy = 0.5301508, train_loss = 18.16632 (0.456 sec)\n",
      "INFO:tensorflow:loss = 18.16632, step = 603 (0.456 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93008\n",
      "INFO:tensorflow:train_accuracy = 0.5590551, train_loss = 18.502983 (0.518 sec)\n",
      "INFO:tensorflow:loss = 18.502983, step = 604 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98631\n",
      "INFO:tensorflow:train_accuracy = 0.552356, train_loss = 16.32982 (0.504 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 16.32982, step = 605 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16719\n",
      "INFO:tensorflow:train_accuracy = 0.5511811, train_loss = 17.094223 (0.461 sec)\n",
      "INFO:tensorflow:loss = 17.094223, step = 606 (0.461 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04089\n",
      "INFO:tensorflow:train_accuracy = 0.5570118, train_loss = 16.538506 (0.490 sec)\n",
      "INFO:tensorflow:loss = 16.538506, step = 607 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89118\n",
      "INFO:tensorflow:train_accuracy = 0.54376656, train_loss = 16.530426 (0.529 sec)\n",
      "INFO:tensorflow:loss = 16.530426, step = 608 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98693\n",
      "INFO:tensorflow:train_accuracy = 0.55280316, train_loss = 16.956192 (0.503 sec)\n",
      "INFO:tensorflow:loss = 16.956192, step = 609 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94814\n",
      "INFO:tensorflow:train_accuracy = 0.55569947, train_loss = 19.923584 (0.514 sec)\n",
      "INFO:tensorflow:loss = 19.923584, step = 610 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05351\n",
      "INFO:tensorflow:train_accuracy = 0.538163, train_loss = 17.286732 (0.487 sec)\n",
      "INFO:tensorflow:loss = 17.286732, step = 611 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13884\n",
      "INFO:tensorflow:train_accuracy = 0.56233424, train_loss = 17.337208 (0.468 sec)\n",
      "INFO:tensorflow:loss = 17.337208, step = 612 (0.468 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17094\n",
      "INFO:tensorflow:train_accuracy = 0.56842107, train_loss = 16.66847 (0.460 sec)\n",
      "INFO:tensorflow:loss = 16.66847, step = 613 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99378\n",
      "INFO:tensorflow:train_accuracy = 0.54155844, train_loss = 18.580301 (0.502 sec)\n",
      "INFO:tensorflow:loss = 18.580301, step = 614 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94372\n",
      "INFO:tensorflow:train_accuracy = 0.53207546, train_loss = 18.80909 (0.514 sec)\n",
      "INFO:tensorflow:loss = 18.80909, step = 615 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08547\n",
      "INFO:tensorflow:train_accuracy = 0.5706522, train_loss = 15.370531 (0.479 sec)\n",
      "INFO:tensorflow:loss = 15.370531, step = 616 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13321\n",
      "INFO:tensorflow:train_accuracy = 0.5284238, train_loss = 20.681435 (0.469 sec)\n",
      "INFO:tensorflow:loss = 20.681435, step = 617 (0.469 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03479\n",
      "INFO:tensorflow:train_accuracy = 0.5472186, train_loss = 17.782854 (0.491 sec)\n",
      "INFO:tensorflow:loss = 17.782854, step = 618 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96064\n",
      "INFO:tensorflow:train_accuracy = 0.55584085, train_loss = 17.917294 (0.510 sec)\n",
      "INFO:tensorflow:loss = 17.917294, step = 619 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07196\n",
      "INFO:tensorflow:train_accuracy = 0.5332497, train_loss = 17.41357 (0.483 sec)\n",
      "INFO:tensorflow:loss = 17.41357, step = 620 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14437\n",
      "INFO:tensorflow:train_accuracy = 0.5809129, train_loss = 13.875093 (0.466 sec)\n",
      "INFO:tensorflow:loss = 13.875093, step = 621 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12619\n",
      "INFO:tensorflow:train_accuracy = 0.5725699, train_loss = 15.211025 (0.471 sec)\n",
      "INFO:tensorflow:loss = 15.211025, step = 622 (0.471 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95763\n",
      "INFO:tensorflow:train_accuracy = 0.54209185, train_loss = 19.007488 (0.510 sec)\n",
      "INFO:tensorflow:loss = 19.007488, step = 623 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09241\n",
      "INFO:tensorflow:train_accuracy = 0.55439055, train_loss = 17.402632 (0.478 sec)\n",
      "INFO:tensorflow:loss = 17.402632, step = 624 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96431\n",
      "INFO:tensorflow:train_accuracy = 0.5429293, train_loss = 18.049608 (0.509 sec)\n",
      "INFO:tensorflow:loss = 18.049608, step = 625 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92133\n",
      "INFO:tensorflow:train_accuracy = 0.5462725, train_loss = 20.96776 (0.520 sec)\n",
      "INFO:tensorflow:loss = 20.96776, step = 626 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09212\n",
      "INFO:tensorflow:train_accuracy = 0.5539846, train_loss = 16.445278 (0.478 sec)\n",
      "INFO:tensorflow:loss = 16.445278, step = 627 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10487\n",
      "INFO:tensorflow:train_accuracy = 0.5468549, train_loss = 17.24619 (0.475 sec)\n",
      "INFO:tensorflow:loss = 17.24619, step = 628 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00481\n",
      "INFO:tensorflow:train_accuracy = 0.52463055, train_loss = 20.557474 (0.499 sec)\n",
      "INFO:tensorflow:loss = 20.557474, step = 629 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.19295\n",
      "INFO:tensorflow:train_accuracy = 0.55921054, train_loss = 17.047009 (0.456 sec)\n",
      "INFO:tensorflow:loss = 17.047009, step = 630 (0.456 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91509\n",
      "INFO:tensorflow:train_accuracy = 0.5266419, train_loss = 20.301659 (0.522 sec)\n",
      "INFO:tensorflow:loss = 20.301659, step = 631 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00167\n",
      "INFO:tensorflow:train_accuracy = 0.56225425, train_loss = 16.00416 (0.500 sec)\n",
      "INFO:tensorflow:loss = 16.00416, step = 632 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16933\n",
      "INFO:tensorflow:train_accuracy = 0.51920694, train_loss = 17.876707 (0.461 sec)\n",
      "INFO:tensorflow:loss = 17.876707, step = 633 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.74876\n",
      "INFO:tensorflow:train_accuracy = 0.55051816, train_loss = 17.9497 (0.572 sec)\n",
      "INFO:tensorflow:loss = 17.9497, step = 634 (0.572 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0658\n",
      "INFO:tensorflow:train_accuracy = 0.55741936, train_loss = 17.933926 (0.484 sec)\n",
      "INFO:tensorflow:loss = 17.933926, step = 635 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04111\n",
      "INFO:tensorflow:train_accuracy = 0.54092073, train_loss = 18.529854 (0.490 sec)\n",
      "INFO:tensorflow:loss = 18.529854, step = 636 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05078\n",
      "INFO:tensorflow:train_accuracy = 0.51537514, train_loss = 18.871994 (0.488 sec)\n",
      "INFO:tensorflow:loss = 18.871994, step = 637 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88887\n",
      "INFO:tensorflow:train_accuracy = 0.5522581, train_loss = 18.54507 (0.529 sec)\n",
      "INFO:tensorflow:loss = 18.54507, step = 638 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06031\n",
      "INFO:tensorflow:train_accuracy = 0.5545335, train_loss = 17.156868 (0.485 sec)\n",
      "INFO:tensorflow:loss = 17.156868, step = 639 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07244\n",
      "INFO:tensorflow:train_accuracy = 0.54801536, train_loss = 15.388743 (0.482 sec)\n",
      "INFO:tensorflow:loss = 15.388743, step = 640 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93377\n",
      "INFO:tensorflow:train_accuracy = 0.5405405, train_loss = 17.4722 (0.517 sec)\n",
      "INFO:tensorflow:loss = 17.4722, step = 641 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05245\n",
      "INFO:tensorflow:train_accuracy = 0.57604307, train_loss = 15.261207 (0.487 sec)\n",
      "INFO:tensorflow:loss = 15.261207, step = 642 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.44607\n",
      "INFO:tensorflow:train_accuracy = 0.5652174, train_loss = 14.228175 (0.184 sec)\n",
      "INFO:tensorflow:loss = 14.228175, step = 643 (0.184 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.80045\n",
      "INFO:tensorflow:train_accuracy = 0.5582902, train_loss = 17.578629 (0.556 sec)\n",
      "INFO:tensorflow:loss = 17.578629, step = 644 (0.556 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09722\n",
      "INFO:tensorflow:train_accuracy = 0.5551181, train_loss = 16.754454 (0.477 sec)\n",
      "INFO:tensorflow:loss = 16.754454, step = 645 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91786\n",
      "INFO:tensorflow:train_accuracy = 0.5012195, train_loss = 19.525963 (0.521 sec)\n",
      "INFO:tensorflow:loss = 19.525963, step = 646 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9691\n",
      "INFO:tensorflow:train_accuracy = 0.5580808, train_loss = 16.873873 (0.508 sec)\n",
      "INFO:tensorflow:loss = 16.873873, step = 647 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04882\n",
      "INFO:tensorflow:train_accuracy = 0.55221933, train_loss = 16.264812 (0.488 sec)\n",
      "INFO:tensorflow:loss = 16.264812, step = 648 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86574\n",
      "INFO:tensorflow:train_accuracy = 0.53875, train_loss = 18.346207 (0.536 sec)\n",
      "INFO:tensorflow:loss = 18.346207, step = 649 (0.536 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07571\n",
      "INFO:tensorflow:train_accuracy = 0.5535248, train_loss = 16.130154 (0.482 sec)\n",
      "INFO:tensorflow:loss = 16.130154, step = 650 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.30645\n",
      "INFO:tensorflow:train_accuracy = 0.5652174, train_loss = 17.226479 (0.434 sec)\n",
      "INFO:tensorflow:loss = 17.226479, step = 651 (0.434 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98889\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.5557012, train_loss = 17.996857 (0.503 sec)\n",
      "INFO:tensorflow:loss = 17.996857, step = 652 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8902\n",
      "INFO:tensorflow:train_accuracy = 0.54497355, train_loss = 15.624594 (0.529 sec)\n",
      "INFO:tensorflow:loss = 15.624594, step = 653 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84327\n",
      "INFO:tensorflow:train_accuracy = 0.55921054, train_loss = 17.04792 (0.543 sec)\n",
      "INFO:tensorflow:loss = 17.04792, step = 654 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85703\n",
      "INFO:tensorflow:train_accuracy = 0.55974025, train_loss = 14.514616 (0.539 sec)\n",
      "INFO:tensorflow:loss = 14.514616, step = 655 (0.538 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16514\n",
      "INFO:tensorflow:train_accuracy = 0.56299216, train_loss = 16.508348 (0.462 sec)\n",
      "INFO:tensorflow:loss = 16.508348, step = 656 (0.462 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21282\n",
      "INFO:tensorflow:train_accuracy = 0.5716216, train_loss = 16.497446 (0.452 sec)\n",
      "INFO:tensorflow:loss = 16.497446, step = 657 (0.452 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15407\n",
      "INFO:tensorflow:train_accuracy = 0.5546772, train_loss = 15.351189 (0.464 sec)\n",
      "INFO:tensorflow:loss = 15.351189, step = 658 (0.464 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24622\n",
      "INFO:tensorflow:train_accuracy = 0.5353925, train_loss = 18.483686 (0.445 sec)\n",
      "INFO:tensorflow:loss = 18.483686, step = 659 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18257\n",
      "INFO:tensorflow:train_accuracy = 0.51715684, train_loss = 19.209564 (0.458 sec)\n",
      "INFO:tensorflow:loss = 19.209564, step = 660 (0.459 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92975\n",
      "INFO:tensorflow:train_accuracy = 0.54699737, train_loss = 14.15717 (0.518 sec)\n",
      "INFO:tensorflow:loss = 14.15717, step = 661 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11447\n",
      "INFO:tensorflow:train_accuracy = 0.56738544, train_loss = 14.965143 (0.473 sec)\n",
      "INFO:tensorflow:loss = 14.965143, step = 662 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89153\n",
      "INFO:tensorflow:train_accuracy = 0.54172015, train_loss = 20.165564 (0.529 sec)\n",
      "INFO:tensorflow:loss = 20.165564, step = 663 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99404\n",
      "INFO:tensorflow:train_accuracy = 0.5621693, train_loss = 17.08111 (0.502 sec)\n",
      "INFO:tensorflow:loss = 17.08111, step = 664 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12904\n",
      "INFO:tensorflow:train_accuracy = 0.5611038, train_loss = 15.272921 (0.470 sec)\n",
      "INFO:tensorflow:loss = 15.272921, step = 665 (0.470 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.70236\n",
      "INFO:tensorflow:train_accuracy = 0.5519481, train_loss = 18.122852 (0.588 sec)\n",
      "INFO:tensorflow:loss = 18.122852, step = 666 (0.588 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.66066\n",
      "INFO:tensorflow:train_accuracy = 0.53759396, train_loss = 18.444584 (0.601 sec)\n",
      "INFO:tensorflow:loss = 18.444584, step = 667 (0.601 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09569\n",
      "INFO:tensorflow:train_accuracy = 0.5670241, train_loss = 16.128094 (0.477 sec)\n",
      "INFO:tensorflow:loss = 16.128094, step = 668 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99072\n",
      "INFO:tensorflow:train_accuracy = 0.5576159, train_loss = 15.306174 (0.502 sec)\n",
      "INFO:tensorflow:loss = 15.306174, step = 669 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17589\n",
      "INFO:tensorflow:train_accuracy = 0.5746073, train_loss = 16.355423 (0.459 sec)\n",
      "INFO:tensorflow:loss = 16.355423, step = 670 (0.459 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98806\n",
      "INFO:tensorflow:train_accuracy = 0.55886155, train_loss = 15.937777 (0.503 sec)\n",
      "INFO:tensorflow:loss = 15.937777, step = 671 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10033\n",
      "INFO:tensorflow:train_accuracy = 0.5485123, train_loss = 18.715916 (0.476 sec)\n",
      "INFO:tensorflow:loss = 18.715916, step = 672 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1447\n",
      "INFO:tensorflow:train_accuracy = 0.51897186, train_loss = 18.585045 (0.466 sec)\n",
      "INFO:tensorflow:loss = 18.585045, step = 673 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24497\n",
      "INFO:tensorflow:train_accuracy = 0.56217617, train_loss = 16.565592 (0.445 sec)\n",
      "INFO:tensorflow:loss = 16.565592, step = 674 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14068\n",
      "INFO:tensorflow:train_accuracy = 0.54333764, train_loss = 16.641264 (0.467 sec)\n",
      "INFO:tensorflow:loss = 16.641264, step = 675 (0.467 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00448\n",
      "INFO:tensorflow:train_accuracy = 0.54967743, train_loss = 17.167444 (0.499 sec)\n",
      "INFO:tensorflow:loss = 17.167444, step = 676 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88459\n",
      "INFO:tensorflow:train_accuracy = 0.51966625, train_loss = 21.339148 (0.530 sec)\n",
      "INFO:tensorflow:loss = 21.339148, step = 677 (0.530 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92844\n",
      "INFO:tensorflow:train_accuracy = 0.54755783, train_loss = 16.636156 (0.519 sec)\n",
      "INFO:tensorflow:loss = 16.636156, step = 678 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24612\n",
      "INFO:tensorflow:train_accuracy = 0.54464287, train_loss = 16.937824 (0.445 sec)\n",
      "INFO:tensorflow:loss = 16.937824, step = 679 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1113\n",
      "INFO:tensorflow:train_accuracy = 0.545108, train_loss = 17.244553 (0.474 sec)\n",
      "INFO:tensorflow:loss = 17.244553, step = 680 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91305\n",
      "INFO:tensorflow:train_accuracy = 0.540404, train_loss = 18.229557 (0.523 sec)\n",
      "INFO:tensorflow:loss = 18.229557, step = 681 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98088\n",
      "INFO:tensorflow:train_accuracy = 0.57011795, train_loss = 15.492936 (0.505 sec)\n",
      "INFO:tensorflow:loss = 15.492936, step = 682 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0332\n",
      "INFO:tensorflow:train_accuracy = 0.5363057, train_loss = 17.823296 (0.492 sec)\n",
      "INFO:tensorflow:loss = 17.823296, step = 683 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06252\n",
      "INFO:tensorflow:train_accuracy = 0.541507, train_loss = 17.999493 (0.485 sec)\n",
      "INFO:tensorflow:loss = 17.999493, step = 684 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03047\n",
      "INFO:tensorflow:train_accuracy = 0.52962965, train_loss = 20.084177 (0.493 sec)\n",
      "INFO:tensorflow:loss = 20.084177, step = 685 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06575\n",
      "INFO:tensorflow:train_accuracy = 0.57104915, train_loss = 15.018486 (0.484 sec)\n",
      "INFO:tensorflow:loss = 15.018486, step = 686 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03523\n",
      "INFO:tensorflow:train_accuracy = 0.57563585, train_loss = 16.288977 (0.491 sec)\n",
      "INFO:tensorflow:loss = 16.288977, step = 687 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05747\n",
      "INFO:tensorflow:train_accuracy = 0.560425, train_loss = 14.044493 (0.486 sec)\n",
      "INFO:tensorflow:loss = 14.044493, step = 688 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.2768\n",
      "INFO:tensorflow:train_accuracy = 0.529148, train_loss = 16.057096 (0.190 sec)\n",
      "INFO:tensorflow:loss = 16.057096, step = 689 (0.190 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79841\n",
      "INFO:tensorflow:train_accuracy = 0.5555556, train_loss = 17.492907 (0.556 sec)\n",
      "INFO:tensorflow:loss = 17.492907, step = 690 (0.555 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85467\n",
      "INFO:tensorflow:train_accuracy = 0.5401003, train_loss = 18.81847 (0.540 sec)\n",
      "INFO:tensorflow:loss = 18.81847, step = 691 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02851\n",
      "INFO:tensorflow:train_accuracy = 0.562341, train_loss = 16.070938 (0.493 sec)\n",
      "INFO:tensorflow:loss = 16.070938, step = 692 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98293\n",
      "INFO:tensorflow:train_accuracy = 0.5409836, train_loss = 17.5421 (0.505 sec)\n",
      "INFO:tensorflow:loss = 17.5421, step = 693 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97769\n",
      "INFO:tensorflow:train_accuracy = 0.525641, train_loss = 15.882059 (0.506 sec)\n",
      "INFO:tensorflow:loss = 15.882059, step = 694 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00735\n",
      "INFO:tensorflow:train_accuracy = 0.56349206, train_loss = 15.326786 (0.498 sec)\n",
      "INFO:tensorflow:loss = 15.326786, step = 695 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0892\n",
      "INFO:tensorflow:train_accuracy = 0.53797466, train_loss = 17.042555 (0.479 sec)\n",
      "INFO:tensorflow:loss = 17.042555, step = 696 (0.478 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20561\n",
      "INFO:tensorflow:train_accuracy = 0.56666666, train_loss = 15.261652 (0.454 sec)\n",
      "INFO:tensorflow:loss = 15.261652, step = 697 (0.454 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00964\n",
      "INFO:tensorflow:train_accuracy = 0.5564202, train_loss = 15.083441 (0.498 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 15.083441, step = 698 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95666\n",
      "INFO:tensorflow:train_accuracy = 0.56958765, train_loss = 15.969747 (0.511 sec)\n",
      "INFO:tensorflow:loss = 15.969747, step = 699 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0085\n",
      "INFO:tensorflow:train_accuracy = 0.5771812, train_loss = 16.2583 (0.499 sec)\n",
      "INFO:tensorflow:loss = 16.2583, step = 700 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0549\n",
      "INFO:tensorflow:train_accuracy = 0.56891894, train_loss = 14.250521 (0.486 sec)\n",
      "INFO:tensorflow:loss = 14.250521, step = 701 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00277\n",
      "INFO:tensorflow:train_accuracy = 0.53992397, train_loss = 18.020039 (0.499 sec)\n",
      "INFO:tensorflow:loss = 18.020039, step = 702 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06814\n",
      "INFO:tensorflow:train_accuracy = 0.5411765, train_loss = 14.789544 (0.483 sec)\n",
      "INFO:tensorflow:loss = 14.789544, step = 703 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87718\n",
      "INFO:tensorflow:train_accuracy = 0.5803453, train_loss = 15.085503 (0.533 sec)\n",
      "INFO:tensorflow:loss = 15.085503, step = 704 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02439\n",
      "INFO:tensorflow:train_accuracy = 0.5541069, train_loss = 16.74493 (0.494 sec)\n",
      "INFO:tensorflow:loss = 16.74493, step = 705 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05456\n",
      "INFO:tensorflow:train_accuracy = 0.5555556, train_loss = 18.732037 (0.487 sec)\n",
      "INFO:tensorflow:loss = 18.732037, step = 706 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04027\n",
      "INFO:tensorflow:train_accuracy = 0.5631579, train_loss = 16.303913 (0.490 sec)\n",
      "INFO:tensorflow:loss = 16.303913, step = 707 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91694\n",
      "INFO:tensorflow:train_accuracy = 0.5489691, train_loss = 16.28897 (0.522 sec)\n",
      "INFO:tensorflow:loss = 16.28897, step = 708 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94775\n",
      "INFO:tensorflow:train_accuracy = 0.5708609, train_loss = 14.813299 (0.513 sec)\n",
      "INFO:tensorflow:loss = 14.813299, step = 709 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92226\n",
      "INFO:tensorflow:train_accuracy = 0.55440414, train_loss = 15.348703 (0.520 sec)\n",
      "INFO:tensorflow:loss = 15.348703, step = 710 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99632\n",
      "INFO:tensorflow:train_accuracy = 0.5627426, train_loss = 15.738184 (0.501 sec)\n",
      "INFO:tensorflow:loss = 15.738184, step = 711 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16309\n",
      "INFO:tensorflow:train_accuracy = 0.5503876, train_loss = 14.581568 (0.462 sec)\n",
      "INFO:tensorflow:loss = 14.581568, step = 712 (0.461 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92625\n",
      "INFO:tensorflow:train_accuracy = 0.5710526, train_loss = 16.150208 (0.519 sec)\n",
      "INFO:tensorflow:loss = 16.150208, step = 713 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01501\n",
      "INFO:tensorflow:train_accuracy = 0.5759577, train_loss = 15.027211 (0.497 sec)\n",
      "INFO:tensorflow:loss = 15.027211, step = 714 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96062\n",
      "INFO:tensorflow:train_accuracy = 0.5708609, train_loss = 16.852825 (0.510 sec)\n",
      "INFO:tensorflow:loss = 16.852825, step = 715 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14482\n",
      "INFO:tensorflow:train_accuracy = 0.5493507, train_loss = 15.125434 (0.466 sec)\n",
      "INFO:tensorflow:loss = 15.125434, step = 716 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94483\n",
      "INFO:tensorflow:train_accuracy = 0.5411471, train_loss = 17.16237 (0.514 sec)\n",
      "INFO:tensorflow:loss = 17.16237, step = 717 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98194\n",
      "INFO:tensorflow:train_accuracy = 0.5643045, train_loss = 15.106334 (0.505 sec)\n",
      "INFO:tensorflow:loss = 15.106334, step = 718 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92168\n",
      "INFO:tensorflow:train_accuracy = 0.5241038, train_loss = 17.4911 (0.520 sec)\n",
      "INFO:tensorflow:loss = 17.4911, step = 719 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01984\n",
      "INFO:tensorflow:train_accuracy = 0.5649351, train_loss = 16.1831 (0.495 sec)\n",
      "INFO:tensorflow:loss = 16.1831, step = 720 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02297\n",
      "INFO:tensorflow:train_accuracy = 0.5406091, train_loss = 18.608055 (0.494 sec)\n",
      "INFO:tensorflow:loss = 18.608055, step = 721 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95471\n",
      "INFO:tensorflow:train_accuracy = 0.5474359, train_loss = 16.48934 (0.512 sec)\n",
      "INFO:tensorflow:loss = 16.48934, step = 722 (0.512 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20985\n",
      "INFO:tensorflow:train_accuracy = 0.545, train_loss = 15.311885 (0.452 sec)\n",
      "INFO:tensorflow:loss = 15.311885, step = 723 (0.452 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98218\n",
      "INFO:tensorflow:train_accuracy = 0.5467349, train_loss = 16.460855 (0.505 sec)\n",
      "INFO:tensorflow:loss = 16.460855, step = 724 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90885\n",
      "INFO:tensorflow:train_accuracy = 0.54285717, train_loss = 16.906317 (0.524 sec)\n",
      "INFO:tensorflow:loss = 16.906317, step = 725 (0.524 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03618\n",
      "INFO:tensorflow:train_accuracy = 0.5466156, train_loss = 17.61096 (0.491 sec)\n",
      "INFO:tensorflow:loss = 17.61096, step = 726 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92377\n",
      "INFO:tensorflow:train_accuracy = 0.5467532, train_loss = 17.46843 (0.520 sec)\n",
      "INFO:tensorflow:loss = 17.46843, step = 727 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08492\n",
      "INFO:tensorflow:train_accuracy = 0.54347825, train_loss = 16.356245 (0.480 sec)\n",
      "INFO:tensorflow:loss = 16.356245, step = 728 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95915\n",
      "INFO:tensorflow:train_accuracy = 0.5767196, train_loss = 14.870685 (0.510 sec)\n",
      "INFO:tensorflow:loss = 14.870685, step = 729 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99386\n",
      "INFO:tensorflow:train_accuracy = 0.53467846, train_loss = 16.13652 (0.502 sec)\n",
      "INFO:tensorflow:loss = 16.13652, step = 730 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.22989\n",
      "INFO:tensorflow:train_accuracy = 0.56662357, train_loss = 16.17419 (0.448 sec)\n",
      "INFO:tensorflow:loss = 16.17419, step = 731 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00448\n",
      "INFO:tensorflow:train_accuracy = 0.53816795, train_loss = 14.918988 (0.499 sec)\n",
      "INFO:tensorflow:loss = 14.918988, step = 732 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79208\n",
      "INFO:tensorflow:train_accuracy = 0.5536869, train_loss = 15.324485 (0.558 sec)\n",
      "INFO:tensorflow:loss = 15.324485, step = 733 (0.559 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24385\n",
      "INFO:tensorflow:train_accuracy = 0.5818674, train_loss = 13.873824 (0.445 sec)\n",
      "INFO:tensorflow:loss = 13.873824, step = 734 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.48546\n",
      "INFO:tensorflow:train_accuracy = 0.5263158, train_loss = 14.581371 (0.154 sec)\n",
      "INFO:tensorflow:loss = 14.581371, step = 735 (0.154 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89156\n",
      "INFO:tensorflow:train_accuracy = 0.57030225, train_loss = 14.90163 (0.529 sec)\n",
      "INFO:tensorflow:loss = 14.90163, step = 736 (0.530 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84371\n",
      "INFO:tensorflow:train_accuracy = 0.5598991, train_loss = 17.01493 (0.542 sec)\n",
      "INFO:tensorflow:loss = 17.01493, step = 737 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11209\n",
      "INFO:tensorflow:train_accuracy = 0.517284, train_loss = 17.897488 (0.473 sec)\n",
      "INFO:tensorflow:loss = 17.897488, step = 738 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79772\n",
      "INFO:tensorflow:train_accuracy = 0.55484694, train_loss = 17.92547 (0.556 sec)\n",
      "INFO:tensorflow:loss = 17.92547, step = 739 (0.556 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00578\n",
      "INFO:tensorflow:train_accuracy = 0.55938697, train_loss = 14.790125 (0.499 sec)\n",
      "INFO:tensorflow:loss = 14.790125, step = 740 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08179\n",
      "INFO:tensorflow:train_accuracy = 0.56763923, train_loss = 13.738331 (0.480 sec)\n",
      "INFO:tensorflow:loss = 13.738331, step = 741 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03873\n",
      "INFO:tensorflow:train_accuracy = 0.5582902, train_loss = 16.026657 (0.491 sec)\n",
      "INFO:tensorflow:loss = 16.026657, step = 742 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00799\n",
      "INFO:tensorflow:train_accuracy = 0.55025774, train_loss = 14.94128 (0.498 sec)\n",
      "INFO:tensorflow:loss = 14.94128, step = 743 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05866\n",
      "INFO:tensorflow:train_accuracy = 0.55307996, train_loss = 13.52043 (0.486 sec)\n",
      "INFO:tensorflow:loss = 13.52043, step = 744 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11097\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.55283505, train_loss = 16.079124 (0.474 sec)\n",
      "INFO:tensorflow:loss = 16.079124, step = 745 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01307\n",
      "INFO:tensorflow:train_accuracy = 0.55496687, train_loss = 14.189272 (0.497 sec)\n",
      "INFO:tensorflow:loss = 14.189272, step = 746 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84892\n",
      "INFO:tensorflow:train_accuracy = 0.5954606, train_loss = 14.728055 (0.541 sec)\n",
      "INFO:tensorflow:loss = 14.728055, step = 747 (0.541 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13185\n",
      "INFO:tensorflow:train_accuracy = 0.55627424, train_loss = 14.389107 (0.469 sec)\n",
      "INFO:tensorflow:loss = 14.389107, step = 748 (0.468 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24031\n",
      "INFO:tensorflow:train_accuracy = 0.5400517, train_loss = 16.996933 (0.447 sec)\n",
      "INFO:tensorflow:loss = 16.996933, step = 749 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96657\n",
      "INFO:tensorflow:train_accuracy = 0.5680317, train_loss = 14.072387 (0.508 sec)\n",
      "INFO:tensorflow:loss = 14.072387, step = 750 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.2621\n",
      "INFO:tensorflow:train_accuracy = 0.5846774, train_loss = 13.110913 (0.442 sec)\n",
      "INFO:tensorflow:loss = 13.110913, step = 751 (0.442 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9192\n",
      "INFO:tensorflow:train_accuracy = 0.5513784, train_loss = 19.525679 (0.521 sec)\n",
      "INFO:tensorflow:loss = 19.525679, step = 752 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.34424\n",
      "INFO:tensorflow:train_accuracy = 0.5363057, train_loss = 16.006866 (0.427 sec)\n",
      "INFO:tensorflow:loss = 16.006866, step = 753 (0.427 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.30336\n",
      "INFO:tensorflow:train_accuracy = 0.56666666, train_loss = 12.44589 (0.434 sec)\n",
      "INFO:tensorflow:loss = 12.44589, step = 754 (0.434 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0409\n",
      "INFO:tensorflow:train_accuracy = 0.568, train_loss = 15.442679 (0.490 sec)\n",
      "INFO:tensorflow:loss = 15.442679, step = 755 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06774\n",
      "INFO:tensorflow:train_accuracy = 0.5375796, train_loss = 15.715843 (0.484 sec)\n",
      "INFO:tensorflow:loss = 15.715843, step = 756 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97002\n",
      "INFO:tensorflow:train_accuracy = 0.5513308, train_loss = 16.569569 (0.508 sec)\n",
      "INFO:tensorflow:loss = 16.569569, step = 757 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03284\n",
      "INFO:tensorflow:train_accuracy = 0.58243245, train_loss = 15.00611 (0.492 sec)\n",
      "INFO:tensorflow:loss = 15.00611, step = 758 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98947\n",
      "INFO:tensorflow:train_accuracy = 0.54909563, train_loss = 13.929445 (0.503 sec)\n",
      "INFO:tensorflow:loss = 13.929445, step = 759 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02075\n",
      "INFO:tensorflow:train_accuracy = 0.5755968, train_loss = 13.150173 (0.495 sec)\n",
      "INFO:tensorflow:loss = 13.150173, step = 760 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99729\n",
      "INFO:tensorflow:train_accuracy = 0.5503268, train_loss = 14.9418125 (0.501 sec)\n",
      "INFO:tensorflow:loss = 14.9418125, step = 761 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9701\n",
      "INFO:tensorflow:train_accuracy = 0.55974025, train_loss = 14.565777 (0.507 sec)\n",
      "INFO:tensorflow:loss = 14.565777, step = 762 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91907\n",
      "INFO:tensorflow:train_accuracy = 0.5440806, train_loss = 16.602589 (0.521 sec)\n",
      "INFO:tensorflow:loss = 16.602589, step = 763 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96863\n",
      "INFO:tensorflow:train_accuracy = 0.5500634, train_loss = 16.288204 (0.508 sec)\n",
      "INFO:tensorflow:loss = 16.288204, step = 764 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20588\n",
      "INFO:tensorflow:train_accuracy = 0.5652742, train_loss = 14.662598 (0.453 sec)\n",
      "INFO:tensorflow:loss = 14.662598, step = 765 (0.453 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9287\n",
      "INFO:tensorflow:train_accuracy = 0.56467664, train_loss = 16.925262 (0.518 sec)\n",
      "INFO:tensorflow:loss = 16.925262, step = 766 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91216\n",
      "INFO:tensorflow:train_accuracy = 0.54303795, train_loss = 17.38128 (0.523 sec)\n",
      "INFO:tensorflow:loss = 17.38128, step = 767 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02814\n",
      "INFO:tensorflow:train_accuracy = 0.5538262, train_loss = 12.782017 (0.493 sec)\n",
      "INFO:tensorflow:loss = 12.782017, step = 768 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06277\n",
      "INFO:tensorflow:train_accuracy = 0.5530587, train_loss = 16.664242 (0.485 sec)\n",
      "INFO:tensorflow:loss = 16.664242, step = 769 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14282\n",
      "INFO:tensorflow:train_accuracy = 0.54533845, train_loss = 15.723038 (0.466 sec)\n",
      "INFO:tensorflow:loss = 15.723038, step = 770 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9097\n",
      "INFO:tensorflow:train_accuracy = 0.54245883, train_loss = 15.116873 (0.524 sec)\n",
      "INFO:tensorflow:loss = 15.116873, step = 771 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95805\n",
      "INFO:tensorflow:train_accuracy = 0.53144264, train_loss = 17.201796 (0.511 sec)\n",
      "INFO:tensorflow:loss = 17.201796, step = 772 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24292\n",
      "INFO:tensorflow:train_accuracy = 0.5625, train_loss = 15.011303 (0.446 sec)\n",
      "INFO:tensorflow:loss = 15.011303, step = 773 (0.446 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92028\n",
      "INFO:tensorflow:train_accuracy = 0.53729457, train_loss = 15.336208 (0.521 sec)\n",
      "INFO:tensorflow:loss = 15.336208, step = 774 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9639\n",
      "INFO:tensorflow:train_accuracy = 0.56970507, train_loss = 14.288024 (0.509 sec)\n",
      "INFO:tensorflow:loss = 14.288024, step = 775 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03041\n",
      "INFO:tensorflow:train_accuracy = 0.54909563, train_loss = 13.415636 (0.493 sec)\n",
      "INFO:tensorflow:loss = 13.415636, step = 776 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.66941\n",
      "INFO:tensorflow:train_accuracy = 0.5496183, train_loss = 15.054264 (0.599 sec)\n",
      "INFO:tensorflow:loss = 15.054264, step = 777 (0.599 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92732\n",
      "INFO:tensorflow:train_accuracy = 0.54452926, train_loss = 15.793951 (0.519 sec)\n",
      "INFO:tensorflow:loss = 15.793951, step = 778 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.19318\n",
      "INFO:tensorflow:train_accuracy = 0.57880795, train_loss = 13.164606 (0.456 sec)\n",
      "INFO:tensorflow:loss = 13.164606, step = 779 (0.456 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03773\n",
      "INFO:tensorflow:train_accuracy = 0.58736557, train_loss = 13.262672 (0.490 sec)\n",
      "INFO:tensorflow:loss = 13.262672, step = 780 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.61528\n",
      "INFO:tensorflow:train_accuracy = 0.5598086, train_loss = 9.79811 (0.178 sec)\n",
      "INFO:tensorflow:loss = 9.79811, step = 781 (0.178 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00625\n",
      "INFO:tensorflow:train_accuracy = 0.5646438, train_loss = 14.093603 (0.499 sec)\n",
      "INFO:tensorflow:loss = 14.093603, step = 782 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23298\n",
      "INFO:tensorflow:train_accuracy = 0.55484694, train_loss = 14.312557 (0.448 sec)\n",
      "INFO:tensorflow:loss = 14.312557, step = 783 (0.448 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92985\n",
      "INFO:tensorflow:train_accuracy = 0.5458996, train_loss = 16.309082 (0.518 sec)\n",
      "INFO:tensorflow:loss = 16.309082, step = 784 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03717\n",
      "INFO:tensorflow:train_accuracy = 0.56072354, train_loss = 14.886456 (0.491 sec)\n",
      "INFO:tensorflow:loss = 14.886456, step = 785 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24301\n",
      "INFO:tensorflow:train_accuracy = 0.5568476, train_loss = 12.397144 (0.446 sec)\n",
      "INFO:tensorflow:loss = 12.397144, step = 786 (0.446 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16153\n",
      "INFO:tensorflow:train_accuracy = 0.5747724, train_loss = 14.298371 (0.462 sec)\n",
      "INFO:tensorflow:loss = 14.298371, step = 787 (0.463 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02136\n",
      "INFO:tensorflow:train_accuracy = 0.53729457, train_loss = 16.579018 (0.495 sec)\n",
      "INFO:tensorflow:loss = 16.579018, step = 788 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.25201\n",
      "INFO:tensorflow:train_accuracy = 0.5750332, train_loss = 13.74535 (0.445 sec)\n",
      "INFO:tensorflow:loss = 13.74535, step = 789 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21924\n",
      "INFO:tensorflow:train_accuracy = 0.5467349, train_loss = 13.207643 (0.451 sec)\n",
      "INFO:tensorflow:loss = 13.207643, step = 790 (0.451 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05376\n",
      "INFO:tensorflow:train_accuracy = 0.56119794, train_loss = 14.665762 (0.487 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 14.665762, step = 791 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90767\n",
      "INFO:tensorflow:train_accuracy = 0.56993467, train_loss = 14.51587 (0.524 sec)\n",
      "INFO:tensorflow:loss = 14.51587, step = 792 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.34502\n",
      "INFO:tensorflow:train_accuracy = 0.5693527, train_loss = 12.016697 (0.426 sec)\n",
      "INFO:tensorflow:loss = 12.016697, step = 793 (0.426 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05305\n",
      "INFO:tensorflow:train_accuracy = 0.5752688, train_loss = 13.68531 (0.487 sec)\n",
      "INFO:tensorflow:loss = 13.68531, step = 794 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96775\n",
      "INFO:tensorflow:train_accuracy = 0.5470738, train_loss = 15.799996 (0.508 sec)\n",
      "INFO:tensorflow:loss = 15.799996, step = 795 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83103\n",
      "INFO:tensorflow:train_accuracy = 0.5737052, train_loss = 12.753255 (0.546 sec)\n",
      "INFO:tensorflow:loss = 12.753255, step = 796 (0.546 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98852\n",
      "INFO:tensorflow:train_accuracy = 0.5719947, train_loss = 14.782744 (0.503 sec)\n",
      "INFO:tensorflow:loss = 14.782744, step = 797 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96723\n",
      "INFO:tensorflow:train_accuracy = 0.5498721, train_loss = 16.023933 (0.509 sec)\n",
      "INFO:tensorflow:loss = 16.023933, step = 798 (0.509 sec)\n",
      "INFO:tensorflow:Saving checkpoints for 800 into asr-minijasper/model.ckpt.\n",
      "INFO:tensorflow:Skip the current checkpoint eval due to throttle secs (120 secs).\n",
      "INFO:tensorflow:global_step/sec: 0.689342\n",
      "INFO:tensorflow:train_accuracy = 0.54215306, train_loss = 14.888674 (1.450 sec)\n",
      "INFO:tensorflow:loss = 14.888674, step = 799 (1.450 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08428\n",
      "INFO:tensorflow:train_accuracy = 0.54953766, train_loss = 13.1358 (0.481 sec)\n",
      "INFO:tensorflow:loss = 13.1358, step = 800 (0.481 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99437\n",
      "INFO:tensorflow:train_accuracy = 0.566838, train_loss = 14.355815 (0.501 sec)\n",
      "INFO:tensorflow:loss = 14.355815, step = 801 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.20936\n",
      "INFO:tensorflow:train_accuracy = 0.51875, train_loss = 15.834797 (0.453 sec)\n",
      "INFO:tensorflow:loss = 15.834797, step = 802 (0.453 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91559\n",
      "INFO:tensorflow:train_accuracy = 0.5883134, train_loss = 12.945238 (0.522 sec)\n",
      "INFO:tensorflow:loss = 12.945238, step = 803 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24859\n",
      "INFO:tensorflow:train_accuracy = 0.5648148, train_loss = 12.934048 (0.445 sec)\n",
      "INFO:tensorflow:loss = 12.934048, step = 804 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95438\n",
      "INFO:tensorflow:train_accuracy = 0.56176853, train_loss = 13.080328 (0.512 sec)\n",
      "INFO:tensorflow:loss = 13.080328, step = 805 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.13663\n",
      "INFO:tensorflow:train_accuracy = 0.56233764, train_loss = 12.0605755 (0.468 sec)\n",
      "INFO:tensorflow:loss = 12.0605755, step = 806 (0.468 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95267\n",
      "INFO:tensorflow:train_accuracy = 0.57310706, train_loss = 15.9175205 (0.512 sec)\n",
      "INFO:tensorflow:loss = 15.9175205, step = 807 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00389\n",
      "INFO:tensorflow:train_accuracy = 0.5716163, train_loss = 13.4543915 (0.499 sec)\n",
      "INFO:tensorflow:loss = 13.4543915, step = 808 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.83056\n",
      "INFO:tensorflow:train_accuracy = 0.56766915, train_loss = 16.539408 (0.547 sec)\n",
      "INFO:tensorflow:loss = 16.539408, step = 809 (0.547 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.14152\n",
      "INFO:tensorflow:train_accuracy = 0.5422446, train_loss = 13.928533 (0.467 sec)\n",
      "INFO:tensorflow:loss = 13.928533, step = 810 (0.467 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0526\n",
      "INFO:tensorflow:train_accuracy = 0.5558376, train_loss = 14.995641 (0.487 sec)\n",
      "INFO:tensorflow:loss = 14.995641, step = 811 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06045\n",
      "INFO:tensorflow:train_accuracy = 0.55641025, train_loss = 14.937708 (0.486 sec)\n",
      "INFO:tensorflow:loss = 14.937708, step = 812 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84022\n",
      "INFO:tensorflow:train_accuracy = 0.56806284, train_loss = 14.76547 (0.543 sec)\n",
      "INFO:tensorflow:loss = 14.76547, step = 813 (0.543 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95448\n",
      "INFO:tensorflow:train_accuracy = 0.5664516, train_loss = 12.805725 (0.511 sec)\n",
      "INFO:tensorflow:loss = 12.805725, step = 814 (0.511 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97867\n",
      "INFO:tensorflow:train_accuracy = 0.54810125, train_loss = 14.219934 (0.506 sec)\n",
      "INFO:tensorflow:loss = 14.219934, step = 815 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89987\n",
      "INFO:tensorflow:train_accuracy = 0.55816996, train_loss = 13.946906 (0.526 sec)\n",
      "INFO:tensorflow:loss = 13.946906, step = 816 (0.526 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88929\n",
      "INFO:tensorflow:train_accuracy = 0.53482586, train_loss = 15.9603 (0.529 sec)\n",
      "INFO:tensorflow:loss = 15.9603, step = 817 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84756\n",
      "INFO:tensorflow:train_accuracy = 0.5471464, train_loss = 14.413484 (0.541 sec)\n",
      "INFO:tensorflow:loss = 14.413484, step = 818 (0.541 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92608\n",
      "INFO:tensorflow:train_accuracy = 0.5499383, train_loss = 16.388279 (0.519 sec)\n",
      "INFO:tensorflow:loss = 16.388279, step = 819 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0881\n",
      "INFO:tensorflow:train_accuracy = 0.5740988, train_loss = 12.612274 (0.479 sec)\n",
      "INFO:tensorflow:loss = 12.612274, step = 820 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11241\n",
      "INFO:tensorflow:train_accuracy = 0.5551331, train_loss = 13.297284 (0.473 sec)\n",
      "INFO:tensorflow:loss = 13.297284, step = 821 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89314\n",
      "INFO:tensorflow:train_accuracy = 0.56832695, train_loss = 14.098337 (0.528 sec)\n",
      "INFO:tensorflow:loss = 14.098337, step = 822 (0.529 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0598\n",
      "INFO:tensorflow:train_accuracy = 0.56241786, train_loss = 13.221878 (0.485 sec)\n",
      "INFO:tensorflow:loss = 13.221878, step = 823 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04394\n",
      "INFO:tensorflow:train_accuracy = 0.555414, train_loss = 13.590097 (0.489 sec)\n",
      "INFO:tensorflow:loss = 13.590097, step = 824 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.32718\n",
      "INFO:tensorflow:train_accuracy = 0.58299595, train_loss = 10.364445 (0.430 sec)\n",
      "INFO:tensorflow:loss = 10.364445, step = 825 (0.430 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0074\n",
      "INFO:tensorflow:train_accuracy = 0.5538262, train_loss = 12.504958 (0.498 sec)\n",
      "INFO:tensorflow:loss = 12.504958, step = 826 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.17028\n",
      "INFO:tensorflow:train_accuracy = 0.5809524, train_loss = 13.688 (0.193 sec)\n",
      "INFO:tensorflow:loss = 13.688, step = 827 (0.193 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.78359\n",
      "INFO:tensorflow:train_accuracy = 0.5551331, train_loss = 13.730155 (0.561 sec)\n",
      "INFO:tensorflow:loss = 13.730155, step = 828 (0.560 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79921\n",
      "INFO:tensorflow:train_accuracy = 0.527881, train_loss = 17.820421 (0.556 sec)\n",
      "INFO:tensorflow:loss = 17.820421, step = 829 (0.556 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93352\n",
      "INFO:tensorflow:train_accuracy = 0.5697823, train_loss = 13.158595 (0.517 sec)\n",
      "INFO:tensorflow:loss = 13.158595, step = 830 (0.517 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8864\n",
      "INFO:tensorflow:train_accuracy = 0.572335, train_loss = 13.070873 (0.530 sec)\n",
      "INFO:tensorflow:loss = 13.070873, step = 831 (0.530 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95963\n",
      "INFO:tensorflow:train_accuracy = 0.556266, train_loss = 12.953985 (0.510 sec)\n",
      "INFO:tensorflow:loss = 12.953985, step = 832 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06464\n",
      "INFO:tensorflow:train_accuracy = 0.5629139, train_loss = 13.001358 (0.484 sec)\n",
      "INFO:tensorflow:loss = 13.001358, step = 833 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95058\n",
      "INFO:tensorflow:train_accuracy = 0.55441743, train_loss = 13.677249 (0.513 sec)\n",
      "INFO:tensorflow:loss = 13.677249, step = 834 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06367\n",
      "INFO:tensorflow:train_accuracy = 0.55878466, train_loss = 11.181491 (0.484 sec)\n",
      "INFO:tensorflow:loss = 11.181491, step = 835 (0.484 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91285\n",
      "INFO:tensorflow:train_accuracy = 0.57087123, train_loss = 14.620616 (0.523 sec)\n",
      "INFO:tensorflow:loss = 14.620616, step = 836 (0.523 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10231\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.5643045, train_loss = 12.0097 (0.476 sec)\n",
      "INFO:tensorflow:loss = 12.0097, step = 837 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04721\n",
      "INFO:tensorflow:train_accuracy = 0.5883152, train_loss = 11.255579 (0.488 sec)\n",
      "INFO:tensorflow:loss = 11.255579, step = 838 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00841\n",
      "INFO:tensorflow:train_accuracy = 0.5595855, train_loss = 14.105899 (0.498 sec)\n",
      "INFO:tensorflow:loss = 14.105899, step = 839 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98664\n",
      "INFO:tensorflow:train_accuracy = 0.57014155, train_loss = 13.172909 (0.503 sec)\n",
      "INFO:tensorflow:loss = 13.172909, step = 840 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96491\n",
      "INFO:tensorflow:train_accuracy = 0.5606258, train_loss = 13.402395 (0.509 sec)\n",
      "INFO:tensorflow:loss = 13.402395, step = 841 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16097\n",
      "INFO:tensorflow:train_accuracy = 0.575, train_loss = 11.21554 (0.463 sec)\n",
      "INFO:tensorflow:loss = 11.21554, step = 842 (0.463 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04058\n",
      "INFO:tensorflow:train_accuracy = 0.5628272, train_loss = 16.271976 (0.490 sec)\n",
      "INFO:tensorflow:loss = 16.271976, step = 843 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03187\n",
      "INFO:tensorflow:train_accuracy = 0.55289674, train_loss = 13.544498 (0.492 sec)\n",
      "INFO:tensorflow:loss = 13.544498, step = 844 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03266\n",
      "INFO:tensorflow:train_accuracy = 0.5833333, train_loss = 10.883051 (0.492 sec)\n",
      "INFO:tensorflow:loss = 10.883051, step = 845 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98512\n",
      "INFO:tensorflow:train_accuracy = 0.57733333, train_loss = 11.618181 (0.504 sec)\n",
      "INFO:tensorflow:loss = 11.618181, step = 846 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94442\n",
      "INFO:tensorflow:train_accuracy = 0.56453717, train_loss = 12.934517 (0.514 sec)\n",
      "INFO:tensorflow:loss = 12.934517, step = 847 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02597\n",
      "INFO:tensorflow:train_accuracy = 0.5631579, train_loss = 13.170454 (0.493 sec)\n",
      "INFO:tensorflow:loss = 13.170454, step = 848 (0.493 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90584\n",
      "INFO:tensorflow:train_accuracy = 0.5707071, train_loss = 14.470033 (0.525 sec)\n",
      "INFO:tensorflow:loss = 14.470033, step = 849 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98649\n",
      "INFO:tensorflow:train_accuracy = 0.5330836, train_loss = 13.392796 (0.503 sec)\n",
      "INFO:tensorflow:loss = 13.392796, step = 850 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03655\n",
      "INFO:tensorflow:train_accuracy = 0.57972974, train_loss = 12.853064 (0.491 sec)\n",
      "INFO:tensorflow:loss = 12.853064, step = 851 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03457\n",
      "INFO:tensorflow:train_accuracy = 0.58607095, train_loss = 12.060406 (0.492 sec)\n",
      "INFO:tensorflow:loss = 12.060406, step = 852 (0.492 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04281\n",
      "INFO:tensorflow:train_accuracy = 0.57444006, train_loss = 12.779227 (0.489 sec)\n",
      "INFO:tensorflow:loss = 12.779227, step = 853 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98005\n",
      "INFO:tensorflow:train_accuracy = 0.57894737, train_loss = 13.526987 (0.505 sec)\n",
      "INFO:tensorflow:loss = 13.526987, step = 854 (0.505 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07169\n",
      "INFO:tensorflow:train_accuracy = 0.5710594, train_loss = 14.354151 (0.483 sec)\n",
      "INFO:tensorflow:loss = 14.354151, step = 855 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03779\n",
      "INFO:tensorflow:train_accuracy = 0.5224242, train_loss = 13.508545 (0.491 sec)\n",
      "INFO:tensorflow:loss = 13.508545, step = 856 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01189\n",
      "INFO:tensorflow:train_accuracy = 0.5476493, train_loss = 13.205686 (0.497 sec)\n",
      "INFO:tensorflow:loss = 13.205686, step = 857 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18221\n",
      "INFO:tensorflow:train_accuracy = 0.5690537, train_loss = 14.356349 (0.458 sec)\n",
      "INFO:tensorflow:loss = 14.356349, step = 858 (0.458 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.73785\n",
      "INFO:tensorflow:train_accuracy = 0.5618557, train_loss = 14.62049 (0.575 sec)\n",
      "INFO:tensorflow:loss = 14.62049, step = 859 (0.575 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15576\n",
      "INFO:tensorflow:train_accuracy = 0.5490446, train_loss = 12.234015 (0.464 sec)\n",
      "INFO:tensorflow:loss = 12.234015, step = 860 (0.464 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99854\n",
      "INFO:tensorflow:train_accuracy = 0.54218364, train_loss = 14.123959 (0.501 sec)\n",
      "INFO:tensorflow:loss = 14.123959, step = 861 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87816\n",
      "INFO:tensorflow:train_accuracy = 0.5535714, train_loss = 13.356516 (0.533 sec)\n",
      "INFO:tensorflow:loss = 13.356516, step = 862 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93107\n",
      "INFO:tensorflow:train_accuracy = 0.5564202, train_loss = 13.835176 (0.518 sec)\n",
      "INFO:tensorflow:loss = 13.835176, step = 863 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93787\n",
      "INFO:tensorflow:train_accuracy = 0.55784065, train_loss = 13.308693 (0.516 sec)\n",
      "INFO:tensorflow:loss = 13.308693, step = 864 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05993\n",
      "INFO:tensorflow:train_accuracy = 0.55037785, train_loss = 13.653299 (0.485 sec)\n",
      "INFO:tensorflow:loss = 13.653299, step = 865 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92105\n",
      "INFO:tensorflow:train_accuracy = 0.5471698, train_loss = 12.49213 (0.520 sec)\n",
      "INFO:tensorflow:loss = 12.49213, step = 866 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90275\n",
      "INFO:tensorflow:train_accuracy = 0.5630573, train_loss = 13.681073 (0.526 sec)\n",
      "INFO:tensorflow:loss = 13.681073, step = 867 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00829\n",
      "INFO:tensorflow:train_accuracy = 0.5620915, train_loss = 13.169129 (0.498 sec)\n",
      "INFO:tensorflow:loss = 13.169129, step = 868 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17613\n",
      "INFO:tensorflow:train_accuracy = 0.5949198, train_loss = 11.102904 (0.460 sec)\n",
      "INFO:tensorflow:loss = 11.102904, step = 869 (0.459 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97314\n",
      "INFO:tensorflow:train_accuracy = 0.55943155, train_loss = 12.78363 (0.507 sec)\n",
      "INFO:tensorflow:loss = 12.78363, step = 870 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11007\n",
      "INFO:tensorflow:train_accuracy = 0.5676741, train_loss = 11.119198 (0.474 sec)\n",
      "INFO:tensorflow:loss = 11.119198, step = 871 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92025\n",
      "INFO:tensorflow:train_accuracy = 0.5694087, train_loss = 10.896395 (0.521 sec)\n",
      "INFO:tensorflow:loss = 10.896395, step = 872 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 5.23114\n",
      "INFO:tensorflow:train_accuracy = 0.58878505, train_loss = 9.391824 (0.191 sec)\n",
      "INFO:tensorflow:loss = 9.391824, step = 873 (0.191 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79562\n",
      "INFO:tensorflow:train_accuracy = 0.5628971, train_loss = 12.977507 (0.557 sec)\n",
      "INFO:tensorflow:loss = 12.977507, step = 874 (0.558 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.6607\n",
      "INFO:tensorflow:train_accuracy = 0.55239326, train_loss = 13.841026 (0.602 sec)\n",
      "INFO:tensorflow:loss = 13.841026, step = 875 (0.603 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87387\n",
      "INFO:tensorflow:train_accuracy = 0.5594493, train_loss = 13.087703 (0.533 sec)\n",
      "INFO:tensorflow:loss = 13.087703, step = 876 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96828\n",
      "INFO:tensorflow:train_accuracy = 0.56675065, train_loss = 12.612929 (0.508 sec)\n",
      "INFO:tensorflow:loss = 12.612929, step = 877 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85306\n",
      "INFO:tensorflow:train_accuracy = 0.5721716, train_loss = 11.831634 (0.540 sec)\n",
      "INFO:tensorflow:loss = 11.831634, step = 878 (0.540 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05336\n",
      "INFO:tensorflow:train_accuracy = 0.57235986, train_loss = 10.775 (0.487 sec)\n",
      "INFO:tensorflow:loss = 10.775, step = 879 (0.487 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.86173\n",
      "INFO:tensorflow:train_accuracy = 0.5779221, train_loss = 11.965365 (0.537 sec)\n",
      "INFO:tensorflow:loss = 11.965365, step = 880 (0.537 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.28302\n",
      "INFO:tensorflow:train_accuracy = 0.5673941, train_loss = 11.539291 (0.438 sec)\n",
      "INFO:tensorflow:loss = 11.539291, step = 881 (0.438 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.16757\n",
      "INFO:tensorflow:train_accuracy = 0.5847569, train_loss = 14.180214 (0.461 sec)\n",
      "INFO:tensorflow:loss = 14.180214, step = 882 (0.462 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0756\n",
      "INFO:tensorflow:train_accuracy = 0.5676741, train_loss = 10.816273 (0.482 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:loss = 10.816273, step = 883 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92051\n",
      "INFO:tensorflow:train_accuracy = 0.59495354, train_loss = 11.062202 (0.520 sec)\n",
      "INFO:tensorflow:loss = 11.062202, step = 884 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.05733\n",
      "INFO:tensorflow:train_accuracy = 0.56031126, train_loss = 11.085577 (0.486 sec)\n",
      "INFO:tensorflow:loss = 11.085577, step = 885 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.18522\n",
      "INFO:tensorflow:train_accuracy = 0.55927837, train_loss = 14.528601 (0.458 sec)\n",
      "INFO:tensorflow:loss = 14.528601, step = 886 (0.457 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.0263\n",
      "INFO:tensorflow:train_accuracy = 0.57588077, train_loss = 10.444466 (0.494 sec)\n",
      "INFO:tensorflow:loss = 10.444466, step = 887 (0.494 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11165\n",
      "INFO:tensorflow:train_accuracy = 0.5770235, train_loss = 11.039472 (0.473 sec)\n",
      "INFO:tensorflow:loss = 11.039472, step = 888 (0.472 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8229\n",
      "INFO:tensorflow:train_accuracy = 0.5716112, train_loss = 14.067532 (0.549 sec)\n",
      "INFO:tensorflow:loss = 14.067532, step = 889 (0.550 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00541\n",
      "INFO:tensorflow:train_accuracy = 0.5782665, train_loss = 12.032055 (0.498 sec)\n",
      "INFO:tensorflow:loss = 12.032055, step = 890 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1756\n",
      "INFO:tensorflow:train_accuracy = 0.563543, train_loss = 10.72806 (0.460 sec)\n",
      "INFO:tensorflow:loss = 10.72806, step = 891 (0.461 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97259\n",
      "INFO:tensorflow:train_accuracy = 0.58629775, train_loss = 10.526976 (0.507 sec)\n",
      "INFO:tensorflow:loss = 10.526976, step = 892 (0.507 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98544\n",
      "INFO:tensorflow:train_accuracy = 0.5710526, train_loss = 12.15822 (0.504 sec)\n",
      "INFO:tensorflow:loss = 12.15822, step = 893 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94451\n",
      "INFO:tensorflow:train_accuracy = 0.58278143, train_loss = 11.402758 (0.514 sec)\n",
      "INFO:tensorflow:loss = 11.402758, step = 894 (0.514 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06262\n",
      "INFO:tensorflow:train_accuracy = 0.5417721, train_loss = 11.58036 (0.485 sec)\n",
      "INFO:tensorflow:loss = 11.58036, step = 895 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.21269\n",
      "INFO:tensorflow:train_accuracy = 0.5663717, train_loss = 12.625377 (0.452 sec)\n",
      "INFO:tensorflow:loss = 12.625377, step = 896 (0.451 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85202\n",
      "INFO:tensorflow:train_accuracy = 0.5895722, train_loss = 9.96394 (0.540 sec)\n",
      "INFO:tensorflow:loss = 9.96394, step = 897 (0.541 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.29167\n",
      "INFO:tensorflow:train_accuracy = 0.56992084, train_loss = 10.583439 (0.436 sec)\n",
      "INFO:tensorflow:loss = 10.583439, step = 898 (0.436 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96275\n",
      "INFO:tensorflow:train_accuracy = 0.5808147, train_loss = 12.781679 (0.509 sec)\n",
      "INFO:tensorflow:loss = 12.781679, step = 899 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.02076\n",
      "INFO:tensorflow:train_accuracy = 0.56453717, train_loss = 11.981394 (0.495 sec)\n",
      "INFO:tensorflow:loss = 11.981394, step = 900 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96478\n",
      "INFO:tensorflow:train_accuracy = 0.5491184, train_loss = 13.41145 (0.509 sec)\n",
      "INFO:tensorflow:loss = 13.41145, step = 901 (0.509 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96536\n",
      "INFO:tensorflow:train_accuracy = 0.5818414, train_loss = 11.086457 (0.509 sec)\n",
      "INFO:tensorflow:loss = 11.086457, step = 902 (0.508 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90803\n",
      "INFO:tensorflow:train_accuracy = 0.54879594, train_loss = 11.79617 (0.524 sec)\n",
      "INFO:tensorflow:loss = 11.79617, step = 903 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06996\n",
      "INFO:tensorflow:train_accuracy = 0.56766915, train_loss = 11.430229 (0.483 sec)\n",
      "INFO:tensorflow:loss = 11.430229, step = 904 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.82153\n",
      "INFO:tensorflow:train_accuracy = 0.56081945, train_loss = 13.222699 (0.549 sec)\n",
      "INFO:tensorflow:loss = 13.222699, step = 905 (0.550 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.17689\n",
      "INFO:tensorflow:train_accuracy = 0.5667539, train_loss = 10.919121 (0.459 sec)\n",
      "INFO:tensorflow:loss = 10.919121, step = 906 (0.459 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00135\n",
      "INFO:tensorflow:train_accuracy = 0.5275, train_loss = 11.703068 (0.500 sec)\n",
      "INFO:tensorflow:loss = 11.703068, step = 907 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99988\n",
      "INFO:tensorflow:train_accuracy = 0.5777778, train_loss = 9.52713 (0.500 sec)\n",
      "INFO:tensorflow:loss = 9.52713, step = 908 (0.500 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79118\n",
      "INFO:tensorflow:train_accuracy = 0.5504926, train_loss = 14.557886 (0.558 sec)\n",
      "INFO:tensorflow:loss = 14.557886, step = 909 (0.558 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10838\n",
      "INFO:tensorflow:train_accuracy = 0.562341, train_loss = 11.871998 (0.475 sec)\n",
      "INFO:tensorflow:loss = 11.871998, step = 910 (0.475 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98601\n",
      "INFO:tensorflow:train_accuracy = 0.58616185, train_loss = 11.386173 (0.503 sec)\n",
      "INFO:tensorflow:loss = 11.386173, step = 911 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08627\n",
      "INFO:tensorflow:train_accuracy = 0.5603015, train_loss = 11.300086 (0.479 sec)\n",
      "INFO:tensorflow:loss = 11.300086, step = 912 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96384\n",
      "INFO:tensorflow:train_accuracy = 0.553137, train_loss = 11.329624 (0.509 sec)\n",
      "INFO:tensorflow:loss = 11.329624, step = 913 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99138\n",
      "INFO:tensorflow:train_accuracy = 0.556266, train_loss = 11.011454 (0.502 sec)\n",
      "INFO:tensorflow:loss = 11.011454, step = 914 (0.502 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90436\n",
      "INFO:tensorflow:train_accuracy = 0.5637066, train_loss = 10.968296 (0.525 sec)\n",
      "INFO:tensorflow:loss = 10.968296, step = 915 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92341\n",
      "INFO:tensorflow:train_accuracy = 0.5706941, train_loss = 10.946623 (0.520 sec)\n",
      "INFO:tensorflow:loss = 10.946623, step = 916 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12938\n",
      "INFO:tensorflow:train_accuracy = 0.5899471, train_loss = 10.883312 (0.470 sec)\n",
      "INFO:tensorflow:loss = 10.883312, step = 917 (0.470 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.10105\n",
      "INFO:tensorflow:train_accuracy = 0.5818908, train_loss = 8.010018 (0.476 sec)\n",
      "INFO:tensorflow:loss = 8.010018, step = 918 (0.476 sec)\n",
      "INFO:tensorflow:global_step/sec: 7.38858\n",
      "INFO:tensorflow:train_accuracy = 0.6084906, train_loss = 9.753752 (0.135 sec)\n",
      "INFO:tensorflow:loss = 9.753752, step = 919 (0.134 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92337\n",
      "INFO:tensorflow:train_accuracy = 0.5667539, train_loss = 11.461674 (0.520 sec)\n",
      "INFO:tensorflow:loss = 11.461674, step = 920 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84832\n",
      "INFO:tensorflow:train_accuracy = 0.56925035, train_loss = 12.340765 (0.541 sec)\n",
      "INFO:tensorflow:loss = 12.340765, step = 921 (0.542 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01041\n",
      "INFO:tensorflow:train_accuracy = 0.54467565, train_loss = 11.976568 (0.497 sec)\n",
      "INFO:tensorflow:loss = 11.976568, step = 922 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93327\n",
      "INFO:tensorflow:train_accuracy = 0.5840708, train_loss = 11.204119 (0.517 sec)\n",
      "INFO:tensorflow:loss = 11.204119, step = 923 (0.517 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9323\n",
      "INFO:tensorflow:train_accuracy = 0.5970938, train_loss = 9.52335 (0.518 sec)\n",
      "INFO:tensorflow:loss = 9.52335, step = 924 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03667\n",
      "INFO:tensorflow:train_accuracy = 0.5522959, train_loss = 11.01152 (0.491 sec)\n",
      "INFO:tensorflow:loss = 11.01152, step = 925 (0.491 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98741\n",
      "INFO:tensorflow:train_accuracy = 0.59292036, train_loss = 11.911322 (0.503 sec)\n",
      "INFO:tensorflow:loss = 11.911322, step = 926 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.12194\n",
      "INFO:tensorflow:train_accuracy = 0.5689655, train_loss = 9.588944 (0.471 sec)\n",
      "INFO:tensorflow:loss = 9.588944, step = 927 (0.471 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92626\n",
      "INFO:tensorflow:train_accuracy = 0.5867876, train_loss = 11.467916 (0.519 sec)\n",
      "INFO:tensorflow:loss = 11.467916, step = 928 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.79464\n",
      "INFO:tensorflow:train_accuracy = 0.5836627, train_loss = 9.54921 (0.558 sec)\n",
      "INFO:tensorflow:loss = 9.54921, step = 929 (0.558 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00909\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:train_accuracy = 0.58539766, train_loss = 10.408207 (0.498 sec)\n",
      "INFO:tensorflow:loss = 10.408207, step = 930 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9597\n",
      "INFO:tensorflow:train_accuracy = 0.5983828, train_loss = 9.776345 (0.510 sec)\n",
      "INFO:tensorflow:loss = 9.776345, step = 931 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01038\n",
      "INFO:tensorflow:train_accuracy = 0.5793651, train_loss = 11.331739 (0.497 sec)\n",
      "INFO:tensorflow:loss = 11.331739, step = 932 (0.497 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89622\n",
      "INFO:tensorflow:train_accuracy = 0.58344287, train_loss = 10.3788 (0.527 sec)\n",
      "INFO:tensorflow:loss = 10.3788, step = 933 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15092\n",
      "INFO:tensorflow:train_accuracy = 0.5704698, train_loss = 9.8031 (0.465 sec)\n",
      "INFO:tensorflow:loss = 9.8031, step = 934 (0.466 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00703\n",
      "INFO:tensorflow:train_accuracy = 0.57215834, train_loss = 10.126661 (0.498 sec)\n",
      "INFO:tensorflow:loss = 10.126661, step = 935 (0.498 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06019\n",
      "INFO:tensorflow:train_accuracy = 0.57583547, train_loss = 8.660117 (0.485 sec)\n",
      "INFO:tensorflow:loss = 8.660117, step = 936 (0.486 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91864\n",
      "INFO:tensorflow:train_accuracy = 0.5743329, train_loss = 11.79207 (0.521 sec)\n",
      "INFO:tensorflow:loss = 11.79207, step = 937 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.8847\n",
      "INFO:tensorflow:train_accuracy = 0.6008119, train_loss = 8.49487 (0.531 sec)\n",
      "INFO:tensorflow:loss = 8.49487, step = 938 (0.531 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.877\n",
      "INFO:tensorflow:train_accuracy = 0.5806029, train_loss = 12.266326 (0.533 sec)\n",
      "INFO:tensorflow:loss = 12.266326, step = 939 (0.533 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93988\n",
      "INFO:tensorflow:train_accuracy = 0.56901044, train_loss = 10.398272 (0.515 sec)\n",
      "INFO:tensorflow:loss = 10.398272, step = 940 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9171\n",
      "INFO:tensorflow:train_accuracy = 0.56171286, train_loss = 11.550349 (0.522 sec)\n",
      "INFO:tensorflow:loss = 11.550349, step = 941 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93981\n",
      "INFO:tensorflow:train_accuracy = 0.55889726, train_loss = 10.573191 (0.516 sec)\n",
      "INFO:tensorflow:loss = 10.573191, step = 942 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07122\n",
      "INFO:tensorflow:train_accuracy = 0.57631576, train_loss = 8.983968 (0.483 sec)\n",
      "INFO:tensorflow:loss = 8.983968, step = 943 (0.483 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94204\n",
      "INFO:tensorflow:train_accuracy = 0.6063408, train_loss = 10.690218 (0.515 sec)\n",
      "INFO:tensorflow:loss = 10.690218, step = 944 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93542\n",
      "INFO:tensorflow:train_accuracy = 0.5803453, train_loss = 11.029001 (0.517 sec)\n",
      "INFO:tensorflow:loss = 11.029001, step = 945 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.23853\n",
      "INFO:tensorflow:train_accuracy = 0.58707124, train_loss = 8.896814 (0.447 sec)\n",
      "INFO:tensorflow:loss = 8.896814, step = 946 (0.447 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.97467\n",
      "INFO:tensorflow:train_accuracy = 0.55025125, train_loss = 11.336752 (0.506 sec)\n",
      "INFO:tensorflow:loss = 11.336752, step = 947 (0.506 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94928\n",
      "INFO:tensorflow:train_accuracy = 0.56615776, train_loss = 11.101131 (0.513 sec)\n",
      "INFO:tensorflow:loss = 11.101131, step = 948 (0.513 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11646\n",
      "INFO:tensorflow:train_accuracy = 0.5710594, train_loss = 9.733561 (0.473 sec)\n",
      "INFO:tensorflow:loss = 9.733561, step = 949 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.84126\n",
      "INFO:tensorflow:train_accuracy = 0.578481, train_loss = 10.97315 (0.542 sec)\n",
      "INFO:tensorflow:loss = 10.97315, step = 950 (0.542 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89657\n",
      "INFO:tensorflow:train_accuracy = 0.5670498, train_loss = 10.30416 (0.527 sec)\n",
      "INFO:tensorflow:loss = 10.30416, step = 951 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92381\n",
      "INFO:tensorflow:train_accuracy = 0.53836477, train_loss = 10.325506 (0.520 sec)\n",
      "INFO:tensorflow:loss = 10.325506, step = 952 (0.520 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91717\n",
      "INFO:tensorflow:train_accuracy = 0.55938697, train_loss = 9.288715 (0.522 sec)\n",
      "INFO:tensorflow:loss = 9.288715, step = 953 (0.521 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94327\n",
      "INFO:tensorflow:train_accuracy = 0.5596919, train_loss = 10.080687 (0.515 sec)\n",
      "INFO:tensorflow:loss = 10.080687, step = 954 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.81932\n",
      "INFO:tensorflow:train_accuracy = 0.56265664, train_loss = 11.224843 (0.550 sec)\n",
      "INFO:tensorflow:loss = 11.224843, step = 955 (0.550 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.90426\n",
      "INFO:tensorflow:train_accuracy = 0.5466999, train_loss = 10.3445015 (0.525 sec)\n",
      "INFO:tensorflow:loss = 10.3445015, step = 956 (0.525 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.99652\n",
      "INFO:tensorflow:train_accuracy = 0.5647668, train_loss = 10.763624 (0.501 sec)\n",
      "INFO:tensorflow:loss = 10.763624, step = 957 (0.501 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.1952\n",
      "INFO:tensorflow:train_accuracy = 0.562982, train_loss = 9.71637 (0.455 sec)\n",
      "INFO:tensorflow:loss = 9.71637, step = 958 (0.455 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94225\n",
      "INFO:tensorflow:train_accuracy = 0.5992167, train_loss = 9.538662 (0.515 sec)\n",
      "INFO:tensorflow:loss = 9.538662, step = 959 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11336\n",
      "INFO:tensorflow:train_accuracy = 0.58709675, train_loss = 12.501841 (0.473 sec)\n",
      "INFO:tensorflow:loss = 12.501841, step = 960 (0.473 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.9282\n",
      "INFO:tensorflow:train_accuracy = 0.55099505, train_loss = 9.980043 (0.519 sec)\n",
      "INFO:tensorflow:loss = 9.980043, step = 961 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92849\n",
      "INFO:tensorflow:train_accuracy = 0.58237547, train_loss = 8.412634 (0.518 sec)\n",
      "INFO:tensorflow:loss = 8.412634, step = 962 (0.518 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.95996\n",
      "INFO:tensorflow:train_accuracy = 0.59042555, train_loss = 9.040008 (0.510 sec)\n",
      "INFO:tensorflow:loss = 9.040008, step = 963 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00809\n",
      "INFO:tensorflow:train_accuracy = 0.6142473, train_loss = 7.7407875 (0.498 sec)\n",
      "INFO:tensorflow:loss = 7.7407875, step = 964 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 6.20172\n",
      "INFO:tensorflow:train_accuracy = 0.60189575, train_loss = 7.679294 (0.161 sec)\n",
      "INFO:tensorflow:loss = 7.679294, step = 965 (0.161 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.73754\n",
      "INFO:tensorflow:train_accuracy = 0.58505154, train_loss = 9.108176 (0.576 sec)\n",
      "INFO:tensorflow:loss = 9.108176, step = 966 (0.576 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.78107\n",
      "INFO:tensorflow:train_accuracy = 0.5758323, train_loss = 11.14966 (0.561 sec)\n",
      "INFO:tensorflow:loss = 11.14966, step = 967 (0.561 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93825\n",
      "INFO:tensorflow:train_accuracy = 0.58980894, train_loss = 9.904249 (0.516 sec)\n",
      "INFO:tensorflow:loss = 9.904249, step = 968 (0.516 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.94142\n",
      "INFO:tensorflow:train_accuracy = 0.60358053, train_loss = 9.179468 (0.515 sec)\n",
      "INFO:tensorflow:loss = 9.179468, step = 969 (0.515 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.03723\n",
      "INFO:tensorflow:train_accuracy = 0.6075269, train_loss = 8.2074585 (0.491 sec)\n",
      "INFO:tensorflow:loss = 8.2074585, step = 970 (0.490 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.96016\n",
      "INFO:tensorflow:train_accuracy = 0.5672589, train_loss = 8.317732 (0.510 sec)\n",
      "INFO:tensorflow:loss = 8.317732, step = 971 (0.510 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.85703\n",
      "INFO:tensorflow:train_accuracy = 0.5636822, train_loss = 12.576158 (0.539 sec)\n",
      "INFO:tensorflow:loss = 12.576158, step = 972 (0.539 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89717\n",
      "INFO:tensorflow:train_accuracy = 0.6031957, train_loss = 8.939763 (0.527 sec)\n",
      "INFO:tensorflow:loss = 8.939763, step = 973 (0.527 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.91867\n",
      "INFO:tensorflow:train_accuracy = 0.5692308, train_loss = 13.04804 (0.521 sec)\n",
      "INFO:tensorflow:loss = 13.04804, step = 974 (0.522 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.11166\n",
      "INFO:tensorflow:train_accuracy = 0.57366365, train_loss = 8.750465 (0.474 sec)\n",
      "INFO:tensorflow:loss = 8.750465, step = 975 (0.474 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.09872\n",
      "INFO:tensorflow:train_accuracy = 0.59466666, train_loss = 8.13993 (0.477 sec)\n",
      "INFO:tensorflow:loss = 8.13993, step = 976 (0.477 sec)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:global_step/sec: 2.17254\n",
      "INFO:tensorflow:train_accuracy = 0.5868421, train_loss = 8.395511 (0.460 sec)\n",
      "INFO:tensorflow:loss = 8.395511, step = 977 (0.460 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08634\n",
      "INFO:tensorflow:train_accuracy = 0.5719745, train_loss = 10.685549 (0.479 sec)\n",
      "INFO:tensorflow:loss = 10.685549, step = 978 (0.479 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04638\n",
      "INFO:tensorflow:train_accuracy = 0.5970549, train_loss = 9.8347 (0.489 sec)\n",
      "INFO:tensorflow:loss = 9.8347, step = 979 (0.489 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98731\n",
      "INFO:tensorflow:train_accuracy = 0.5978836, train_loss = 9.349877 (0.503 sec)\n",
      "INFO:tensorflow:loss = 9.349877, step = 980 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.88134\n",
      "INFO:tensorflow:train_accuracy = 0.5839793, train_loss = 9.844684 (0.532 sec)\n",
      "INFO:tensorflow:loss = 9.844684, step = 981 (0.532 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00263\n",
      "INFO:tensorflow:train_accuracy = 0.5710526, train_loss = 7.979488 (0.499 sec)\n",
      "INFO:tensorflow:loss = 7.979488, step = 982 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.04859\n",
      "INFO:tensorflow:train_accuracy = 0.58928573, train_loss = 9.297625 (0.488 sec)\n",
      "INFO:tensorflow:loss = 9.297625, step = 983 (0.488 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01838\n",
      "INFO:tensorflow:train_accuracy = 0.59096944, train_loss = 8.863863 (0.496 sec)\n",
      "INFO:tensorflow:loss = 8.863863, step = 984 (0.496 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.87002\n",
      "INFO:tensorflow:train_accuracy = 0.58629775, train_loss = 10.273321 (0.535 sec)\n",
      "INFO:tensorflow:loss = 10.273321, step = 985 (0.535 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.01841\n",
      "INFO:tensorflow:train_accuracy = 0.5816994, train_loss = 9.157484 (0.495 sec)\n",
      "INFO:tensorflow:loss = 9.157484, step = 986 (0.495 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.93537\n",
      "INFO:tensorflow:train_accuracy = 0.5726928, train_loss = 8.284483 (0.517 sec)\n",
      "INFO:tensorflow:loss = 8.284483, step = 987 (0.517 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.06354\n",
      "INFO:tensorflow:train_accuracy = 0.5837629, train_loss = 10.283245 (0.485 sec)\n",
      "INFO:tensorflow:loss = 10.283245, step = 988 (0.485 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.92693\n",
      "INFO:tensorflow:train_accuracy = 0.57198954, train_loss = 8.445904 (0.519 sec)\n",
      "INFO:tensorflow:loss = 8.445904, step = 989 (0.519 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98386\n",
      "INFO:tensorflow:train_accuracy = 0.58933336, train_loss = 8.676499 (0.504 sec)\n",
      "INFO:tensorflow:loss = 8.676499, step = 990 (0.504 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.98809\n",
      "INFO:tensorflow:train_accuracy = 0.5857143, train_loss = 9.29236 (0.503 sec)\n",
      "INFO:tensorflow:loss = 9.29236, step = 991 (0.503 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.07568\n",
      "INFO:tensorflow:train_accuracy = 0.5847569, train_loss = 8.966092 (0.482 sec)\n",
      "INFO:tensorflow:loss = 8.966092, step = 992 (0.482 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.82393\n",
      "INFO:tensorflow:train_accuracy = 0.57706535, train_loss = 11.038257 (0.548 sec)\n",
      "INFO:tensorflow:loss = 11.038257, step = 993 (0.548 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.15445\n",
      "INFO:tensorflow:train_accuracy = 0.55025774, train_loss = 9.369637 (0.464 sec)\n",
      "INFO:tensorflow:loss = 9.369637, step = 994 (0.465 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.08474\n",
      "INFO:tensorflow:train_accuracy = 0.58492976, train_loss = 10.73138 (0.480 sec)\n",
      "INFO:tensorflow:loss = 10.73138, step = 995 (0.480 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.24752\n",
      "INFO:tensorflow:train_accuracy = 0.5703422, train_loss = 9.216164 (0.445 sec)\n",
      "INFO:tensorflow:loss = 9.216164, step = 996 (0.445 sec)\n",
      "INFO:tensorflow:global_step/sec: 2.00395\n",
      "INFO:tensorflow:train_accuracy = 0.57942706, train_loss = 8.553196 (0.499 sec)\n",
      "INFO:tensorflow:loss = 8.553196, step = 997 (0.499 sec)\n",
      "INFO:tensorflow:global_step/sec: 1.89423\n",
      "INFO:tensorflow:train_accuracy = 0.57253885, train_loss = 9.20272 (0.528 sec)\n",
      "INFO:tensorflow:loss = 9.20272, step = 998 (0.528 sec)\n",
      "INFO:tensorflow:Saving checkpoints for 1000 into asr-minijasper/model.ckpt.\n",
      "WARNING:tensorflow:From /home/husein/.local/lib/python3.6/site-packages/tensorflow_core/python/training/saver.py:963: remove_checkpoint (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use standard file APIs to delete files with this prefix.\n",
      "INFO:tensorflow:Calling model_fn.\n",
      "INFO:tensorflow:Done calling model_fn.\n",
      "INFO:tensorflow:Starting evaluation at 2020-10-25T19:16:58Z\n",
      "INFO:tensorflow:Graph was finalized.\n",
      "INFO:tensorflow:Restoring parameters from asr-minijasper/model.ckpt-1000\n",
      "INFO:tensorflow:Running local_init_op.\n",
      "INFO:tensorflow:Done running local_init_op.\n",
      "INFO:tensorflow:Evaluation [1/10]\n",
      "INFO:tensorflow:Evaluation [2/10]\n",
      "INFO:tensorflow:Evaluation [3/10]\n",
      "INFO:tensorflow:Evaluation [4/10]\n",
      "INFO:tensorflow:Evaluation [5/10]\n",
      "INFO:tensorflow:Evaluation [6/10]\n",
      "INFO:tensorflow:Evaluation [7/10]\n",
      "INFO:tensorflow:Evaluation [8/10]\n",
      "INFO:tensorflow:Evaluation [9/10]\n",
      "INFO:tensorflow:Evaluation [10/10]\n",
      "INFO:tensorflow:Finished evaluation at 2020-10-25-19:17:03\n",
      "INFO:tensorflow:Saving dict for global step 1000: WER = 1.0, accuracy = 0.57361966, global_step = 1000, loss = 23.33398\n",
      "INFO:tensorflow:Saving 'checkpoint_path' summary for global step 1000: asr-minijasper/model.ckpt-1000\n",
      "INFO:tensorflow:global_step/sec: 0.0973404\n",
      "INFO:tensorflow:train_accuracy = 0.5541872, train_loss = 9.266422 (10.273 sec)\n",
      "INFO:tensorflow:loss = 9.266422, step = 999 (10.273 sec)\n",
      "INFO:tensorflow:Loss for final step: 9.266422.\n"
     ]
    }
   ],
   "source": [
    "train.run_training(\n",
    "    train_fn = train_dataset,\n",
    "    model_fn = model_fn,\n",
    "    model_dir = 'asr-minijasper',\n",
    "    num_gpus = 1,\n",
    "    log_step = 1,\n",
    "    save_checkpoint_step = 200,\n",
    "    max_steps = train_steps,\n",
    "    eval_fn = dev_dataset,\n",
    "    train_hooks = train_hooks,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reference\n",
    "\n",
    "1. Nvidia OpenSeq2Seq, https://github.com/NVIDIA/OpenSeq2Seq/blob/master/example_configs/speech2text/jasper-Mini-for-Jetson.py"
   ]
  },
  {
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}