tf_mnist_example.py

''' Demonstrate the usage of the telegram DL bot on the official Tensorflow MNIST example.
(https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/tutorials/mnist/fully_connected_feed.py)
All rights to the backbone of this example reserved to Tensorflow.
Trains and Evaluates the MNIST network using a feed dictionary
By: Eyal Zakkay, 2019
https://eyalzk.github.io/
'''

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

# pylint: disable=missing-docstring
import argparse
import os
import sys
import time

from six.moves import xrange  # pylint: disable=redefined-builtin
import tensorflow as tf

from tensorflow.examples.tutorials.mnist import input_data
from tensorflow.examples.tutorials.mnist import mnist

# Telegram Bot imports
from dl_bot import DLBot

telegram_token = "TOKEN"  # replace TOKEN with your bot's token

# user id is optional, however highly recommended as it limits the access to you alone.
telegram_user_id = None   # replace None with your telegram user id (integer):

# Create a DLBot instance
bot = DLBot(token=telegram_token, user_id=telegram_user_id)
# Activate the bot
bot.activate_bot()
# Basic model parameters as external flags.
FLAGS = None


# Unmodified function
def placeholder_inputs(batch_size):
  """Generate placeholder variables to represent the input tensors.
  These placeholders are used as inputs by the rest of the model building
  code and will be fed from the downloaded data in the .run() loop, below.
  Args:
    batch_size: The batch size will be baked into both placeholders.
  Returns:
    images_placeholder: Images placeholder.
    labels_placeholder: Labels placeholder.
  """
  # Note that the shapes of the placeholders match the shapes of the full
  # image and label tensors, except the first dimension is now batch_size
  # rather than the full size of the train or test data sets.
  images_placeholder = tf.placeholder(tf.float32, shape=(batch_size,
                                                         mnist.IMAGE_PIXELS))
  labels_placeholder = tf.placeholder(tf.int32, shape=(batch_size))
  return images_placeholder, labels_placeholder

# Unmodified function
def fill_feed_dict(data_set, images_pl, labels_pl):
  """Fills the feed_dict for training the given step.
  A feed_dict takes the form of:
  feed_dict = {
      <placeholder>: <tensor of values to be passed for placeholder>,
      ....
  }
  Args:
    data_set: The set of images and labels, from input_data.read_data_sets()
    images_pl: The images placeholder, from placeholder_inputs().
    labels_pl: The labels placeholder, from placeholder_inputs().
  Returns:
    feed_dict: The feed dictionary mapping from placeholders to values.
  """
  # Create the feed_dict for the placeholders filled with the next
  # `batch size` examples.
  images_feed, labels_feed = data_set.next_batch(FLAGS.batch_size,
                                                 FLAGS.fake_data)
  feed_dict = {
      images_pl: images_feed,
      labels_pl: labels_feed,
  }
  return feed_dict


def do_eval(sess,
            eval_correct,
            images_placeholder,
            labels_placeholder,
            data_set):
  """Runs one evaluation against the full epoch of data.
  Args:
    sess: The session in which the model has been trained.
    eval_correct: The Tensor that returns the number of correct predictions.
    images_placeholder: The images placeholder.
    labels_placeholder: The labels placeholder.
    data_set: The set of images and labels to evaluate, from
      input_data.read_data_sets().

  Returns:
      message: string containing the printed message
  """
  # And run one epoch of eval.
  true_count = 0  # Counts the number of correct predictions.
  steps_per_epoch = data_set.num_examples // FLAGS.batch_size
  num_examples = steps_per_epoch * FLAGS.batch_size
  for step in xrange(steps_per_epoch):
    feed_dict = fill_feed_dict(data_set,
                               images_placeholder,
                               labels_placeholder)
    true_count += sess.run(eval_correct, feed_dict=feed_dict)
  precision = float(true_count) / num_examples
  message = 'Num examples: %d  Num correct: %d  Precision @ 1: %0.04f' %(num_examples, true_count, precision)
  print(message)
  # BOT: returns the printed message so we can feed it to bot
  return message


def run_training():
  """Train MNIST for a number of steps."""
  # Get the sets of images and labels for training, validation, and
  # test on MNIST.
  data_sets = input_data.read_data_sets(FLAGS.input_data_dir, FLAGS.fake_data)

  # Tell TensorFlow that the model will be built into the default Graph.
  with tf.Graph().as_default():
    # Generate placeholders for the images and labels.
    images_placeholder, labels_placeholder = placeholder_inputs(
        FLAGS.batch_size)

    # Build a Graph that computes predictions from the inference model.
    logits = mnist.inference(images_placeholder,
                             FLAGS.hidden1,
                             FLAGS.hidden2)

    # Add to the Graph the Ops for loss calculation.
    loss = mnist.loss(logits, labels_placeholder)

    # Add to the Graph the Ops that calculate and apply gradients.
    # BOT: making the lr a variable so we can update it using our bot
    learning_rate = tf.Variable(FLAGS.learning_rate, trainable=False)
    train_op = mnist.training(loss, learning_rate)
    bot.lr = FLAGS.learning_rate
    # Add the Op to compare the logits to the labels during evaluation.
    eval_correct = mnist.evaluation(logits, labels_placeholder)

    # Build the summary Tensor based on the TF collection of Summaries.
    summary = tf.summary.merge_all()

    # Add the variable initializer Op.
    init = tf.global_variables_initializer()

    # Create a saver for writing training checkpoints.
    saver = tf.train.Saver()

    # Create a session for running Ops on the Graph.
    sess = tf.Session()

    # Instantiate a SummaryWriter to output summaries and the Graph.
    summary_writer = tf.summary.FileWriter(FLAGS.log_dir, sess.graph)

    # And then after everything is built:

    # Run the Op to initialize the variables.
    sess.run(init)

    # Start the training loop.
    for step in xrange(FLAGS.max_steps):
      start_time = time.time()

      # Fill a feed dictionary with the actual set of images and labels
      # for this particular training step.
      feed_dict = fill_feed_dict(data_sets.train,
                                 images_placeholder,
                                 labels_placeholder)

      # Run one step of the model.  The return values are the activations
      # from the `train_op` (which is discarded) and the `loss` Op.  To
      # inspect the values of your Ops or variables, you may include them
      # in the list passed to sess.run() and the value tensors will be
      # returned in the tuple from the call.
      _, loss_value = sess.run([train_op, loss],
                               feed_dict=feed_dict)

      duration = time.time() - start_time

      # Write the summaries and print an overview fairly often.
      if step % 100 == 0:
        # Print status to stdout.
        print('Step %d: loss = %.2f (%.3f sec)' % (step, loss_value, duration))
        # Update the events file.
        summary_str = sess.run(summary, feed_dict=feed_dict)
        summary_writer.add_summary(summary_str, step)
        summary_writer.flush()

      # Save a checkpoint and evaluate the model periodically.
      if (step + 1) % 1000 == 0 or (step + 1) == FLAGS.max_steps:
        checkpoint_file = os.path.join(FLAGS.log_dir, 'model.ckpt')
        saver.save(sess, checkpoint_file, global_step=step)
        # Print step number:
        print("step: {}".format(step))

        # Evaluate against the training set.
        print('Training Data Eval:')
        message_trn = 'Training Data Eval: \n' + do_eval(sess,
                                                         eval_correct,
                                                         images_placeholder,
                                                         labels_placeholder,
                                                         data_sets.train)

        # Evaluate against the validation set.
        print('Validation Data Eval:')
        message_val = 'Validation Data Eval:\n' + do_eval(sess,
                                                          eval_correct,
                                                          images_placeholder,
                                                          labels_placeholder,
                                                          data_sets.validation)
        # Evaluate validation loss
        val_loss_value = sess.run(loss, feed_dict=fill_feed_dict(data_sets.validation,
                                                           images_placeholder,
                                                           labels_placeholder))

        # Evaluate against the test set.
        print('Test Data Eval:')
        message_tst = 'Test Data Eval:\n' + do_eval(sess,
                                                    eval_correct,
                                                    images_placeholder,
                                                    labels_placeholder,
                                                    data_sets.test)


        ## BOT: handling of all bot commands ##
        # Prepare bot update message
        message = "\n".join(["step: {}".format(step+1), message_trn, message_val, message_tst])
        bot.set_status(message)
        # Send update message
        if bot.verbose:
          bot.send_message(message)

        # Stop training command from bot
        if bot.stop_train_flag:
          bot.send_message('Training stopped!')
          print('Training Stopped! Stop command sent via Telegram bot.')
          break

        # Update bot's loss history (for /plot command)
        bot.loss_hist.append(loss_value)
        bot.val_loss_hist.append(val_loss_value)

        # Modify learning rate via bot
        if bot.modify_lr != 1:
          curr_lr = sess.run(learning_rate)
          new_lr = curr_lr*bot.modify_lr
          learning_rate = tf.assign(learning_rate, new_lr)
          message = '\nStep %05d: setting learning rate to %f.' % (step + 1, new_lr)
          print(message)
          bot.send_message(message)
          bot.modify_lr = 1
          bot.lr = new_lr


def main(_):
  if tf.gfile.Exists(FLAGS.log_dir):
    tf.gfile.DeleteRecursively(FLAGS.log_dir)
  tf.gfile.MakeDirs(FLAGS.log_dir)
  run_training()


if __name__ == '__main__':
  parser = argparse.ArgumentParser()
  parser.add_argument(
      '--learning_rate',
      type=float,
      default=0.01,
      help='Initial learning rate.'
  )
  parser.add_argument(
      '--max_steps',
      type=int,
      default=50000,
      help='Number of steps to run trainer.'
  )
  parser.add_argument(
      '--hidden1',
      type=int,
      default=128,
      help='Number of units in hidden layer 1.'
  )
  parser.add_argument(
      '--hidden2',
      type=int,
      default=32,
      help='Number of units in hidden layer 2.'
  )
  parser.add_argument(
      '--batch_size',
      type=int,
      default=100,
      help='Batch size.  Must divide evenly into the dataset sizes.'
  )
  parser.add_argument(
      '--input_data_dir',
      type=str,
      default=os.path.join(os.getenv('TEST_TMPDIR', '/tmp'),
                           'tensorflow/mnist/input_data'),
      help='Directory to put the input data.'
  )
  parser.add_argument(
      '--log_dir',
      type=str,
      default=os.path.join(os.getenv('TEST_TMPDIR', '/tmp'),
                           'tensorflow/mnist/logs/fully_connected_feed'),
      help='Directory to put the log data.'
  )
  parser.add_argument(
      '--fake_data',
      default=False,
      help='If true, uses fake data for unit testing.',
      action='store_true'
  )

  FLAGS, unparsed = parser.parse_known_args()
  tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)