train.py

#!/usr/bin/env python
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import os
import time

import numpy as np
import tensorflow as tf
import util
import coref_model as cm

if __name__ == "__main__":
  # tf.logging.set_verbosity(tf.logging.INFO)

  args = util.get_args()
  config = util.initialize_from_env(args.experiment, args.logdir)

  report_frequency = config["report_frequency"]
  eval_frequency = config["eval_frequency"]

  model = cm.CorefModel(config)

  print('# parameters:', np.sum([np.prod(v.get_shape().as_list()) for v in model.trainable_variables]))
  saver = tf.train.Saver()
  initial_step = 0

  log_dir = config["log_dir"]
  writer = tf.summary.FileWriter(log_dir, flush_secs=20)

  max_f1 = 0

  with tf.Session() as session:
    session.run(tf.global_variables_initializer())
    model.start_enqueue_thread(session)
    accumulated_loss = 0.0

    ckpt = tf.train.get_checkpoint_state(log_dir)
    if ckpt and ckpt.model_checkpoint_path:
      print("Restoring from: {}".format(ckpt.model_checkpoint_path))
      saver.restore(session, ckpt.model_checkpoint_path)
      initial_step = session.run(model.global_step)

    initial_time = time.time()
    while True:
      tf_loss, tf_global_step, _ = session.run([model.loss, model.global_step, model.train_op])
      accumulated_loss += tf_loss

      if tf_global_step % report_frequency == 0:
        steps_per_second = (tf_global_step - initial_step) / (time.time() - initial_time)

        average_loss = accumulated_loss / report_frequency
        print("[{}] loss={:.2f}, steps/s={:.2f}".format(tf_global_step,
                                                        average_loss,
                                                        steps_per_second))
        writer.add_summary(util.make_summary({"loss": average_loss,
                                              "learning_rate": session.run(model.learning_rate)}), tf_global_step)
        accumulated_loss = 0.0
        initial_time = time.time()
        initial_step = tf_global_step

      if tf_global_step % eval_frequency == 0:
        saver.save(session, os.path.join(log_dir, "model"), global_step=tf_global_step)
        eval_summary, eval_f1 = model.evaluate(session)

        if eval_f1 > max_f1:
          max_f1 = eval_f1
          util.copy_checkpoint(os.path.join(log_dir, "model-{}".format(tf_global_step)), os.path.join(log_dir, "model.max.ckpt"))

        writer.add_summary(eval_summary, tf_global_step)
        writer.add_summary(util.make_summary({"max_eval_f1": max_f1}), tf_global_step)

        print("[{}] evaL_f1={:.2f}, max_f1={:.2f}".format(tf_global_step, eval_f1, max_f1))