slim train their own classification model

Train your classifier model

运行环境为(在1.10及以上版本会出现cudnn无法初始化问题)：
tensorflow-1.9-gpu
ubuntu16.0.4
gtx1080x2

• Data is structured as follows:

dataset
│           ├── class1
│           │   ├── class1.jpg
│           │   ├── ...
│           │   └── classn.jpg
│           ├── class2
│           │   ├── class2.jpg
│           │   ├── ...
│           │   ├── classn.jpg

The best data set on the dataset directory, otherwise generate tfrecord code needs to be modified.
In download_and_convert_data.pyadding support for new data set (where the new data set is pocmans, download the script is written because the handling of different data sets , you need to add support for new raw data).

 elif FLAGS.dataset_name == 'pocmans'

In the dataset directory is added download_and_convert_pocmans.py, flowers script contains a download, convert tfrecord, clear temporary files. The data presented here has been so good to download the code is not required. Modify the code as follows (the following code import dataset_utilsneed to add pocmanssupport datasets):

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

import math
import os
import random
import sys

import tensorflow as tf

from datasets import dataset_utils


# The number of images in the validation set.
_NUM_VALIDATION = 150

# Seed for repeatability.
_RANDOM_SEED = 0

# The number of shards per dataset split.
_NUM_SHARDS = 5


class ImageReader(object):
  """Helper class that provides TensorFlow image coding utilities."""

  def __init__(self):
    # Initializes function that decodes RGB JPEG data.
    self._decode_jpeg_data = tf.placeholder(dtype=tf.string)
    self._decode_jpeg = tf.image.decode_jpeg(self._decode_jpeg_data, channels=3)

  def read_image_dims(self, sess, image_data):
    image = self.decode_jpeg(sess, image_data)
    return image.shape[0], image.shape[1]

  def decode_jpeg(self, sess, image_data):
    image = sess.run(self._decode_jpeg,
                     feed_dict={self._decode_jpeg_data: image_data})
    assert len(image.shape) == 3
    assert image.shape[2] == 3
    return image


def _get_filenames_and_classes(dataset_dir):
  """Returns a list of filenames and inferred class names.

  Args:
    dataset_dir: A directory containing a set of subdirectories representing
      class names. Each subdirectory should contain PNG or JPG encoded images.

  Returns:
    A list of image file paths, relative to `dataset_dir` and the list of
    subdirectories, representing class names.
  """
  flower_root = os.path.join(dataset_dir, 'dataset')
  directories = []
  class_names = []
  for filename in os.listdir(flower_root):
    path = os.path.join(flower_root, filename)
    if os.path.isdir(path):
      directories.append(path)
      class_names.append(filename)

  photo_filenames = []
  for directory in directories:
    for filename in os.listdir(directory):
      path = os.path.join(directory, filename)
      photo_filenames.append(path)

  return photo_filenames, sorted(class_names)


def _get_dataset_filename(dataset_dir, split_name, shard_id):
  output_filename = 'pocmans_%s_%05d-of-%05d.tfrecord' % (
      split_name, shard_id, _NUM_SHARDS)
  return os.path.join(dataset_dir, output_filename)


def _convert_dataset(split_name, filenames, class_names_to_ids, dataset_dir):
  """Converts the given filenames to a TFRecord dataset.

  Args:
    split_name: The name of the dataset, either 'train' or 'validation'.
    filenames: A list of absolute paths to png or jpg images.
    class_names_to_ids: A dictionary from class names (strings) to ids
      (integers).
    dataset_dir: The directory where the converted datasets are stored.
  """
  assert split_name in ['train', 'validation']

  num_per_shard = int(math.ceil(len(filenames) / float(_NUM_SHARDS)))

  with tf.Graph().as_default():
    image_reader = ImageReader()

    with tf.Session('') as sess:

      for shard_id in range(_NUM_SHARDS):
        output_filename = _get_dataset_filename(
            dataset_dir, split_name, shard_id)

        with tf.python_io.TFRecordWriter(output_filename) as tfrecord_writer:
          start_ndx = shard_id * num_per_shard
          end_ndx = min((shard_id+1) * num_per_shard, len(filenames))
          for i in range(start_ndx, end_ndx):
            sys.stdout.write('\r>> Converting image %d/%d shard %d' % (
                i+1, len(filenames), shard_id))
            sys.stdout.flush()

            # Read the filename:
            image_data = tf.gfile.FastGFile(filenames[i], 'rb').read()
            height, width = image_reader.read_image_dims(sess, image_data)

            class_name = os.path.basename(os.path.dirname(filenames[i]))
            class_id = class_names_to_ids[class_name]

            example = dataset_utils.image_to_tfexample(
                image_data, b'jpg', height, width, class_id)
            tfrecord_writer.write(example.SerializeToString())

  sys.stdout.write('\n')
  sys.stdout.flush()



def _dataset_exists(dataset_dir):
  for split_name in ['train', 'validation']:
    for shard_id in range(_NUM_SHARDS):
      output_filename = _get_dataset_filename(
          dataset_dir, split_name, shard_id)
      if not tf.gfile.Exists(output_filename):
        return False
  return True


def run(dataset_dir):
  """Runs the download and conversion operation.

  Args:
    dataset_dir: The dataset directory where the dataset is stored.
  """
  if not tf.gfile.Exists(dataset_dir):
    tf.gfile.MakeDirs(dataset_dir)

  if _dataset_exists(dataset_dir):
    print('Dataset files already exist. Exiting without re-creating them.')
    return

  photo_filenames, class_names = _get_filenames_and_classes(dataset_dir)
  class_names_to_ids = dict(zip(class_names, range(len(class_names))))

  # Divide into train and test:
  random.seed(_RANDOM_SEED)
  random.shuffle(photo_filenames)
  training_filenames = photo_filenames[_NUM_VALIDATION:]
  validation_filenames = photo_filenames[:_NUM_VALIDATION]

  # First, convert the training and validation sets.
  _convert_dataset('train', training_filenames, class_names_to_ids,
                   dataset_dir)
  _convert_dataset('validation', validation_filenames, class_names_to_ids,
                   dataset_dir)

  # Finally, write the labels file:
  labels_to_class_names = dict(zip(range(len(class_names)), class_names))
  dataset_utils.write_label_file(labels_to_class_names, dataset_dir)

  print('\nFinished converting the Pocmans dataset!')

dataset added pocman.py :

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

import os
import tensorflow as tf

from datasets import dataset_utils

slim = tf.contrib.slim

_FILE_PATTERN = 'pocmans_%s_*.tfrecord'

SPLITS_TO_SIZES = {'train': 1168, 'validation': 150}

_NUM_CLASSES = 5

_ITEMS_TO_DESCRIPTIONS = {
    'image': 'A color image of varying size.',
    'label': 'A single integer between 0 and 4',
}


def get_split(split_name, dataset_dir, file_pattern=None, reader=None):
  if split_name not in SPLITS_TO_SIZES:
    raise ValueError('split name %s was not recognized.' % split_name)

  if not file_pattern:
    file_pattern = _FILE_PATTERN
  file_pattern = os.path.join(dataset_dir, file_pattern % split_name)

  # Allowing None in the signature so that dataset_factory can use the default.
  if reader is None:
    reader = tf.TFRecordReader

  keys_to_features = {
      'image/encoded': tf.FixedLenFeature((), tf.string, default_value=''),
      'image/format': tf.FixedLenFeature((), tf.string, default_value='png'),
      'image/class/label': tf.FixedLenFeature(
          [], tf.int64, default_value=tf.zeros([], dtype=tf.int64)),
  }

  items_to_handlers = {
      'image': slim.tfexample_decoder.Image(),
      'label': slim.tfexample_decoder.Tensor('image/class/label'),
  }

  decoder = slim.tfexample_decoder.TFExampleDecoder(
      keys_to_features, items_to_handlers)

  labels_to_names = None
  if dataset_utils.has_labels(dataset_dir):
    labels_to_names = dataset_utils.read_label_file(dataset_dir)

  return slim.dataset.Dataset(
      data_sources=file_pattern,
      reader=reader,
      decoder=decoder,
      num_samples=SPLITS_TO_SIZES[split_name],
      items_to_descriptions=_ITEMS_TO_DESCRIPTIONS,
      num_classes=_NUM_CLASSES,
      labels_to_names=labels_to_names)

Training script ( nception_v3 ):

DATASET_DIR=/home/liushuai/pocman_dataset/tfrecord
TRAIN_DIR=/home/liushuai/models/research/slim/pocman_output
CHECKPOINT_PATH=/tmp/data/inception_v3.ckpt
python train_image_classifier.py \
    --train_dir=${TRAIN_DIR} \
    --dataset_dir=${DATASET_DIR} \
    --dataset_name=pocmans \
    --dataset_split_name=train \
    --model_name=inception_v3 \
    --checkpoint_path=${CHECKPOINT_PATH} \
    --checkpoint_exclude_scopes=InceptionV3/Logits,InceptionV3/AuxLogits \
    --trainable_scopes=InceptionV3/Logits,InceptionV3/AuxLogits

Training output is as follows:

INFO:tensorflow:global step 22300: loss = 0.7281 (0.269 sec/step)
INFO:tensorflow:global step 22310: loss = 0.4304 (0.186 sec/step)
INFO:tensorflow:global step 22320: loss = 0.4609 (0.175 sec/step)
INFO:tensorflow:global step 22330: loss = 0.4486 (0.194 sec/step)
INFO:tensorflow:global step 22340: loss = 0.5034 (0.220 sec/step)
INFO:tensorflow:global step 22350: loss = 0.6333 (0.290 sec/step)
INFO:tensorflow:global step 22360: loss = 0.5172 (0.235 sec/step)
INFO:tensorflow:global step 22370: loss = 0.7086 (0.275 sec/step)

tensorboard output

Evaluation data (eval_pocman.sh)

python eval_image_classifier.py \
    --alsologtostderr \
    --checkpoint_path=${CHECKPOINT_FILE} \
    --dataset_dir=${DATASET_DIR} \
    --dataset_name=pocmans \
    --dataset_split_name=validation \
    --model_name=inception_v3

Output is as follows:

INFO:tensorflow:Evaluating /home/liushuai/models/research/slim/pocman_output/model.ckpt-49593
INFO:tensorflow:Starting evaluation at 2019-01-22-09:29:13
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /home/liushuai/models/research/slim/pocman_output/model.ckpt-49593
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Evaluation [1/2]
INFO:tensorflow:Evaluation [2/2]
eval/Accuracy[0.935]
eval/Recall_5[1]
INFO:tensorflow:Finished evaluation at 2019-01-22-09:29:49

Export pb map:

python export_inference_graph.py \
  --alsologtostderr \
  --model_name=inception_v3 \
  --output_file=./pocman_inception_v3_inf_graph.pb