Upgrade to Pro — share decks privately, control downloads, hide ads and more …

20171209 Sakura ML Night

ARIYAMA Keiji
December 09, 2017

20171209 Sakura ML Night

2017年12月9日に大阪で開催された「さくらの機械学習ナイト」の発表資料です。

「TensorFlowによるNSFW(職場で不適切な)画像検出」について。

ARIYAMA Keiji

December 09, 2017

Transcript

65 slides
  1. C-LIS CO., LTD.  ༗ࢁܓೋʢ,FJKJ"3*:"."ʣ $-*4$0 -5% "OESPJEΞϓϦ։ൃνϣοτσΩϧ Photo by

    Koji MORIGUCHI (MORIGCHOWDER) ػցֶश͸ͪΐͬͱ΍ͬͨ͜ͱ͋Γ·͢ Twitter΍ͬͯ·ͤΜ
  2. { "generator": "Region Cropper", "file_name": "haruki_g17.png", "regions": [ { "probability":

    1.0, "label": 2, "rect": { "left": 97.0, "top": 251.0, "right": 285.0, "bottom": 383.0 } }, { "probability": 1.0, "label": 2, "rect": { "left": 536.0, "top": 175.0, "right": 730.0, "bottom": 321.0 } } ] } Region Cropper: https://github.com/keiji/region_cropper 
  3. ཧ૝ͷߏ੒  Downloader Face Detection Megane Detection ֬ೝɾमਖ਼ ೝࣝ݁Ռ ֶशʢ܇࿅ʣ

    λΠϜϥΠϯ ϝσΟΞ σʔληοτ ֶशʢ܇࿅ʣ TensorFlow  rsync
  4. Ϟσϧͷߏ଄  conv 3x3x64 stride 1 conv 3x3x64
 stride 1

    ReLU ReLU conv 3x3x128
 stride 1 conv 3x3x128
 stride 1 ReLU conv 3x3x256
 stride 1 conv 3x3x256
 stride 1 ReLU output 1 256x256x1 max_pool 2x2 stride 2 max_pool 2x2 stride 2 ReLU ReLU Sigmoid max_pool 2x2 stride 2 conv 3x3x64
 stride 1 ReLU fc 768 ReLU bn bn bn
  5. # モデル定義 NUM_CLASSES = 1 NAME = 'model3' IMAGE_SIZE =

    256 CHANNELS = 3 def prepare_layers(image, training=False): with tf.variable_scope('inference'): conv1 = tf.layers.conv2d(image, 64, [3, 3], [1, 1], padding='SAME', activation=tf.nn.relu, use_bias=False, trainable=training, name='conv1_1') conv1 = tf.layers.conv2d(conv1, 64, [3, 3], [1, 1], padding='VALID', activation=tf.nn.relu, use_bias=False, trainable=training, name='conv1_2') conv1 = tf.layers.batch_normalization(conv1, trainable=training, name='bn_1') 
  6. conv2 = tf.layers.conv2d(pool1, 128, [3, 3], [1, 1], padding='VALID', activation=tf.nn.relu,

    use_bias=False, trainable=training, name='conv2_1') conv2 = tf.layers.conv2d(conv2, 128, [3, 3], [1, 1], padding='VALID', activation=tf.nn.relu, use_bias=False, trainable=training, name='conv2_2') conv2 = tf.layers.batch_normalization(conv2, trainable=training, name='bn_2') pool2 = tf.layers.max_pooling2d(conv2, [2, 2], [2, 2]) 
  7. conv3 = tf.layers.conv2d(pool2, 256, [3, 3], [1, 1], padding='VALID', activation=tf.nn.relu,

    use_bias=False, trainable=training, name='conv4_1') conv3 = tf.layers.conv2d(conv3, 256, [3, 3], [1, 1], padding='VALID', activation=tf.nn.relu, use_bias=False, trainable=training, name='conv4_2') conv3 = tf.layers.batch_normalization(conv3, trainable=training, name='bn_4') pool3 = tf.layers.max_pooling2d(conv3, [2, 2], [2, 2]) conv = tf.layers.conv2d(pool3, 64, [1, 1], [1, 1], padding='VALID', activation=tf.nn.relu, use_bias=True, trainable=training, name='conv') return conv 
  8. def output_layers(prev, batch_size, keep_prob=0.8, training=False): flatten = tf.reshape(prev, [batch_size, -1])

    fc1 = tf.layers.dense(flatten, 768, trainable=training, activation=tf.nn.relu, name='fc1') fc1 = tf.layers.dropout(fc1, rate=keep_prob, training=training) output = tf.layers.dense(fc1, NUM_CLASSES, trainable=training, activation=None, name='output') return output 
  9. def _loss(logits, labels, batch_size, positive_ratio): cross_entropy = tf.nn.sigmoid_cross_entropy_with_logits( labels=labels, logits=logits)

    loss = tf.reduce_mean(cross_entropy) return loss def _init_optimizer(learning_rate): return tf.train.AdamOptimizer(learning_rate=learning_rate)  ޡࠩؔ਺ͱ࠷దԽΞϧΰϦζϜ
  10. def _hard_negative_mining(loss, labels, batch_size): positive_count = tf.reduce_sum(labels) positive_count = tf.reduce_max((positive_count,

    1)) negative_count = positive_count * HARD_SAMPLE_MINING_RATIO negative_count = tf.reduce_max((negative_count, 1)) negative_count = tf.reduce_min((negative_count, batch_size)) positive_losses = loss * labels negative_losses = loss - positive_losses top_negative_losses, _ = tf.nn.top_k(negative_losses, k=tf.cast(negative_count, tf.int32)) loss = (tf.reduce_sum(positive_losses / positive_count) + tf.reduce_sum(top_negative_losses / negative_count)) return loss  )BSE/FHBUJWF.JOJOH
  11. 

  12. 

  13. 

  14. TAGS = [ 'original_art', 'nsfw', 'like', 'photo', 'illust', 'comic', 'face',

    'girl', 'megane',  ϥϕϧʢλάʣ 'school_uniform', 'blazer_uniform', 'sailor_uniform', 'gl', 'kemono', 'boy', 'bl', 'cat', 'dog', 'food', 'dislike', ]
  15. Ϋϥε൑ఆϞσϧ  conv 3x3x64 stride 1 conv 3x3x64
 stride 1

    ReLU ReLU conv 3x3x128
 stride 1 conv 3x3x128
 stride 1 ReLU conv 3x3x256
 stride 1 conv 3x3x256
 stride 1 ReLU output 20 256x256x1 max_pool 2x2 stride 2 max_pool 2x2 stride 2 ReLU ReLU Sigmoid max_pool 2x2 stride 2 conv 3x3x64
 stride 1 ReLU fc 768 ReLU bn bn bn

More Decks by ARIYAMA Keiji

Other Decks in Technology