cnn.py

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

# Imports
import numpy as np
import tensorflow as tf


def cnn_model_fn(features, labels, mode):
	"""Model function for CNN."""
	# Input Layer
	# array[0] = batch size = Size of the subset of examples to use when performing gradient descent during training.
	# array[1] = width of images
	# array[2] = height of images
	# array[3] = channels = Number of color channels in the example images (1 for grey, 3 for rgb)
	input_layer = tf.reshape(features, [-1, 40, 40, 1])

	# Convolutional Layer #1
	conv1 = tf.layers.conv2d(
			inputs=input_layer,
			filters=32,
			kernel_size=[5, 5],
			#  To specify that the output tensor should have the same width and height values as the input tensor
			# value can be "same" ou "valid"
			padding="same",
			activation=tf.nn.relu)

	# Pooling Layer #1
	pool1 = tf.layers.max_pooling2d(inputs=conv1, pool_size=[2, 2], strides=2)

	# Convolutional Layer #2 and Pooling Layer #2
	conv2 = tf.layers.conv2d(
			inputs=pool1,
			filters=64,
			kernel_size=[5, 5],
			padding="same",
			activation=tf.nn.relu)
	pool2 = tf.layers.max_pooling2d(inputs=conv2, pool_size=[2, 2], strides=2)

	# Dense Layer
	# pool2 width * pool2 height * pool2 channels
	pool2_shape = pool2.get_shape().as_list()
	pool2_flat = tf.reshape(pool2, [-1, pool2_shape[1] * pool2_shape[2] * pool2_shape[3]])
	dense = tf.layers.dense(inputs=pool2_flat, units=1024, activation=tf.nn.relu)
	dropout = tf.layers.dropout(
			inputs=dense, rate=0.4, training=mode == tf.estimator.ModeKeys.TRAIN)

	# Logits Layer
	logits = tf.layers.dense(inputs=dropout, units=1)

	print("================")
	print(logits)
	print("================")
	print("================")
	print(labels)
	print("================")

	predictions = {
		# Generate predictions (for PREDICT and EVAL mode)
		"classes":       tf.argmax(input=logits, axis=1),
		# Add `softmax_tensor` to the graph. It is used for PREDICT and by the
		# `logging_hook`.
		"probabilities": tf.nn.softmax(logits, name="softmax_tensor")
	}

	if mode == tf.estimator.ModeKeys.PREDICT:
		return tf.estimator.EstimatorSpec(mode=mode, predictions=predictions)

	# Calculate Loss (for both TRAIN and EVAL modes)
	loss = tf.losses.sparse_softmax_cross_entropy(labels=labels, logits=logits)

	# Configure the Training Op (for TRAIN mode)
	if mode == tf.estimator.ModeKeys.TRAIN:
		optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001)
		train_op = optimizer.minimize(
				loss=loss,
				global_step=tf.train.get_global_step())
		return tf.estimator.EstimatorSpec(mode=mode, loss=loss, train_op=train_op)

	# Add evaluation metrics (for EVAL mode)
	eval_metric_ops = {
		"accuracy": tf.metrics.accuracy(
				labels=labels, predictions=predictions["classes"])}
	return tf.estimator.EstimatorSpec(
			mode=mode, loss=loss, eval_metric_ops=eval_metric_ops)


def train_input_fn_custom(filenames_array, labels_array, batch_size):
	# Reads an image from a file, decodes it into a dense tensor, and resizes it to a fixed shape.
	def _parse_function(filename, label):
		image_string = tf.read_file(filename)
		image_decoded = tf.image.decode_png(image_string)
		image_resized = tf.image.resize_images(image_decoded, [40, 40])
		return image_resized, label

	filenames = tf.constant(filenames_array)
	labels = tf.constant(labels_array)

	# Convert the inputs to a Dataset
	dataset = tf.data.Dataset.from_tensor_slices((filenames, labels))

	# Convert filenames to a decoded image
	dataset = dataset.map(_parse_function)

	# Shuffle, repeat, and batch the examples.
	dataset = dataset.shuffle(1000).repeat().batch(batch_size)

	# Build the Iterator, and return the read end of the pipeline.
	return dataset.make_one_shot_iterator().get_next()


def train_label_custom():
	return [0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1]


def eval_input_fn_custom(filenames_array, labels_array, batch_size):
	# Reads an image from a file, decodes it into a dense tensor, and resizes it to a fixed shape.
	def _parse_function(filename, label):
		image_string = tf.read_file(filename)
		image_decoded = tf.image.decode_png(image_string)
		image_resized = tf.image.resize_images(image_decoded, [40, 40])
		return image_resized, label

	filenames = tf.constant(filenames_array)
	labels = tf.constant(labels_array)

	# Convert the inputs to a Dataset
	dataset = tf.data.Dataset.from_tensor_slices((filenames, labels))

	# Convert filenames to a decoded image
	dataset = dataset.map(_parse_function)

	# Shuffle, repeat, and batch the examples.
	dataset = dataset.batch(batch_size)

	# Build the Iterator, and return the read end of the pipeline.
	return dataset.make_one_shot_iterator().get_next()


def eval_label_custom():
	return [0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1]


def main(self):
	tf.logging.set_verbosity(tf.logging.INFO)

	filenames_train = ['blackcorner-data/1.png', 'blackcorner-data/2.png', 'blackcorner-data/3.png',
	                   'blackcorner-data/4.png', 'blackcorner-data/n1.png', 'blackcorner-data/n2.png',
	                   'blackcorner-data/n3.png', 'blackcorner-data/n4.png',
	                   'blackcorner-data/11.png', 'blackcorner-data/21.png', 'blackcorner-data/31.png',
	                   'blackcorner-data/41.png', 'blackcorner-data/n11.png', 'blackcorner-data/n21.png',
	                   'blackcorner-data/n31.png', 'blackcorner-data/n41.png'
	                   ]

	filenames_eval = ['blackcorner-data-eval/1.png', 'blackcorner-data-eval/2.png', 'blackcorner-data-eval/3.png',
	                  'blackcorner-data-eval/4.png', 'blackcorner-data-eval/n1.png', 'blackcorner-data-eval/n2.png',
	                  'blackcorner-data-eval/n3.png', 'blackcorner-data-eval/n4.png',
	                  'blackcorner-data-eval/11.png', 'blackcorner-data-eval/21.png', 'blackcorner-data-eval/31.png',
	                  'blackcorner-data-eval/41.png', 'blackcorner-data-eval/n11.png', 'blackcorner-data-eval/n21.png',
	                  'blackcorner-data-eval/n31.png', 'blackcorner-data-eval/n41.png'
	                  ]

	# Create the Estimator
	mnist_classifier = tf.estimator.Estimator(model_fn=cnn_model_fn, model_dir="/tmp/test_convnet_model")

	# Set up logging for predictions
	tensors_to_log = {"probabilities": "softmax_tensor"}
	logging_hook = tf.train.LoggingTensorHook(
			tensors=tensors_to_log, every_n_iter=50)

	# Train the model
	cust_train_input_fn = lambda: train_input_fn_custom(
			filenames_array=filenames_train, labels_array=train_label_custom(), batch_size=1)
	mnist_classifier.train(
			input_fn=cust_train_input_fn,
			steps=20000,
			hooks=[logging_hook])

	# Evaluate the model and print results
	# eval_input_fn = tf.estimator.inputs.numpy_input_fn(
	# 		x={"x": eval_data},
	# 		y=eval_labels,
	# 		num_epochs=1,
	# 		shuffle=False)
	# eval_results = mnist_classifier.evaluate(input_fn=cust_eval_input_fn)
	# print("======================")
	# print("Eval results : ")
	# print(eval_results)


if __name__ == "__main__":
	tf.app.run()