tensor_stream-opencl 0.2.5 → 0.2.6

data/samples/build_mnist_model.rb

@@ -0,0 +1,130 @@
+#!/usr/bin/env ruby
+
+# builds a frozen mnist image recognition model from scratch
+# based on mnist_data_3.0.rb
+
+require "bundler/setup"
+require 'tensor_stream'
+require 'mnist-learn'
+require 'csv'
+
+# Enable OpenCL hardware accelerated computation, not using OpenCL can be very slow
+require 'tensor_stream/opencl'
+require 'tensor_stream/utils/freezer'
+
+epochs = (ARGV[0] || 2000).to_i
+
+tf = TensorStream
+puts "Tensorstream version #{tf.__version__} with OpenCL lib #{TensorStream::Opencl::VERSION}"
+
+# Import MNIST data
+puts "downloading minst data"
+# Download images and labels into mnist.test (10K images+labels) and mnist.train (60K images+labels)
+mnist = Mnist.read_data_sets('/tmp/data', one_hot: true)
+
+puts "downloading finished"
+
+x = tf.placeholder(:float32, shape: [nil, 28, 28, 1])
+y_ = tf.placeholder(:float32, shape: [nil, 10])
+step = tf.placeholder(:int32)
+pkeep = tf.placeholder(tf.float32)
+
+# Network parameters
+K = 4 # first convolutional layer output depth
+L = 8 # second convolutional layer output depth
+M = 12 # third convolutional layer
+N = 200 # fully connected layer
+
+
+w1 = tf.variable(tf.truncated_normal([6, 6, 1, K], stddev: 0.1))
+b1 = tf.variable(tf.ones([K])/10)
+
+w2 = tf.variable(tf.truncated_normal([5, 5, K, L], stddev: 0.1))
+b2 = tf.variable(tf.ones([L])/10)
+
+w3 = tf.variable(tf.truncated_normal([4, 4, L, M], stddev: 0.1))
+b3 = tf.variable(tf.ones([M])/10)
+
+w4 = tf.variable(tf.truncated_normal([7 * 7 * M, N], stddev: 0.1))
+b4 = tf.variable(tf.ones([N])/10)
+
+w5 = tf.variable(tf.truncated_normal([N, 10], stddev: 0.1))
+b5 = tf.variable(tf.ones([10])/10)
+
+# The model
+stride = 1  # output is 28x28
+y1 = tf.nn.relu(tf.nn.conv2d(tf.reshape(x, [-1, 28, 28, 1]), w1, [1, stride, stride, 1], 'SAME') + b1)
+stride = 2  # output is 14x14
+y2 = tf.nn.relu(tf.nn.conv2d(y1, w2, [1, stride, stride, 1], 'SAME') + b2)
+stride = 2  # output is 7x7
+y3 = tf.nn.relu(tf.nn.conv2d(y2, w3, [1, stride, stride, 1], 'SAME') + b3)
+
+# reshape the output from the third convolution for the fully connected layer
+yy = tf.reshape(y3, [-1, 7 * 7 * M])
+y4 = tf.nn.relu(tf.matmul(yy, w4) + b4)
+
+ylogits = tf.matmul(y4, w5) + b5
+
+# model
+y = tf.nn.softmax(ylogits, name: 'out')
+
+# cross-entropy loss function (= -sum(Y_i * log(Yi)) ), normalised for batches of 100  images
+# TensorFlow provides the softmax_cross_entropy_with_logits function to avoid numerical stability
+# problems with log(0) which is NaN
+cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits: ylogits, labels: y_)
+cross_entropy = tf.reduce_mean(cross_entropy)*100
+
+is_correct = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
+accuracy =  tf.reduce_mean(tf.cast(is_correct, :float32))
+
+# training step, learning rate = 0.003
+lr = 0.0001.t +  tf.train.exponential_decay(0.003, step, 2000, 1/Math::E)
+train_step = TensorStream::Train::AdamOptimizer.new(lr).minimize(cross_entropy)
+
+sess = tf.session
+# Add ops to save and restore all the variables.
+
+init = tf.global_variables_initializer
+
+sess.run(init)
+
+#Setup save and restore
+model_save_path = "test_models/mnist_data_3.0"
+saver = tf::Train::Saver.new
+saver.restore(sess, model_save_path)
+
+mnist_train = mnist.train
+test_data = { x => mnist.test.images, y_ => mnist.test.labels, pkeep => 1.0 }
+
+puts "training #{epochs} epochs"
+(0..epochs).each do |i|
+  # load batch of images and correct answers
+  batch_x, batch_y = mnist_train.next_batch(100)
+  train_data = { x => batch_x, y_ => batch_y, step => i, pkeep => 0.75 }
+
+  # train
+  sess.run(train_step, feed_dict: train_data)
+
+  if (i % 10 == 0)
+
+    a_train, c_train, l = sess.run([accuracy, cross_entropy, lr], feed_dict: { x => batch_x, y_ => batch_y, step => i, pkeep => 1.0})
+    puts "#{i}: accuracy:#{a_train} loss:#{c_train} (lr:#{l})"
+  end
+
+  if (i % 100 == 0)
+    # success on test data?
+    a_test, c_test = sess.run([accuracy, cross_entropy], feed_dict: test_data, pkeep => 1.0)
+    puts("#{i}: ******** test accuracy: #{a_test} test loss: #{c_test}")
+
+    # save current state of the model
+    save_path = saver.save(sess, model_save_path)
+  end
+end
+
+# Freezing removes variables and turns them into constants
+puts "saving frozen model"
+freezer = TensorStream::Freezer.new
+freezer.convert(sess, model_save_path, 'mnist_model.yml')
+puts "done. model written at mnist_model.yml"
+
+

data/samples/classify.rb

@@ -1,5 +1,16 @@
 #!/usr/bin/env ruby
 
+# This script reads a grayscale PNG and predicts what digit it is
+# Note: You can use the dump_mnist.rb script to get those images
+#
+# This also requires an existing model:
+#
+# You can run
+#
+# ruby samples/build_mnist_model.rb
+#
+# to build a model that you can use
+
 require "bundler/setup"
 require "tensor_stream"
 require 'mnist-learn'
@@ -14,6 +25,7 @@ input = target_graph['Placeholder']
 output = TensorStream.argmax(target_graph['out'], 1)
 sess = TensorStream.session
 
+# invert since model was trained on a white on black png image
 reshaped_image = 255.0.t - decoded_image.reshape([1, 28, 28, 1]).cast(:float32)
 result = sess.run(output, feed_dict: { input => reshaped_image})
 

data/samples/mnist_data_3.0.rb

@@ -9,7 +9,6 @@
 require "bundler/setup"
 require 'tensor_stream'
 require 'mnist-learn'
-require 'pry-byebug'
 require 'csv'
 
 # Enable OpenCL hardware accelerated computation, not using OpenCL can be very slow

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: tensor_stream-opencl
 version: !ruby/object:Gem::Version
-  version: 0.2.5
+  version: 0.2.6
 platform: ruby
 authors:
 - Joseph Dayo
@@ -246,9 +246,10 @@ files:
 - lib/tensor_stream/opencl/opencl_evaluator.rb
 - lib/tensor_stream/opencl/opencl_template_helper.rb
 - lib/tensor_stream/opencl/version.rb
+- mnist_model.yml
+- samples/build_mnist_model.rb
 - samples/classify.rb
 - samples/dump_mnist.rb
-- samples/image_sort.rb
 - samples/iris.data
 - samples/iris.rb
 - samples/logistic_regression.rb

data/samples/image_sort.rb

@@ -1,9 +0,0 @@
-require "bundler/setup"
-require 'tensor_stream'
-require 'mnist-learn'
-require 'csv'
-
-# Enable OpenCL hardware accelerated computation, not using OpenCL can be very slow
-require 'tensor_stream/opencl'
-
-