tensor_stream-opencl 0.1.3 → 0.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/Gemfile.lock +11 -4
- data/benchmark/benchmark.rb +91 -0
- data/benchmark_intel.txt +36 -0
- data/lib/tensor_stream/opencl/array_ops.rb +395 -0
- data/lib/tensor_stream/opencl/images_ops.rb +62 -0
- data/lib/tensor_stream/opencl/kernels/abs.cl +6 -8
- data/lib/tensor_stream/opencl/kernels/acos.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/apply_adadelta.cl +2 -4
- data/lib/tensor_stream/opencl/kernels/apply_adagrad.cl +12 -0
- data/lib/tensor_stream/opencl/kernels/apply_adam.cl +2 -5
- data/lib/tensor_stream/opencl/kernels/apply_centered_rms_prop.cl +19 -0
- data/lib/tensor_stream/opencl/kernels/apply_gradient.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/apply_momentum.cl +2 -4
- data/lib/tensor_stream/opencl/kernels/apply_rms_prop.cl +16 -0
- data/lib/tensor_stream/opencl/kernels/asin.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/ceil.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/concat.cl +21 -0
- data/lib/tensor_stream/opencl/kernels/cos.cl +3 -5
- data/lib/tensor_stream/opencl/kernels/exp.cl +3 -5
- data/lib/tensor_stream/opencl/kernels/floor.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/log.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/log1p.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/negate.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/reciprocal.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/sigmoid.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/sign.cl +7 -8
- data/lib/tensor_stream/opencl/kernels/sin.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/split.cl +17 -0
- data/lib/tensor_stream/opencl/kernels/split_n.cl +18 -0
- data/lib/tensor_stream/opencl/kernels/sqrt.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/square.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/tan.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/tanh.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/tanh_grad.cl +3 -4
- data/lib/tensor_stream/opencl/kernels/unpack.cl +23 -0
- data/lib/tensor_stream/opencl/nn_ops.rb +111 -26
- data/lib/tensor_stream/opencl/opencl_buffer.rb +9 -0
- data/lib/tensor_stream/opencl/opencl_evaluator.rb +129 -172
- data/lib/tensor_stream/opencl/version.rb +1 -1
- data/samples/iris.data +150 -0
- data/samples/iris.rb +110 -0
- data/samples/mnist_data.rb +65 -0
- data/samples/multigpu.rb +73 -0
- data/samples/nearest_neighbor.rb +56 -0
- data/samples/rnn.rb +108 -0
- data/tensor_stream-opencl.gemspec +4 -1
- metadata +62 -3
data/samples/rnn.rb
ADDED
@@ -0,0 +1,108 @@
|
|
1
|
+
# RNN sample
|
2
|
+
#
|
3
|
+
# Ruby port Example based on article by Erik Hallström
|
4
|
+
# https://medium.com/@erikhallstrm/hello-world-rnn-83cd7105b767
|
5
|
+
#
|
6
|
+
#
|
7
|
+
|
8
|
+
require "bundler/setup"
|
9
|
+
require 'tensor_stream'
|
10
|
+
require 'tensor_stream/opencl'
|
11
|
+
require 'pry-byebug'
|
12
|
+
|
13
|
+
|
14
|
+
tf = TensorStream
|
15
|
+
|
16
|
+
num_epochs = 100
|
17
|
+
total_series_length = 50000
|
18
|
+
truncated_backprop_length = 15
|
19
|
+
state_size = 4
|
20
|
+
num_classes = 2
|
21
|
+
echo_step = 3
|
22
|
+
batch_size = 5
|
23
|
+
num_batches = total_series_length / batch_size / truncated_backprop_length
|
24
|
+
randomizer = TensorStream.random_uniform([total_series_length], minval: 0, maxval: 2)
|
25
|
+
|
26
|
+
|
27
|
+
def generate_data(randomizer, total_series_length, batch_size, echo_step)
|
28
|
+
x = randomizer.eval
|
29
|
+
y = x.rotate(-echo_step)
|
30
|
+
|
31
|
+
y[echo_step] = 0
|
32
|
+
|
33
|
+
x = TensorStream::TensorShape.reshape(x, [batch_size, -1]) # The first index changing slowest, subseries as rows
|
34
|
+
y = TensorStream::TensorShape.reshape(y, [batch_size, -1])
|
35
|
+
[x, y]
|
36
|
+
end
|
37
|
+
|
38
|
+
batchX_placeholder = tf.placeholder(:float32, shape: [batch_size, truncated_backprop_length], name: 'batch_x')
|
39
|
+
batchY_placeholder = tf.placeholder(:int32, shape: [batch_size, truncated_backprop_length], name: 'batch_y')
|
40
|
+
|
41
|
+
init_state = tf.placeholder(:float32, shape: [batch_size, state_size], name: 'init_state')
|
42
|
+
|
43
|
+
|
44
|
+
W = tf.variable(tf.random_uniform([state_size+1, state_size]), dtype: :float32, name: 'W')
|
45
|
+
b = tf.variable(tf.zeros([state_size]), dtype: :float32, name: 'b')
|
46
|
+
|
47
|
+
W2 = tf.variable(tf.random_uniform([state_size, num_classes]), dtype: :float32, name: 'W2')
|
48
|
+
b2 = tf.variable(tf.zeros([num_classes]), dtype: :float32, name: 'b2')
|
49
|
+
|
50
|
+
|
51
|
+
inputs_series = tf.unpack(batchX_placeholder, axis: 1)
|
52
|
+
labels_series = tf.unpack(batchY_placeholder, axis: 1)
|
53
|
+
|
54
|
+
current_state = init_state
|
55
|
+
states_series = []
|
56
|
+
|
57
|
+
inputs_series.each do |current_input|
|
58
|
+
current_input = tf.reshape(current_input, [batch_size, 1])
|
59
|
+
input_and_state_concatenated = tf.concat([current_input, current_state], 1) # Increasing number of columns
|
60
|
+
next_state = tf.tanh(tf.matmul(input_and_state_concatenated, W) + b) # Broadcasted addition
|
61
|
+
states_series << next_state
|
62
|
+
current_state = next_state
|
63
|
+
end
|
64
|
+
|
65
|
+
logits_series = states_series.collect do |state|
|
66
|
+
tf.matmul(state, W2) + b2
|
67
|
+
end
|
68
|
+
|
69
|
+
predictions_series = logits_series.collect do |logits|
|
70
|
+
tf.nn.softmax(logits)
|
71
|
+
end
|
72
|
+
|
73
|
+
losses = logits_series.zip(labels_series).collect do |logits, labels|
|
74
|
+
tf.nn.sparse_softmax_cross_entropy_with_logits(logits: logits, labels: labels)
|
75
|
+
end
|
76
|
+
total_loss = tf.reduce_mean(losses)
|
77
|
+
|
78
|
+
train_step = TensorStream::Train::AdagradOptimizer.new(0.3).minimize(total_loss)
|
79
|
+
|
80
|
+
puts "#{tf.get_default_graph.nodes.keys.size} nodes created"
|
81
|
+
zeros_state = tf.zeros([batch_size, state_size]).eval
|
82
|
+
tf.session do |sess|
|
83
|
+
sess.run(tf.global_variables_initializer)
|
84
|
+
(0..num_epochs).each do |epoch_idx|
|
85
|
+
x, y = generate_data(randomizer, total_series_length, batch_size, echo_step)
|
86
|
+
_current_state = zeros_state
|
87
|
+
print("New data, epoch", epoch_idx, "\n")
|
88
|
+
(0..num_batches - 1).each do |batch_idx|
|
89
|
+
start_idx = batch_idx * truncated_backprop_length
|
90
|
+
end_idx = start_idx + truncated_backprop_length
|
91
|
+
|
92
|
+
batchX = x.map { |x| x[start_idx...end_idx] }
|
93
|
+
batchY = y.map { |y| y[start_idx...end_idx] }
|
94
|
+
|
95
|
+
_total_loss, _train_step, _current_state, _predictions_series = sess.run(
|
96
|
+
[total_loss, train_step, current_state, predictions_series],
|
97
|
+
feed_dict: {
|
98
|
+
batchX_placeholder => batchX,
|
99
|
+
batchY_placeholder => batchY,
|
100
|
+
init_state => _current_state
|
101
|
+
})
|
102
|
+
|
103
|
+
if batch_idx%10 == 0
|
104
|
+
print("Step",batch_idx, " Loss ", _total_loss, "\n")
|
105
|
+
end
|
106
|
+
end
|
107
|
+
end
|
108
|
+
end
|
@@ -36,6 +36,9 @@ Gem::Specification.new do |spec|
|
|
36
36
|
spec.add_development_dependency "rake", "~> 10.0"
|
37
37
|
spec.add_development_dependency "rspec", "~> 3.0"
|
38
38
|
spec.add_development_dependency "pry-byebug"
|
39
|
-
spec.
|
39
|
+
spec.add_development_dependency "awesome_print"
|
40
|
+
spec.add_development_dependency "mnist-learn"
|
41
|
+
spec.add_dependency "tensor_stream", "~> 0.9.0"
|
40
42
|
spec.add_dependency "opencl_ruby_ffi"
|
43
|
+
spec.add_dependency "oily_png"
|
41
44
|
end
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: tensor_stream-opencl
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.2.0
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Joseph Dayo
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2018-
|
11
|
+
date: 2018-10-05 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|
@@ -66,8 +66,50 @@ dependencies:
|
|
66
66
|
- - ">="
|
67
67
|
- !ruby/object:Gem::Version
|
68
68
|
version: '0'
|
69
|
+
- !ruby/object:Gem::Dependency
|
70
|
+
name: awesome_print
|
71
|
+
requirement: !ruby/object:Gem::Requirement
|
72
|
+
requirements:
|
73
|
+
- - ">="
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: '0'
|
76
|
+
type: :development
|
77
|
+
prerelease: false
|
78
|
+
version_requirements: !ruby/object:Gem::Requirement
|
79
|
+
requirements:
|
80
|
+
- - ">="
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: '0'
|
83
|
+
- !ruby/object:Gem::Dependency
|
84
|
+
name: mnist-learn
|
85
|
+
requirement: !ruby/object:Gem::Requirement
|
86
|
+
requirements:
|
87
|
+
- - ">="
|
88
|
+
- !ruby/object:Gem::Version
|
89
|
+
version: '0'
|
90
|
+
type: :development
|
91
|
+
prerelease: false
|
92
|
+
version_requirements: !ruby/object:Gem::Requirement
|
93
|
+
requirements:
|
94
|
+
- - ">="
|
95
|
+
- !ruby/object:Gem::Version
|
96
|
+
version: '0'
|
69
97
|
- !ruby/object:Gem::Dependency
|
70
98
|
name: tensor_stream
|
99
|
+
requirement: !ruby/object:Gem::Requirement
|
100
|
+
requirements:
|
101
|
+
- - "~>"
|
102
|
+
- !ruby/object:Gem::Version
|
103
|
+
version: 0.9.0
|
104
|
+
type: :runtime
|
105
|
+
prerelease: false
|
106
|
+
version_requirements: !ruby/object:Gem::Requirement
|
107
|
+
requirements:
|
108
|
+
- - "~>"
|
109
|
+
- !ruby/object:Gem::Version
|
110
|
+
version: 0.9.0
|
111
|
+
- !ruby/object:Gem::Dependency
|
112
|
+
name: opencl_ruby_ffi
|
71
113
|
requirement: !ruby/object:Gem::Requirement
|
72
114
|
requirements:
|
73
115
|
- - ">="
|
@@ -81,7 +123,7 @@ dependencies:
|
|
81
123
|
- !ruby/object:Gem::Version
|
82
124
|
version: '0'
|
83
125
|
- !ruby/object:Gem::Dependency
|
84
|
-
name:
|
126
|
+
name: oily_png
|
85
127
|
requirement: !ruby/object:Gem::Requirement
|
86
128
|
requirements:
|
87
129
|
- - ">="
|
@@ -110,23 +152,31 @@ files:
|
|
110
152
|
- LICENSE.txt
|
111
153
|
- README.md
|
112
154
|
- Rakefile
|
155
|
+
- benchmark/benchmark.rb
|
156
|
+
- benchmark_intel.txt
|
113
157
|
- bin/console
|
114
158
|
- bin/setup
|
115
159
|
- lib/tensor_stream/opencl.rb
|
160
|
+
- lib/tensor_stream/opencl/array_ops.rb
|
161
|
+
- lib/tensor_stream/opencl/images_ops.rb
|
116
162
|
- lib/tensor_stream/opencl/kernels/_bool_operand.cl
|
117
163
|
- lib/tensor_stream/opencl/kernels/_operand.cl
|
118
164
|
- lib/tensor_stream/opencl/kernels/abs.cl
|
119
165
|
- lib/tensor_stream/opencl/kernels/acos.cl
|
120
166
|
- lib/tensor_stream/opencl/kernels/add.cl
|
121
167
|
- lib/tensor_stream/opencl/kernels/apply_adadelta.cl
|
168
|
+
- lib/tensor_stream/opencl/kernels/apply_adagrad.cl
|
122
169
|
- lib/tensor_stream/opencl/kernels/apply_adam.cl
|
170
|
+
- lib/tensor_stream/opencl/kernels/apply_centered_rms_prop.cl
|
123
171
|
- lib/tensor_stream/opencl/kernels/apply_gradient.cl
|
124
172
|
- lib/tensor_stream/opencl/kernels/apply_momentum.cl
|
173
|
+
- lib/tensor_stream/opencl/kernels/apply_rms_prop.cl
|
125
174
|
- lib/tensor_stream/opencl/kernels/argmax.cl
|
126
175
|
- lib/tensor_stream/opencl/kernels/argmin.cl
|
127
176
|
- lib/tensor_stream/opencl/kernels/asin.cl
|
128
177
|
- lib/tensor_stream/opencl/kernels/cast.cl
|
129
178
|
- lib/tensor_stream/opencl/kernels/ceil.cl
|
179
|
+
- lib/tensor_stream/opencl/kernels/concat.cl
|
130
180
|
- lib/tensor_stream/opencl/kernels/cond.cl.erb
|
131
181
|
- lib/tensor_stream/opencl/kernels/cos.cl
|
132
182
|
- lib/tensor_stream/opencl/kernels/div.cl.erb
|
@@ -156,6 +206,8 @@ files:
|
|
156
206
|
- lib/tensor_stream/opencl/kernels/softmax_cross.cl
|
157
207
|
- lib/tensor_stream/opencl/kernels/softmax_cross_grad.cl
|
158
208
|
- lib/tensor_stream/opencl/kernels/softmax_grad.cl
|
209
|
+
- lib/tensor_stream/opencl/kernels/split.cl
|
210
|
+
- lib/tensor_stream/opencl/kernels/split_n.cl
|
159
211
|
- lib/tensor_stream/opencl/kernels/sqrt.cl
|
160
212
|
- lib/tensor_stream/opencl/kernels/square.cl
|
161
213
|
- lib/tensor_stream/opencl/kernels/squared_difference.cl
|
@@ -163,6 +215,7 @@ files:
|
|
163
215
|
- lib/tensor_stream/opencl/kernels/tan.cl
|
164
216
|
- lib/tensor_stream/opencl/kernels/tanh.cl
|
165
217
|
- lib/tensor_stream/opencl/kernels/tanh_grad.cl
|
218
|
+
- lib/tensor_stream/opencl/kernels/unpack.cl
|
166
219
|
- lib/tensor_stream/opencl/kernels/where.cl
|
167
220
|
- lib/tensor_stream/opencl/math_ops.rb
|
168
221
|
- lib/tensor_stream/opencl/nn_ops.rb
|
@@ -171,6 +224,12 @@ files:
|
|
171
224
|
- lib/tensor_stream/opencl/opencl_evaluator.rb
|
172
225
|
- lib/tensor_stream/opencl/opencl_template_helper.rb
|
173
226
|
- lib/tensor_stream/opencl/version.rb
|
227
|
+
- samples/iris.data
|
228
|
+
- samples/iris.rb
|
229
|
+
- samples/mnist_data.rb
|
230
|
+
- samples/multigpu.rb
|
231
|
+
- samples/nearest_neighbor.rb
|
232
|
+
- samples/rnn.rb
|
174
233
|
- tensor_stream-opencl.gemspec
|
175
234
|
homepage: https://github.com/jedld/tensor_stream-opencl
|
176
235
|
licenses:
|