tensor_stream-opencl 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 0d1536e4ed46b18be21e6c0a12a6c06a854d6695
+  data.tar.gz: 458e8fcf33d992c3e0ee3660bf632ef5bab49b8e
+SHA512:
+  metadata.gz: 9841051b9a4ef0809eb243158aeb1bfa19e8efeb81f30e99795a019e86b08ff65b02b0bd0c64e4d31350bc496a6371298e945d77f52456439cfd419b02fef2ce
+  data.tar.gz: 29d8efc7a82aac8ecd959a46d04e02c96efeca311ef9340fd74be6bbe1b89a62d8789a2cc867c1d62c2860dffcf03c203bd825ebe159fc0be4db3ec457eb2ceb

data/.gitignore

@@ -0,0 +1,11 @@
+/.bundle/
+/.yardoc
+/_yardoc/
+/coverage/
+/doc/
+/pkg/
+/spec/reports/
+/tmp/
+
+# rspec failure tracking
+.rspec_status

data/.rspec

@@ -0,0 +1,3 @@
+--format documentation
+--color
+--require spec_helper

data/.travis.yml

@@ -0,0 +1,5 @@
+sudo: false
+language: ruby
+rvm:
+  - 2.5.1
+before_install: gem install bundler -v 1.16.2

data/CODE_OF_CONDUCT.md

@@ -0,0 +1,74 @@
+# Contributor Covenant Code of Conduct
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as
+contributors and maintainers pledge to making participation in our project and
+our community a harassment-free experience for everyone, regardless of age, body
+size, disability, ethnicity, gender identity and expression, level of experience,
+nationality, personal appearance, race, religion, or sexual identity and
+orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment
+include:
+
+* Using welcoming and inclusive language
+* Being respectful of differing viewpoints and experiences
+* Gracefully accepting constructive criticism
+* Focusing on what is best for the community
+* Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery and unwelcome sexual attention or
+advances
+* Trolling, insulting/derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or electronic
+  address, without explicit permission
+* Other conduct which could reasonably be considered inappropriate in a
+  professional setting
+
+## Our Responsibilities
+
+Project maintainers are responsible for clarifying the standards of acceptable
+behavior and are expected to take appropriate and fair corrective action in
+response to any instances of unacceptable behavior.
+
+Project maintainers have the right and responsibility to remove, edit, or
+reject comments, commits, code, wiki edits, issues, and other contributions
+that are not aligned to this Code of Conduct, or to ban temporarily or
+permanently any contributor for other behaviors that they deem inappropriate,
+threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies both within project spaces and in public spaces
+when an individual is representing the project or its community. Examples of
+representing a project or community include using an official project e-mail
+address, posting via an official social media account, or acting as an appointed
+representative at an online or offline event. Representation of a project may be
+further defined and clarified by project maintainers.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be
+reported by contacting the project team at joseph.dayo@gmail.com. All
+complaints will be reviewed and investigated and will result in a response that
+is deemed necessary and appropriate to the circumstances. The project team is
+obligated to maintain confidentiality with regard to the reporter of an incident.
+Further details of specific enforcement policies may be posted separately.
+
+Project maintainers who do not follow or enforce the Code of Conduct in good
+faith may face temporary or permanent repercussions as determined by other
+members of the project's leadership.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
+available at [http://contributor-covenant.org/version/1/4][version]
+
+[homepage]: http://contributor-covenant.org
+[version]: http://contributor-covenant.org/version/1/4/

data/Gemfile

@@ -0,0 +1,6 @@
+source "https://rubygems.org"
+
+git_source(:github) {|repo_name| "https://github.com/#{repo_name}" }
+
+# Specify your gem's dependencies in tensor_stream-opencl.gemspec
+gemspec

data/Gemfile.lock

@@ -0,0 +1,51 @@
+PATH
+  remote: .
+  specs:
+    tensor_stream-opencl (0.1.0)
+      opencl_ruby_ffi
+      tensor_stream
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    concurrent-ruby (1.0.5)
+    deep_merge (1.2.1)
+    diff-lcs (1.3)
+    ffi (1.9.25)
+    narray (0.6.1.2)
+    narray_ffi (1.4.3)
+      ffi (~> 1.9, >= 1.9.3)
+      narray (~> 0.6, >= 0.6.0.8)
+    opencl_ruby_ffi (1.3.4)
+      ffi (~> 1.9, >= 1.9.3)
+      narray (~> 0.6, >= 0.6.0.8)
+      narray_ffi (~> 1.0, >= 1.0.0)
+    rake (10.5.0)
+    rspec (3.8.0)
+      rspec-core (~> 3.8.0)
+      rspec-expectations (~> 3.8.0)
+      rspec-mocks (~> 3.8.0)
+    rspec-core (3.8.0)
+      rspec-support (~> 3.8.0)
+    rspec-expectations (3.8.1)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.8.0)
+    rspec-mocks (3.8.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.8.0)
+    rspec-support (3.8.0)
+    tensor_stream (0.8.1)
+      concurrent-ruby
+      deep_merge
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bundler (~> 1.16)
+  rake (~> 10.0)
+  rspec (~> 3.0)
+  tensor_stream-opencl!
+
+BUNDLED WITH
+   1.16.2

data/LICENSE.txt

@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2018 Joseph Dayo
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.

data/README.md

@@ -0,0 +1,58 @@
+# TensorStream::Opencl
+
+This gem provides an OpenCL backend for TensorStream (https://github.com/jedld/tensor_stream). OpenCL is an open standard
+that allows running compute applications on heterogenous platforms like CPUs and GPUs.
+
+## Installation
+
+Make sure OpenCL device drivers are installed in your system. You may refer to the following links:
+
+### Nvidia
+
+https://developer.nvidia.com/opencl
+
+### AMD
+
+https://support.amd.com/en-us/kb-articles/Pages/OpenCL2-Driver.aspx
+
+
+### Intel
+
+https://software.intel.com/en-us/articles/opencl-drivers
+
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'tensor_stream-opencl'
+```
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install tensor_stream-opencl
+
+## Usage
+
+Simply including this gem will allow tensor_stream to automatically select opencl devices for use in your computation
+
+## Development
+
+After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
+
+To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
+
+## Contributing
+
+Bug reports and pull requests are welcome on GitHub at https://github.com/jedld/tensor_stream-opencl. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [Contributor Covenant](http://contributor-covenant.org) code of conduct.
+
+## License
+
+The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
+
+## Code of Conduct
+
+Everyone interacting in the TensorStream::Opencl project’s codebases, issue trackers, chat rooms and mailing lists is expected to follow the [code of conduct](https://github.com/[USERNAME]/tensor_stream-opencl/blob/master/CODE_OF_CONDUCT.md).

data/Rakefile

@@ -0,0 +1,6 @@
+require "bundler/gem_tasks"
+require "rspec/core/rake_task"
+
+RSpec::Core::RakeTask.new(:spec)
+
+task :default => :spec

data/bin/console

@@ -0,0 +1,14 @@
+#!/usr/bin/env ruby
+
+require "bundler/setup"
+require "tensor_stream/opencl"
+
+# You can add fixtures and/or initialization code here to make experimenting
+# with your gem easier. You can also use a different console, if you like.
+
+# (If you use this, don't forget to add pry to your Gemfile!)
+# require "pry"
+# Pry.start
+
+require "irb"
+IRB.start(__FILE__)

data/bin/setup

@@ -0,0 +1,8 @@
+#!/usr/bin/env bash
+set -euo pipefail
+IFS=$'\n\t'
+set -vx
+
+bundle install
+
+# Do any other automated setup that you need to do here

data/lib/tensor_stream/opencl.rb

@@ -0,0 +1,7 @@
+require "tensor_stream/opencl/version"
+require "tensor_stream/opencl/opencl_evaluator"
+
+module TensorStream
+  module Opencl
+  end
+end

data/lib/tensor_stream/opencl/kernels/_bool_operand.cl

@@ -0,0 +1,45 @@
+ // same dimension add floating point op
+ __kernel void <%= fname%>_<%= dtype %>(const int M, const int N, const int switch_op, __global const <%= a_dtype %> *A, __global <%= b_dtype %> *B, __global <%= result_t %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    C[globalRow * N + globalCol] = A[globalRow * N + globalCol] <%= op %> B[globalRow * N + globalCol] ? 1 : 0;
+}
+
+ // 1D + Scalar floating point add op
+ __kernel void <%=fname%>_c_<%= dtype %>(const int M, const int N, const int switch_op, __global const <%= a_dtype %> *A, __global <%= b_dtype %> *B, __global <%= result_t %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    if (switch_op == 0) {
+      C[globalRow * N + globalCol] = A[globalRow * N + globalCol] <%= op %> B[0] ? 1 : 0;
+    } else {
+      C[globalRow * N + globalCol] = B[0] <%= op %> A[globalRow * N + globalCol] ? 1 : 0;
+    }
+}
+
+ // 1D + Scalar floating point add op broadcast
+ __kernel void <%= fname%>_b_<%= dtype %>(const int M, const int N, const int M2, const int N2, const int switch_op,__global const <%= a_dtype %> *A, __global <%= b_dtype %> *B, __global <%= result_t %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    int b_m_index = globalRow;
+    int b_n_index = globalCol;
+
+    if ( b_m_index >= M2) {
+      b_m_index = b_m_index % M2;
+    };
+
+    if (b_n_index >= N2) {
+      b_n_index = b_n_index % N2;
+    }
+
+    if (switch_op == 0) {
+      C[globalRow * N + globalCol] = A[globalRow * N + globalCol] <%= op %> B[b_m_index * N2 + b_n_index] ? 1 : 0;
+    } else {
+      C[globalRow * N + globalCol] = B[b_m_index * N2 + b_n_index] <%= op %> A[globalRow * N + globalCol] ? 1 : 0;
+    }
+}

data/lib/tensor_stream/opencl/kernels/_operand.cl

@@ -0,0 +1,45 @@
+ // same dimension add floating point op
+ __kernel void <%= fname%>_<%= dtype %>(const int M, const int N, const int switch_op, __global const <%= c_dtype %> *A, __global <%= c_dtype %> *B, __global <%= result_t %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    C[globalRow * N + globalCol] = A[globalRow * N + globalCol] <%= op %> B[globalRow * N + globalCol];
+}
+
+ // 1D + Scalar floating point add op
+ __kernel void <%=fname%>_c_<%= dtype %>(const int M, const int N, const int switch_op, __global const <%= c_dtype %> *A, __global <%= c_dtype %> *B, __global <%= result_t %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    if (switch_op == 0) {
+      C[globalRow * N + globalCol] = A[globalRow * N + globalCol] <%= op %> B[0];
+    } else {
+      C[globalRow * N + globalCol] = B[0] <%= op %> A[globalRow * N + globalCol];
+    }
+}
+
+ // 1D + Scalar floating point add op broadcast
+ __kernel void <%= fname%>_b_<%= dtype %>(const int M, const int N, const int M2, const int N2, const int switch_op,__global const <%= c_dtype %> *A, __global <%= c_dtype %> *B, __global <%= result_t %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    int b_m_index = globalRow;
+    int b_n_index = globalCol;
+
+    if ( b_m_index >= M2) {
+      b_m_index = b_m_index % M2;
+    };
+
+    if (b_n_index >= N2) {
+      b_n_index = b_n_index % N2;
+    }
+
+    if (switch_op == 0) {
+      C[globalRow * N + globalCol] = A[globalRow * N + globalCol] <%= op %> B[b_m_index * N2 + b_n_index];
+    } else {
+      C[globalRow * N + globalCol] = B[b_m_index * N2 + b_n_index] <%= op %> A[globalRow * N + globalCol];
+    }
+}

data/lib/tensor_stream/opencl/kernels/abs.cl

@@ -0,0 +1,20 @@
+% c_dtype = dtype_to_c_type(dtype)
+% if TensorStream::Ops::FLOATING_POINT_TYPES.include?(dtype)
+__kernel void abs_<%= dtype%>(const int M, const int N, __global const <%= c_dtype %> *A, __global <%= c_dtype %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    C[globalRow * N + globalCol] = fabs(A[globalRow * N + globalCol]);
+}
+% else
+% %w[int int32].each do |dt|
+__kernel void abs_<%= dt %>(const int M, const int N, __global const <%= c_dtype %> *A, __global <%= c_dtype %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    C[globalRow * N + globalCol] = fabs((float)A[globalRow * N + globalCol]);
+}
+% end
+%end

data/lib/tensor_stream/opencl/kernels/acos.cl

@@ -0,0 +1,8 @@
+% c_dtype = dtype_to_c_type(dtype)
+__kernel void acos_<%= dtype %>(const int M, const int N, __global const <%= c_dtype %> *A, __global <%= c_dtype %> *C) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+
+    C[globalRow * N + globalCol] = acos(A[globalRow * N + globalCol]);
+}

data/lib/tensor_stream/opencl/kernels/add.cl

@@ -0,0 +1,3 @@
+% c_dtype = dtype_to_c_type(dtype)
+% op = operator_to_c('add')
+<%= render 'operand.cl', c_dtype: c_dtype, op: op, fname: 'add', dtype: "#{a}_#{b}", result_t: c_dtype %>

data/lib/tensor_stream/opencl/kernels/apply_adam.cl

@@ -0,0 +1,23 @@
+% c_dtype = dtype_to_c_type(dtype)
+ // same dimension add floating point op
+ __kernel void apply_adam_<%= dtype %>(const int M, const int N,
+                                       __global const <%= c_dtype %> *grad,
+                                       __global const <%= c_dtype %> *learning_rate,
+                                       __global const <%= c_dtype %> *beta1_power,
+                                       __global const <%= c_dtype %> *beta2_power,
+                                       __global const <%= c_dtype %> *beta1,
+                                       __global const <%= c_dtype %> *beta2,
+                                       __global const <%= c_dtype %> *epsilon,
+                                       __global <%= c_dtype %> *momentum,
+                                       __global <%= c_dtype %> *output, __global <%= c_dtype %> *v) {
+    // Get the index of the current element to be processed
+    const int globalRow = get_global_id(0); // Row ID of C (0..M)
+    const int globalCol = get_global_id(1); // Col ID of C (0..N)
+    const int index = globalRow * N + globalCol;
+
+    <%= c_dtype %> alpha = learning_rate[0] * sqrt(1.0 - beta2_power[0]) / (1.0 - beta1_power[0]);
+
+    momentum[index] += (grad[index] - momentum[index]) * (1.0 - beta1[0]);
+    v[index] += (grad[index] * grad[index] - v[index]) * (1.0 - beta2[0]);
+    output[index] -= (momentum[index] * alpha) / ( sqrt(v[index]) + epsilon[0] );
+}