tensor_stream-opencl 0.1.0 → 0.1.1
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
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: d33b253833ef9d0db6be85834b2af716d5c2b1fb
|
|
4
|
+
data.tar.gz: b2e18fcf323ff2353bd251c0d6a9d01c6ef7d729
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 3933ab4454e758dd9bdfd0bf3bff4c856369da6704f6ded41e35ff50a271b259381eadda0e285a95d92a4f277975da56fb084cc03fb4b408403ace6021a146d9
|
|
7
|
+
data.tar.gz: 5f39e60e3b435d24a49cbeb0af399545c98a031999a26ff07f4c4d6badef58ce3799b5c39dfbe2f3899a2dd44b729b61cdd7e6b9a47f8a5d35e5b6487626fcc4
|
data/.gitignore
CHANGED
data/Gemfile.lock
CHANGED
|
@@ -1,13 +1,14 @@
|
|
|
1
1
|
PATH
|
|
2
2
|
remote: .
|
|
3
3
|
specs:
|
|
4
|
-
tensor_stream-opencl (0.1.
|
|
4
|
+
tensor_stream-opencl (0.1.1)
|
|
5
5
|
opencl_ruby_ffi
|
|
6
6
|
tensor_stream
|
|
7
7
|
|
|
8
8
|
GEM
|
|
9
9
|
remote: https://rubygems.org/
|
|
10
10
|
specs:
|
|
11
|
+
chunky_png (1.3.10)
|
|
11
12
|
concurrent-ruby (1.0.5)
|
|
12
13
|
deep_merge (1.2.1)
|
|
13
14
|
diff-lcs (1.3)
|
|
@@ -34,7 +35,8 @@ GEM
|
|
|
34
35
|
diff-lcs (>= 1.2.0, < 2.0)
|
|
35
36
|
rspec-support (~> 3.8.0)
|
|
36
37
|
rspec-support (3.8.0)
|
|
37
|
-
tensor_stream (0.8.
|
|
38
|
+
tensor_stream (0.8.5)
|
|
39
|
+
chunky_png
|
|
38
40
|
concurrent-ruby
|
|
39
41
|
deep_merge
|
|
40
42
|
|
|
@@ -2,5 +2,5 @@
|
|
|
2
2
|
% a_dtype = dtype_to_c_type(a)
|
|
3
3
|
% b_dtype = dtype_to_c_type(b)
|
|
4
4
|
% op = operator_to_c(fname)
|
|
5
|
-
<%= render 'bool_operand.cl', a_dtype: a_dtype, b_dtype: b_dtype, op: op, fname: fname, dtype: "#{a}_#{b}", result_t: '
|
|
5
|
+
<%= render 'bool_operand.cl', a_dtype: a_dtype, b_dtype: b_dtype, op: op, fname: fname, dtype: "#{a}_#{b}", result_t: 'uchar' %>
|
|
6
6
|
% end
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
% c_dtype = dtype_to_c_type(dtype)
|
|
2
|
-
__kernel void where_<%= dtype %>(const int M, const int N, __global const
|
|
2
|
+
__kernel void where_<%= dtype %>(const int M, const int N, __global const uchar *PRED, __global const <%= c_dtype %> *A, __global const <%= c_dtype %> *B, __global <%= c_dtype %> *C) {
|
|
3
3
|
// Get the index of the current element to be processed
|
|
4
4
|
const int globalRow = get_global_id(0); // Row ID of C (0..M)
|
|
5
5
|
const int globalCol = get_global_id(1); // Col ID of C (0..N)
|
|
@@ -5,7 +5,7 @@ require 'tensor_stream/evaluator/buffer'
|
|
|
5
5
|
require 'tensor_stream/opencl/opencl_buffer'
|
|
6
6
|
require 'tensor_stream/opencl/opencl_template_helper'
|
|
7
7
|
require 'tensor_stream/device'
|
|
8
|
-
require 'tensor_stream/
|
|
8
|
+
require 'tensor_stream/opencl/opencl_device'
|
|
9
9
|
require 'opencl_ruby_ffi'
|
|
10
10
|
require 'narray_ffi'
|
|
11
11
|
require 'tensor_stream/evaluator/base_evaluator'
|
|
@@ -781,7 +781,7 @@ module TensorStream
|
|
|
781
781
|
when :int16
|
|
782
782
|
NArray.sint(narray_size)
|
|
783
783
|
when :boolean
|
|
784
|
-
NArray.
|
|
784
|
+
NArray.byte(narray_size)
|
|
785
785
|
when :unknown
|
|
786
786
|
nil
|
|
787
787
|
else
|