Gitiles
Code Review
Sign In
review.mlplatform.org
/
ml
/
ComputeLibrary
/
906443f38fe2c168f4a0b2db5cd9852865fc4922
/
src
/
core
906443f
COMPMID-415 - Fixed bug in CLDepthConcatenateKernel
by Gian Marco Iodice
· 7 years ago
64ebe5b
COMPMID-519: Add support for Lower and Upper Bounded RELU for CL/NEON
by Georgios Pinitas
· 7 years ago
07781ac
COMPMID-518 - Bare metal support
by Michalis Spyrou
· 7 years ago
1fab09f
COMPMID-424 Implemented reference implementation, new output valid region and validation tests (NEON and CL) for Scale
by Isabella Gottardi
· 7 years ago
04f089c
COMPMID-476 L2 Normalization for CL
by Michalis Spyrou
· 7 years ago
6dba6d1
COMPMID-417: Fix invalid read in CL GEMM accumulate biases
by Moritz Pflanzer
· 7 years ago
3e36369
COMPMID-358 Implement OpenCL ROI Pooling
by SiCong Li
· 7 years ago
edfa9f4
COMPMID-477 - Optimized batched case in CLConvolutionLayer
by Gian Marco Iodice
· 7 years ago
dc460f1
COMPMID-417: Port PoolingLayer to new validation.
by Georgios Pinitas
· 7 years ago
32982d8
COMPMID-439: Implement NEON Dequantization Layer.
by Michele Di Giorgio
· 7 years ago
a184836
COMPMID-417: Fix valid region
by Moritz Pflanzer
· 7 years ago
5f91072
COMPMID-513 Choose maximum local workgroup size at run time
by steniu01
· 7 years ago
93a690e
COMPMID-452 CL Depthwise Separable Convolution Layer kernel implementation, validation and benchmarking for 3x3xC depthwise filter and DataType::F32.
by Giorgio Arena
· 7 years ago
06da39d
COMPMID-345: Added support for 5x5 kernels in NEDirectConvolution
by Pablo Tello
· 7 years ago
d60a6b9
COMPMID-477 - Optimized CLNormalizationLayer
by Gian Marco Iodice
· 7 years ago
4e09b38
COMPMID-439: Implement NEON Quantization Layer.
by Michele Di Giorgio
· 7 years ago
62eeae4
COMPMID-345: Optimization for NEFillBorder kernel.
by Pablo Tello
· 7 years ago
0c7614f
COMPMID-431 Port OpenCL pooling layer to use fixed point
by steniu01
· 7 years ago
cb29283
COMPMID-477 - Optimizing Pooling 3x3 with stride_x <= 3 on OpenCL
by Gian Marco Iodice
· 7 years ago
484e7b3
COMPMID-417: Cleanup NEON FullyConnectedLayer
by Moritz Pflanzer
· 7 years ago
768e9f1
COMPMID-417: Cleanup CL FullyConnectedLayer
by Moritz Pflanzer
· 7 years ago
75af28e
COMPMID-417 - Fixed call of direct convolution 1x1 for bifrost
by Gian Marco Iodice
· 7 years ago
1246b63
COMPMID-477 - Optimized Direct Convolution 3x3 and 5x5 (f32) for Bifrost.
by Gian Marco Iodice
· 7 years ago
409ee0a
COMPMID-417: Add in-place support for batch-normalization.
by Georgios Pinitas
· 7 years ago
1e5c157
COMPMID-450 Add YOLOV2 benchmark tests
by SiCong Li
· 7 years ago
db00668
COMPMID-478 Implemnt CL direct convolution 5x5
by steniu01
· 7 years ago
def665a
COMPMID-474 - Add support for QS8/QS16 DirectConvolution CL
by Michalis Spyrou
· 7 years ago
fc2817d
COMPMID-424 NEON/CL Harris Corners validation tests.
by Giorgio Arena
· 7 years ago
bbd9fb9
COMPMID-412: Port PoolingLayer to use fixed point 16.
by Michalis Spyrou
· 7 years ago
2eac5bd
COMPMID-417 - Fixed bug in CLCol2ImKernek related to the stride passed during the configuration
by Gian Marco Iodice
· 7 years ago
63d76a7
COMPMID-417 - Added clFinish to CLSymbols
by Gian Marco Iodice
· 7 years ago
40ff03b
COMPMID-417 - Bug Fix WarpPerspective kernel
by Isabella Gottardi
· 7 years ago
868e541
COMPMID-459 Collapse CL Im2col's higher dimensions
by steniu01
· 7 years ago
5cb4d6a
COMPMID-477 - Optimizing CLDirectConvolution 3x3 on OpenCL and added the auto configuration
by Gian Marco Iodice
· 7 years ago
6203153
COMPMID-424 Implemented reference implementation and validation tests (NEON and CL) for Warp Perspective
by Isabella Gottardi
· 7 years ago
d625360
COMPMID-417: Fix invalid read in GEMM
by Moritz Pflanzer
· 7 years ago
d976958
COMPMID-: Implement L2NormalizationLayer.
by Georgios Pinitas
· 7 years ago
fcdbc66
COMPMID-415: Add autoconfigure to CLCol2ImKernel
by Anthony Barbier
· 7 years ago
218cc52
COMPMID-415 Fixed bug in autoconfigure of NEPoolingLayerKernel
by Anthony Barbier
· 7 years ago
8901096
COMPMID-417: Fix CLNormalization error issue.
by Georgios Pinitas
· 7 years ago
d8e765b
COMPMID-472 : Implement Floor for CL and NEON.
by Georgios Pinitas
· 7 years ago
cfc6fe8
COMPMID-443 collapse higher dimension for CL col2im kernel
by steniu01
· 7 years ago
a261181
COMPMID-417 NEON/CL MeanStdDev bugfix using FillBorderKernel
by Giorgio Arena
· 7 years ago
c51b72f
COMPMID-355 Implement CL DirectConvolution1x1
by SiCong Li
· 7 years ago
ef4b4ae
COMPMID-438: Add support for floating point Min-Max Location layer.
by Michele Di Giorgio
· 7 years ago
f87cc7f
COMPMID-417: Port NEDirectConvolution 1x1 to QS16.
by Pablo Tello
· 7 years ago
6c92834
COMPMID-413: Add support for QS8 and QS16 CLNormalizationLayer.
by Michele Di Giorgio
· 7 years ago
d5e65c7
COMPMID-456: Add support for QS16 NEON Normalization Layer.
by Michele Di Giorgio
· 7 years ago
f9bae2e
COMPMID-417 - Bug Fix WarpPerspective kernel
by Isabella Gottardi
· 7 years ago
02dfb2c
COMPMID-457 Fix F16 NormalizationLayer CL kernel
by SiCong Li
· 7 years ago
a36ccf1
COMPMID-417: Fix CL F16 ActivationLayer
by Moritz Pflanzer
· 7 years ago
8fda1cb
COMPMID-421: Added FP16 support in BatchNormalizationLayer.
by Pablo Tello
· 7 years ago
afde732
COMPMID-421: Added FP16 support in the Neon Locally Connected Layer.
by Pablo Tello
· 7 years ago
3a62324
COMPMID-455 - Optimizing CLIm2ColKernel
by Gian Marco Iodice
· 7 years ago
ba9b3f5
COMPMID-417: Fix F16 CLSoftmaxLayer
by Moritz Pflanzer
· 7 years ago
e49e266
COMPMID-415: Use half_float library for F16
by Moritz Pflanzer
· 7 years ago
27b386c
COMPMID-355 Implement 3x3 CL direct convolution
by steniu01
· 7 years ago
1b80b6c
COMPMID-447: Support scaling factors different than 1 for QS8/QS16 NEPixelWiseMultiplication.
by Michele Di Giorgio
· 7 years ago
b49a715
COMPMID-421: Added FP16 support to Softmax.
by Pablo Tello
· 7 years ago
0d17614
COMPMID-421: Added FP16 support to the NEON Direct Convolution function.
by Pablo Tello
· 7 years ago
0c34fe2
COMPMID-421: Added FP16 support in Pooling Layer
by Pablo Tello
· 7 years ago
725788e
COMPMID-417: Allow loading of custom OpenCL library
by Moritz Pflanzer
· 7 years ago
91654c4
COMPMID-421: Added FP16 support in ActivationLayer.
by Pablo Tello
· 7 years ago
d7a5d22
COMPMID-421: Added FP16 support to Arithmetic Subtraction.
by Pablo Tello
· 7 years ago
7281834
COMPMID-446: Add support for QS8/QS16 CL Arithmetic Add/Sub
by Michele Di Giorgio
· 7 years ago
e222941
COMPMID-401: Implement FixedPointPosition conversion for NEON.
by Georgios Pinitas
· 7 years ago
bbd3d60
COMPMID-410 Port BatchNormalization to use fixed point 16
by Michalis Spyrou
· 7 years ago
172e570
COMPMID-425 Port CLBatchnormalization to support QS8/QS16
by Michalis Spyrou
· 7 years ago
579c049
COMPMID-417: Add Leaky RELU support for both NEON/CL.
by Georgios Pinitas
· 7 years ago
81f0d15
COMPMID-444: Add support for QS8/QS16 NEON Arithmetic Add/Sub/Mul.
by Michele Di Giorgio
· 7 years ago
f70256b
COMPMID-443 Collapse higher dimension for pooling layer and normalization layer
by steniu01
· 7 years ago
0d523cc
COMPMID-443 Change CLSoftMaxLayerKernel to use 3D tensor and collapse the higer dimension
by steniu01
· 7 years ago
00394ae
COMPMID-406: Port CLActivationLayer to use QS8/QS16.
by Georgios Pinitas
· 7 years ago
ac4e873
COMPMID-417: Port DepthConcatenate to QS8/QS16 for NEON/CL.
by Georgios Pinitas
· 7 years ago
df24618
COMPMID-421: Added FP16 suppot to NENormalizationLayer and NEPixelWiseMultiplication.
by Pablo Tello
· 7 years ago
d1b0ecc
COMPMID-421: Added FP16 support to Arithmetic Addition.
by Pablo Tello
· 7 years ago
9a7182e
COMPMID-443 Use 3D tensor for pixel multiply (Needed for Normalization Layer)
by Anthony Barbier
· 7 years ago
be35b0e
COMPMID-443: Collapse higher dimensions for activation layer
by Anthony Barbier
· 7 years ago
7ff47a3
COMPMID-443: Use 3D tensors for fill_border_image
by Anthony Barbier
· 7 years ago
da37e2f
COMPMID-431 Port CLDepthConvert to use 8-bit and 16-bit fixed point
by steniu01
· 7 years ago
3470247
COMPMID-417 Checking CL non uniform support at runtime.
by steniu01
· 7 years ago
9247c92
COMPMID-428: Port NESoftmaxLayer to 16-bit fixed point.
by Georgios Pinitas
· 7 years ago
0979675
COMPMID-429: Port CLSoftmaxLayer to QS16.
by Georgios Pinitas
· 7 years ago
dcdc85e
COMPMID-421: Added F16 support in FC Layer.
by Pablo Tello
· 7 years ago
7d323a6
COMPMID-440, COMPMID-441 - Port CLConvolutionLayer and CLFullyConnectedLayer to support 16 bit fixed point
by Gian Marco Iodice
· 7 years ago
ab0a77e
COMPMID-409: Add support for QS8 and QS16 CLPixelWiseMultiplication.
by Michele Di Giorgio
· 7 years ago
ccc65d4
COMPMID-427: Port NEActivationLayer in 16bit fixed point.
by Georgios Pinitas
· 7 years ago
0745a98
COMPMID-417: Fix assert in GEMMTranspose
by Moritz Pflanzer
· 7 years ago
b158fba
COMPMID-417: Fix output access window in ChannelExtract Kernels.
by Georgios Pinitas
· 7 years ago
d368df3
COMPMID-417: Auto initialize for SoftmaxLayer NEON/CL.
by Georgios Pinitas
· 7 years ago
21efeb4
COMPMID-417: DepthConvert NEON for QS8/QS16.
by Georgios Pinitas
· 7 years ago
368da83
COMPMID-420, COMPMID-414 - Port CLConvolutionLayer and CLFullyConnectedLayer to use 8 bit fixed point
by Gian Marco Iodice
· 7 years ago
b93f5de
COMPMID-417 - Fixed bug in gemm_interleave_16bit and gemm_interleave_32_bit due to the non non representable numbers in half and float
by Gian Marco Iodice
· 7 years ago
2bbd964
COMPMID-436, COMPMID-437 - Port NEConvolutionLayer & NEFullyConnectedLayer to support 16 bit fixed point
by Gian Marco Iodice
· 7 years ago
8a38369
COMPMID-434 - Port CLGEMM to support 16 bit fixed point
by Gian Marco Iodice
· 7 years ago
bdb6b0b
COMPMID-433 - Port NEGEMM to support 16 bit fixed point
by Gian Marco Iodice
· 7 years ago
ac69aa1
COMPMID-418 Add check and fix comments after preprocessor conditions
by Anthony Barbier
· 7 years ago
05da6dd
COMPMID-417: Remove val_to_string
by Moritz Pflanzer
· 7 years ago
221f381
COMPMID-421: Fixed FP16 support in Neon GEMM.
by Pablo Tello
· 7 years ago
f0dea70
COMPMID-417: Auto configuration for Add/Sub/Mul Neon/CL.
by Georgios Pinitas
· 7 years ago
Next »