| COMPMID-2491: (Nightly) Bug in NEON LSTMLayerQuantized |
|
More...
|
over 7 years ago
|
| COMPMID-2055: Fusion of ConvolutionLayer with BatchNormalization at graph only for CL |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Additional warnings for import_memory() |
|
More...
|
over 7 years ago
|
| Update copyright years |
|
More...
|
over 7 years ago
|
| COMPMID-2484: Support depthwise convolution for depth multipliers > 3 |
|
More...
|
over 7 years ago
|
| COMPMID-1850: Port TFLite mobilenetSSD as a graph example |
|
More...
|
over 7 years ago
|
| COMPMID 1528: Fixed compilation error when opencl=0 |
|
More...
|
over 7 years ago
|
| COMPMID-2483: Fixed tuner in benchmark tests |
|
More...
|
over 7 years ago
|
| COMPMID-1816: Use parallel reduction on 0 axis in CL ARG_MIN/ARG_MAX |
|
More...
|
over 7 years ago
|
| Use .inst directives instead of .int directives. |
|
More...
|
over 7 years ago
|
| COMPMID-2466: Improved ConvLayer documentation. |
|
More...
|
over 7 years ago
|
| COMPMID-1979: Fuse Activation Function in CLGEMM - part 3 |
|
More...
|
over 7 years ago
|
| COMPMID-2307: QUANTIZED_16BIT_LSTM operator for CL |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix validation for quantized NEDepthwiseConvolutionLayer |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Change GPUTarget regex |
|
More...
|
over 7 years ago
|
| COMPMID-1979: Fuse Activation Function in CLGEMM - part 2 |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix InPlaceMutator condition and add SaveNumpyAccessor |
|
More...
|
over 7 years ago
|
| COMPMID-1849: Add DetectorPostProcess operator Part1 - Rework of CPPNonMaximumSuppression |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Add missing macro comment in GLES scale |
|
More...
|
over 7 years ago
|
| COMPMID-2236: Move assembly implementation interfaces to src folder |
|
More...
|
over 7 years ago
|
| COMPMID-2179 New generic depthwise convolution for NEON F32 NHWC |
|
More...
|
over 7 years ago
|
| COMPMID-2468: (Nightly) Bug in CL QSYMM16 |
|
More...
|
over 7 years ago
|
| COMPMID-2443: CL example use program cache by default. |
|
More...
|
over 7 years ago
|
| COMPMID-2414 Add add_2D_tensor_argument_if for OpenCL |
|
More...
|
over 7 years ago
|
| Revert "COMPMID-2425: Fix memory leak reported by coverity" |
|
More...
|
over 7 years ago
|
| COMPMID-2421: Add 5x5 DepthwiseConv2d in the optimized path for NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2458: Initialize uninitialized variables |
|
More...
|
over 7 years ago
|
| COMPMID-2456: NEDeconvolutionLayer.cpp, NHWC is not supported |
|
More...
|
over 7 years ago
|
| COMPMID-2463: Add deprecation macro to deprecate interfaces |
|
More...
|
over 7 years ago
|
| COMPMID-2447: Align TFlite nearest neighbor NE/CL functions with ACL |
|
More...
|
over 7 years ago
|
