| /external/tensorflow/tensorflow/docs_src/performance/ |
| datasets_performance.md | 6 `tf.data` API helps to build flexible and efficient input pipelines. This
8 high performance TensorFlow input pipelines across a variety of models and
13 * Illustrates that TensorFlow input pipelines are essentially an
20 pipelines.
37 the heavy lifting of training your model. In addition, viewing input pipelines
75 to design input pipelines that effectively utilize the CPU, optimizing each step
305 Here is a summary of the best practices for designing input pipelines:
|
| /external/deqp/external/vulkancts/modules/vulkan/multiview/ |
| vktMultiViewRenderTests.cpp | 137 vector<PipelineSp>& pipelines);
201 // pipelines
203 vector<PipelineSp> pipelines(subpassCount);
210 pipelines[subpassNdx] = (PipelineSp(new Unique<VkPipeline>(makeGraphicsPipeline(*renderPass, *pipelineLayout, static_cast<deUint32>(shaderStageParams.size()), shaderStageParams.data(), subpassNdx, vertexInputRate))));
217 draw(subpassCount, *renderPass, *frameBuffer, pipelines);
265 void MultiViewRenderTestInstance::draw (const deUint32 subpassCount,VkRenderPass renderPass, VkFramebuffer frameBuffer, vector<PipelineSp>& pipelines)
293 m_device->cmdBindPipeline(*m_cmdBuffer, VK_PIPELINE_BIND_POINT_GRAPHICS, **pipelines[subpassNdx]);
[all...] |
| /prebuilts/clang/host/darwin-x86/clang-4053586/prebuilt_include/llvm/include/llvm/Passes/ |
| PassBuilder.h | 1 //===- Parsing, selection, and construction of pass pipelines --*- C++ -*--===//
12 /// configurations, and parsing of pass pipelines.
187 /// Note that \p Level cannot be `O0` here. The pipelines produced are
202 /// Note that \p Level cannot be `O0` here. The pipelines produced are
217 /// Note that \p Level cannot be `O0` here. The pipelines produced are
231 /// Note that \p Level cannot be `O0` here. The pipelines produced are
|
| /prebuilts/clang/host/linux-x86/clang-4053586/prebuilt_include/llvm/include/llvm/Passes/ |
| PassBuilder.h | 1 //===- Parsing, selection, and construction of pass pipelines --*- C++ -*--===//
12 /// configurations, and parsing of pass pipelines.
187 /// Note that \p Level cannot be `O0` here. The pipelines produced are
202 /// Note that \p Level cannot be `O0` here. The pipelines produced are
217 /// Note that \p Level cannot be `O0` here. The pipelines produced are
231 /// Note that \p Level cannot be `O0` here. The pipelines produced are
|
| /external/ImageMagick/api_examples/ |
| magick_script | 9 # exit - not really needed here -- more for pipelines and debugging
|
| /external/adhd/cras/src/server/ |
| cras_dsp.h | 67 /* Re-reads the ini file and reloads all pipelines in the system. */
|
| /external/curl/lib/ |
| url.h | 64 /* remove the specified connection from all (possible) pipelines and related
|
| /external/mesa3d/src/mesa/state_tracker/ |
| st_atom.h | 42 * Enumeration of state tracker pipelines.
|
| /external/tensorflow/tensorflow/docs_src/programmers_guide/ |
| index.md | 11 set up data pipelines to read data sets into your TensorFlow program.
|
| /external/vboot_reference/tests/futility/ |
| test_main.sh | 32 # all progs in the pipelines should work
|
| /libcore/ojluni/src/main/java/java/util/stream/ |
| DoubleStream.java | 68 * specification of streams, stream operations, stream pipelines, and
201 * <p>For parallel stream pipelines, the action may be called at
233 * stream pipelines, it can be quite expensive on ordered parallel pipelines,
239 * speedups of {@code limit()} in parallel pipelines, if the semantics of
242 * {@code limit()} in parallel pipelines, switching to sequential execution
262 * stream pipelines, it can be quite expensive on ordered parallel pipelines,
268 * speedups of {@code skip()} in parallel pipelines, if the semantics of
271 * {@code skip()} in parallel pipelines, switching to sequential executio
[all...] |
| IntStream.java | 66 * specification of streams, stream operations, stream pipelines, and
196 * <p>For parallel stream pipelines, the action may be called at
228 * stream pipelines, it can be quite expensive on ordered parallel pipelines,
234 * speedups of {@code limit()} in parallel pipelines, if the semantics of
237 * {@code limit()} in parallel pipelines, switching to sequential execution
257 * stream pipelines, it can be quite expensive on ordered parallel pipelines,
263 * speedups of {@code skip()} in parallel pipelines, if the semantics of
266 * {@code skip()} in parallel pipelines, switching to sequential executio
[all...] |
| /external/python/cpython2/Doc/library/ |
| pipes.rst | 1 :mod:`pipes` --- Interface to shell pipelines
6 :synopsis: A Python interface to Unix shell pipelines.
|
| /external/tensorflow/tensorflow/contrib/eager/python/examples/notebooks/ |
| 3_datasets.ipynb | 12 "This notebook demonstrates the use of the [`tf.data.Dataset` API](https://www.tensorflow.org/programmers_guide/datasets) to build pipelines to feed data to your program. It covers:\n",
17 "We recommend using the `Dataset`s API for building performant, complex input pipelines from simple, re-usable pieces that will feed your model's training or evaluation loops.\n",
|
| /frameworks/native/opengl/tools/glgen/specs/gles11/ |
| GLES31.spec | 16 void glDeleteProgramPipelines ( GLsizei n, const GLuint *pipelines )
17 void glGenProgramPipelines ( GLsizei n, GLuint *pipelines )
|
| /external/llvm/lib/Target/ARM/ |
| ARMInstrVFP.td | 109 // pipelines.
128 // pipelines.
181 // VFP pipelines.
194 // VFP pipelines.
207 // VFP pipelines.
347 // VFP pipelines on A8.
369 // VFP pipelines on A8.
409 // VFP pipelines on A8.
429 // VFP pipelines on A8.
511 // VFP pipelines on A8
[all...] |
| /external/llvm/lib/Target/Mips/ |
| MipsScheduleP5600.td | 21 // ALQ Pipelines
36 // AGQ Pipelines
162 // Either ALU or AL2 Pipelines
183 // FPU Pipelines
315 // These instructions are split across multiple uops (in different pipelines)
|
| /external/gemmlowp/doc/ |
| low-precision.md | 21 ### The current flexible design with arbitrary "output pipelines".
23 See [output.md](output.md) for more details about output pipelines. This is a
32 first principles to the actual output pipelines to assemble for successful
|
| /frameworks/base/core/jni/ |
| android_opengl_GLES31.cpp | 1033 GLuint *pipelines = (GLuint *) 0; local
1077 GLuint *pipelines = (GLuint *) 0; local
1113 GLuint *pipelines = (GLuint *) 0; local
1157 GLuint *pipelines = (GLuint *) 0; local
[all...] |
| /external/guice/extensions/servlet/test/com/google/inject/servlet/ |
| MultipleServletInjectorsTest.java | 36 * This gorgeous test asserts that multiple servlet pipelines can
|
| /external/mesa3d/src/mesa/drivers/dri/i965/ |
| brw_program_cache.c | 452 brw->state.pipelines[BRW_RENDER_PIPELINE].mesa = ~0;
453 brw->state.pipelines[BRW_RENDER_PIPELINE].brw = ~0ull;
454 brw->state.pipelines[BRW_COMPUTE_PIPELINE].mesa = ~0;
455 brw->state.pipelines[BRW_COMPUTE_PIPELINE].brw = ~0ull;
|
| /external/mesa3d/src/mesa/tnl/ |
| NOTES | 49 starting point for driver pipelines. Some drivers will remove a stage
|
| /external/swiftshader/third_party/LLVM/lib/Target/Alpha/ |
| AlphaSchedule.td | 13 //Pipelines
|
| /external/tensorflow/tensorflow/contrib/data/ |
| __init__.py | 15 """Experimental API for building input pipelines.
|
| /external/tensorflow/tensorflow/docs_src/api_guides/python/ |
| input_dataset.md | 4 @{tf.data.Dataset} allows you to build complex input pipelines. See the
|
