|
DLPrimitives
|
| Cdlprim::ActivationConfig | |
| Cdlprim::AXPBY | |
| ►Cbad_cast | |
| Cdlprim::BatchNormConfig | |
| Cdlprim::core::BatchNormFwdBwd | Performs batch normalization computations over channel #1 (when #0 is batch) |
| Cdlprim::core::BiasBackwardFilter | Calculate filter |
| Cdlprim::BWBias | |
| Cdlprim::gpu::Cache | |
| Cdlprim::ConcatConfig | |
| ►Cdlprim::ConcatSliceBase | |
| Cdlprim::Context | This is main object that represent the pair of OpenCL platform and device all other objects use it |
| ►Cdlprim::core::Conv2DBase | |
| ►Cdlprim::Convolution2DBase | |
| ►Cdlprim::Convolution2DConfigBase | Convolution settings |
| Cdlprim::json::copy_ptr< T > | Smart pointer similar to std::unique_ptr but it copies underlying object on pointer copy instead of moving its ownership |
| Cdlprim::json::copy_ptr< _data > | |
| Cdlprim::TimingData::Data | |
| Cdlprim::ElementwiseConfig | |
| Cdlprim::ExecGuard | |
| Cdlprim::ExecutionContext | This class is used to pass cl::Events that the kernel should wait for and/or signal event completion |
| Cdlprim::FlattenConfig | |
| Cdlprim::gpu::GEMM | |
| Cdlprim::HardtanhConfig | |
| Cdlprim::InnerProductConfig | |
| Cdlprim::core::IPBackwardData | Perform InnerProduct/FullyConnected/Dense backward data calculations |
| Cdlprim::core::IPBackwardFilter | Perform InnerProduct/FullyConnected/Dense backward filter calcilations |
| Cdlprim::core::IPForward | Perform InnerProduct/FullyConnected/Dense forward calulations, allow fusing bias and activation into same GPU kernel |
| Cdlprim::core::IPSettings | Configuration of InnerProduct layer |
| ►Cdlprim::ModelBase | Base class used for loading non-native model formats to dlprimitives |
| Cdlprim::MSELossConfig | |
| Cdlprim::Net | Major object used for inference |
| Cdlprim::NLLLossConfig | |
| Cdlprim::json::null | Special object that is convertible to null json value |
| ►Cdlprim::Operator | Base class for backward/forward propogation calculations for internal network |
| Cdlprim::gpu::Parameter | |
| Cdlprim::core::PointwiseOperationBroadcastReduce | Perform pointwise operation with both boradcasting and reduction |
| ►Cdlprim::core::Pooling2DBackwardBase | Backward pooling computation |
| Cdlprim::core::Pooling2DForward | 2d pooling |
| ►Cdlprim::PoolingBase | |
| Cdlprim::RandomState | |
| Cdlprim::ReshapeConfig | |
| ►Cruntime_error | |
| Cdlprim::Scal | |
| Cdlprim::core::Scale | Scale tensor by factor inplace, if s==0 fills with zero so nan is not propagated of s==0 |
| Cdlprim::TimingData::Section | |
| Cdlprim::Shape | Tensor shape |
| Cdlprim::SharedResource | Resources shared by the entire network |
| Cdlprim::SliceConfig | |
| Cdlprim::core::SliceCopy | Class for copying a slice of an tensor |
| ►Cdlprim::SoftmaxBase | |
| Cdlprim::SoftmaxConfig | |
| ►Cdlprim::solvers::SolverBase | Base class for SGD based optimizers |
| Cdlprim::SqueezeConfig | |
| Cdlprim::json::string_key | This is a special object that may hold an std::string or alternatively reference to external (unowned) chunk of text |
| Cdlprim::Tensor | Central Data Contrainer - Tensor |
| Cdlprim::TensorAndGradient | Pair of tensor and its gradient for backpropogation |
| Cdlprim::TensorSpecs | Definition of Tensor without actual memory/object |
| Cdlprim::ThresholdConfig | |
| Cdlprim::TimingData | Class used for benchmarking of the model |
| Cdlprim::json::traits< T > | |
| Cdlprim::TypeTraits< T > | |
| Cdlprim::TypeTraits< double > | |
| Cdlprim::TypeTraits< float > | |
| Cdlprim::TypeTraits< int16_t > | |
| Cdlprim::TypeTraits< int32_t > | |
| Cdlprim::TypeTraits< int64_t > | |
| Cdlprim::TypeTraits< int8_t > | |
| Cdlprim::TypeTraits< uint16_t > | |
| Cdlprim::TypeTraits< uint32_t > | |
| Cdlprim::TypeTraits< uint64_t > | |
| Cdlprim::TypeTraits< uint8_t > | |
| Cdlprim::json::undefined | Special object that is convertible to undefined json value |
| Cdlprim::core::Upscale | |
| Cdlprim::json::value | This class is central representation of json objects |
1.8.11