|
Mila
Deep Neural Network Library
|
Exported Modules | |
| module | Dnn.Components.Residual |
| module | Dnn.TensorDataType |
| module | Dnn.Components.QwenConfig |
| module | Compute.CpuMemoryResource |
| module | Compute.DeviceTypeTraits |
| module | Serialization.ModelArchive |
| module | Dnn.Quantization.Weight.Policies |
| module | Dnn.CompositeComponent |
| module | Compute.IExecutionContext |
| module | Serialization.Mode |
| module | Dnn.Components.Swiglu |
| module | Dnn.ITensor |
| module | Dnn.Components.AttentionOutputGate |
| module | Compute.DeviceType |
| module | Compute.ExecutionContext |
| module | Dnn.ComponentType |
| module | Compute.Device |
| module | Compute.DeviceId |
| module | Dnn.Component |
| module | Dnn.Components.CausalConv1d |
| module | Dnn.Components.RmsNorm |
| module | Dnn.Quantization.Weight.PrecisionPlan |
| module | Dnn.Components.QwenPrecisionPlan |
| module | Dnn.ActivationType |
| module | Dnn.TensorOps |
| module | Dnn.Components.GatedDeltaRule |
| module | Dnn.Components.ITransformerBlock |
| module | Dnn.TensorDataTypeTraits |
| module | Dnn.TensorTypes |
| module | Compute.GqaState |
| module | Compute.DeviceAllocation |
| module | Dnn.Components.Activation |
| module | Dnn.Components.Linear |
| module | Dnn.Tensor |
| module | Compute.Observation |
| module | Compute.ExecutionContextFactory |
Classes | |
| struct | Mila::Dnn::QwenDeltaNetBlock< TDeviceType, TPrecision, TWeightPlan >::BlockContexts |
| class | Mila::Dnn::QwenDeltaNetBlock< TDeviceType, TPrecision, TWeightPlan > |
| One Gated DeltaNet decoder block. More... | |
| struct | Mila::Dnn::QwenDeltaNetBlockWorkspace< TDeviceType, TPrecision > |
| Transformer-owned shared activation workspace for QwenDeltaNetBlock (pooling). More... | |
| struct | Mila::Dnn::QwenDeltaNetBlock< TDeviceType, TPrecision, TWeightPlan >::StateSnapshot |
| Everything this block carries between chunks – all of it, in one value. More... | |
Functions | |
| template<DeviceType TDeviceType, TensorDataType TPrecision> requires PrecisionSupportedOnDevice<TPrecision, TDeviceType> | |
| QwenDeltaNetBlockWorkspace< TDeviceType, TPrecision > | Mila::Dnn::makeQwenDeltaNetBlockWorkspace (const QwenConfig &config, DeviceId device, dim_t B, dim_t prefill_chunk, const std::string &name_prefix) |
| Allocate the shared DeltaNet-block workspace. | |
| std::vector< dim_t > | Mila::Dnn::qwenDeltaNetWorkspaceSlotWidths (const QwenConfig &config) |
| The width of each workspace slot, one entry per [B, chunk, width] allocation. | |
Files | |
| file | Mila/Src/Dnn/Components/Transformers/Qwen/Qwen.DeltaNetBlock.ixx |
| Qwen 3.8 Gated DeltaNet decoder block (inference: prefill + decode). | |