Mila
Deep Neural Network Library
Loading...
Searching...
No Matches
Mila::Dnn::LlamaModel< TDeviceType, TPrecision > Member List

This is the complete list of members for Mila::Dnn::LlamaModel< TDeviceType, TPrecision >, including all inherited members.

awaitSampledToken()Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineprotected
Base typedef (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >
enqueueSampleNext(const TensorType &logits, TokenTensor &token_out, const SamplingParams &params)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineprotected
finalLogitSoftcap() const noexceptMila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineprotectedvirtual
fromPretrained(const std::filesystem::path &path, const LlamaModelConfig &model_config, DeviceId device_id=DeviceId{ TDeviceType, 0 })Mila::Dnn::LlamaModel< TDeviceType, TPrecision >inlinestatic
generate(std::span< const int32_t > prompt_tokens, const std::function< void(int32_t)> &on_token, const GenerateParams &params={}, std::stop_token stop={})Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inline
getConfig() const noexcept (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >inline
getDeviceId() const noexceptMila::Dnn::Model< TDeviceType, TPrecision >inline
getLanguageNetwork() noexcept (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineprotected
getLanguageNetwork() const noexcept (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineprotected
getMemoryStats() constMila::Dnn::Model< TDeviceType, TPrecision >inline
getRequiredMemory(const std::filesystem::path &path, const LlamaModelConfig &model_config, DeviceId device_id=DeviceId{ TDeviceType, 0 })Mila::Dnn::LlamaModel< TDeviceType, TPrecision >inlinestatic
getRuntimeMode() const noexceptMila::Dnn::Model< TDeviceType, TPrecision >inline
getScratchHighWaterBytes() constMila::Dnn::Model< TDeviceType, TPrecision >inline
isEval() const noexceptMila::Dnn::Model< TDeviceType, TPrecision >inline
isInferenceMode() const noexceptMila::Dnn::Model< TDeviceType, TPrecision >inline
isTrainingMode() const noexceptMila::Dnn::Model< TDeviceType, TPrecision >inline
LanguageModel(const LanguageModel &)=delete (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >
LanguageModel(LanguageModel &&)=default (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >
LanguageModel(std::unique_ptr< LanguageNetwork< TDeviceType, TPrecision > > network, RuntimeMode runtime_mode, Serialization::PretrainedMetadata source_metadata={}, WeightQuantization weight_quantization=WeightQuantization::None)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineexplicitprotected
LlamaKvPolicy typedefMila::Dnn::LlamaModel< TDeviceType, TPrecision >
LlamaModel(const LlamaModel &)=delete (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
LlamaModel(LlamaModel &&)=default (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
maxSequenceLength() const noexcept overrideMila::Dnn::LlamaModel< TDeviceType, TPrecision >inlineprotectedvirtual
Model(const Model &)=delete (defined in Mila::Dnn::Model< TDeviceType, TPrecision >)Mila::Dnn::Model< TDeviceType, TPrecision >
Model(Model &&)=default (defined in Mila::Dnn::Model< TDeviceType, TPrecision >)Mila::Dnn::Model< TDeviceType, TPrecision >
Model(std::unique_ptr< NetworkType > network, RuntimeMode runtime_mode)Mila::Dnn::Model< TDeviceType, TPrecision >inlineexplicitprotected
ModelBase typedef (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
MR typedef (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
network_Mila::Dnn::Model< TDeviceType, TPrecision >protected
NetworkType typedef (defined in Mila::Dnn::Model< TDeviceType, TPrecision >)Mila::Dnn::Model< TDeviceType, TPrecision >
onGenerating(std::span< const int32_t > prompt_tokens, const std::function< void(int32_t)> &on_token, const GenerateParams &params, std::stop_token stop) overrideMila::Dnn::LlamaModel< TDeviceType, TPrecision >inlineprotectedvirtual
onTraining() overrideMila::Dnn::LlamaModel< TDeviceType, TPrecision >inlineprotectedvirtual
operator=(const LlamaModel &)=delete (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
operator=(LlamaModel &&)=default (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
operator=(const LanguageModel &)=delete (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >
operator=(LanguageModel &&)=default (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >
operator=(const Model &)=delete (defined in Mila::Dnn::Model< TDeviceType, TPrecision >)Mila::Dnn::Model< TDeviceType, TPrecision >
operator=(Model &&)=default (defined in Mila::Dnn::Model< TDeviceType, TPrecision >)Mila::Dnn::Model< TDeviceType, TPrecision >
profilePrefill(const std::vector< int32_t > &token_ids) (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >inline
sampleNext(const TensorType &logits, TokenTensor &token_out, const SamplingParams &params)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inlineprotected
savePretrained(const std::filesystem::path &path) constMila::Dnn::LanguageModel< TDeviceType, TPrecision >inline
seedSampler(uint64_t seed)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >inline
setEval(bool eval)Mila::Dnn::Model< TDeviceType, TPrecision >inline
source_metadata_Mila::Dnn::LanguageModel< TDeviceType, TPrecision >protected
StagingMR typedef (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
TensorType typedef (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
TokenIndexType typedef (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
TokenTensor typedef (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >protected
toString() const overrideMila::Dnn::LlamaModel< TDeviceType, TPrecision >inlinevirtual
train()Mila::Dnn::Model< TDeviceType, TPrecision >inline
vocabSize() const noexcept overrideMila::Dnn::LlamaModel< TDeviceType, TPrecision >inlineprotectedvirtual
weight_quantization_Mila::Dnn::LanguageModel< TDeviceType, TPrecision >protected
~LanguageModel()=default (defined in Mila::Dnn::LanguageModel< TDeviceType, TPrecision >)Mila::Dnn::LanguageModel< TDeviceType, TPrecision >virtual
~LlamaModel()=default (defined in Mila::Dnn::LlamaModel< TDeviceType, TPrecision >)Mila::Dnn::LlamaModel< TDeviceType, TPrecision >
~Model()=default (defined in Mila::Dnn::Model< TDeviceType, TPrecision >)Mila::Dnn::Model< TDeviceType, TPrecision >virtual