获取请求类型。
mindie_llm::InferReqType GetReqType() const;
无。
构造一个请求对象后,为该请求对象设置请求类型,然后获取该请求的请求类型。
1 2 3 4 | mindie_llm::InferRequestId runtimeReqId(0); std::shared_ptr<mindie_llm::InferRequest> runtimeRequest(runtimeReqId); runtimeRequest->SetReqType(mindie_llm::InferReqType::REQ_STAND_INFER); mindie_llm::InferReqType type = runtimeRequest->GetReqType(); |
enum class InferReqType {
REQ_STAND_INFER = 0,
REQ_PREFILL = 1,
REQ_DECODE = 2
};