Hi3559AV100 NNIE開發（4）mobilefacenet.cfg參數配置挖坑解決與SVP_NNIE_Cnn實現分析

2021 年 3 月 11 日
筆記
hi3559, Hi35xx項目開發, mobilefacenet.cfg, NNIE_SVP_CNN

　　前面隨筆給出了NNIE開發的基本知識，下面幾篇隨筆將着重於Mobilefacenet NNIE開發，實現mobilefacenet.wk的chip版本，並在Hi3559AV100上實現mobilefacenet網絡功能，外接USB攝像頭通過MPP平台輸出至VO HDMI顯示結果。下文是Hi3559AV100 NNIE開發（4）mobilefacenet.cfg參數配置挖坑解決與SVP_NNIE_Cnn實現分析，目前項目需要對mobilefacenet網絡進行.wk的開發，下面給出在.wk生成過程中遇到的坑與解決方式，並給出SVP_NNIE_Cnn整體實現的各個step分析，為後面在板載上實現mobilefacenet網絡打下基礎。

1、mobilefacenet.cfg參數配置挖坑解決

　　CNN_convert_bin_and_print_featuremap.py和Get Caffe Output這裡的預處理方式都是先乘以【data_scale】，再減均值【mean_file】，而在量化生成 .mk 文件時卻是先減均值再乘以scale的。

　　給出預處理這一個環節對輸入數據data的處理方式：

1     data = inputs
2     if norm_type == '4' or norm_type == '5':
3        data = data * float(data_scale)

　　data是uint8類型的array，是先乘以了【data_scale】的，也就是說和NNIE 生成wk中的操作順序是不一致的，對於mobilefacenet.cfg網絡輸入數據預處理方法時，當norm_type = 5時，輸入數據減通道均值後再乘以 data_scale，如下所示：

　　所在在實際操作中，需要對均值文件進行處理，轉換方式如下：

(data – 128.0) * 0.0078125 <==> data * 0.0078125 – 1

　　因此這裡需要做的修改就是需要將【mean_file】pixel_mean_compare.txt修設置為1.0：

　　最終生成mobilefacenet.wk，結果如下所示，具體的測試需要下一步進行。

 1 begin parameter compressing....
 2 
 3 end parameter compressing
 4 
 5 begin compress index generating....
 6 
 7 end compress index generating
 8 
 9 begin binary code generating....
10 
11 .........................................................................................................................
12 ....................................................................end binary code generating
13 
14 begin quant files writing....
15 
16 end quant files writing
17 
18 .
19 ===============D:\Hi3559_NNIE\3559\mobileface\mobileface.cfg Successfully!===============
20 
21 End [RuyiStudio Wk NNIE Mapper] [D:\Hi3559_NNIE\3559\mobileface\mobileface.cfg] mobileface

2、SVP_NNIE_Cnn實現分析

　　下面給出SAMPLE_SVP_NNIE_Cnn函數的執行過程，主要分為下面八個步驟：

 1     HI_CHAR *pcSrcFile = "./data/nnie_image/y/0_28x28.y";
 2     HI_CHAR *pcModelName = "./data/nnie_model/classification/inst_mnist_cycle.wk";
 3 
 4 
 5     /*Set configuration parameter  */
 6     stNnieCfg.pszPic= pcSrcFile;
 7     stNnieCfg.u32MaxInputNum = u32PicNum; //max input image num in each batch
 8     stNnieCfg.u32MaxRoiNum = 0;
 9     stNnieCfg.aenNnieCoreId[0] = SVP_NNIE_ID_0;//set NNIE core
10     s_stCnnSoftwareParam.u32TopN = 5;
11 
12 
13 
14     /*Sys init ---step1*/
15     SAMPLE_COMM_SVP_CheckSysInit();
16 
17     /*CNN Load model  ------step2*/
18     s32Ret = SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel);
19 
20 
21     /*CNN parameter initialization   -------step3*/
22     /*Cnn software parameters are set in SAMPLE_SVP_NNIE_Cnn_SoftwareParaInit,
23      if user has changed net struct, please make sure the parameter settings in
24      SAMPLE_SVP_NNIE_Cnn_SoftwareParaInit function are correct*/
25     s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
26     s32Ret = SAMPLE_SVP_NNIE_Cnn_ParamInit(&stNnieCfg,&s_stCnnNnieParam,&s_stCnnSoftwareParam);
27 
28 
29     /*record tskBuf  -------step4*/
30     s32Ret = HI_MPI_SVP_NNIE_AddTskBuf(&(s_stCnnNnieParam.astForwardCtrl[0].stTskBuf));
31 
32 
33     /*Fill src data -------step5*/
34     SAMPLE_SVP_TRACE_INFO("Cnn start!\n");
35     stInputDataIdx.u32SegIdx = 0;
36     stInputDataIdx.u32NodeIdx = 0;
37     s32Ret = SAMPLE_SVP_NNIE_FillSrcData(&stNnieCfg,&s_stCnnNnieParam,&stInputDataIdx);
38 
39 
40     /*NNIE process(process the 0-th segment)   -------step6*/
41     stProcSegIdx.u32SegIdx = 0;
42     s32Ret = SAMPLE_SVP_NNIE_Forward(&s_stCnnNnieParam,&stInputDataIdx,&stProcSegIdx,HI_TRUE);
43 
44 
45 
46     /*Software process     --------step7*/
47     /*if user has changed net struct, please make sure SAMPLE_SVP_NNIE_Cnn_GetTopN
48      function's input datas are correct*/
49     s32Ret = SAMPLE_SVP_NNIE_Cnn_GetTopN(&s_stCnnNnieParam,&s_stCnnSoftwareParam);
50 
51 
52 
53     /*Print result      --------step8*/
54     SAMPLE_SVP_TRACE_INFO("Cnn result:\n");
55     s32Ret = SAMPLE_SVP_NNIE_Cnn_PrintResult(&(s_stCnnSoftwareParam.stGetTopN),
56              s_stCnnSoftwareParam.u32TopN);

　　（1）step1為SAMPLE_COMM_SVP_CheckSysInit()，完成的是MPP系統的初始化，主要實現的是Sys_Init和VB_Init，實現MPP內存池的配置，具體實現如下：

 1 HI_VOID SAMPLE_COMM_SVP_CheckSysInit(HI_VOID)
 2 {
 3     ..............
 4     SAMPLE_COMM_SVP_SysInit()
 5     {
 6         //省略了部分過程，列出實現關鍵函數
 7         HI_MPI_SYS_Exit();
 8         HI_MPI_VB_Exit();
 9 
10         memset(&struVbConf,0,sizeof(VB_CONFIG_S));
11 
12         struVbConf.u32MaxPoolCnt             = 2;
13         struVbConf.astCommPool[1].u64BlkSize = 768*576*2;
14         struVbConf.astCommPool[1].u32BlkCnt  = 1;
15 
16         s32Ret = HI_MPI_VB_SetConfig((const VB_CONFIG_S *)&struVbConf);  //設置MPP視頻緩存池屬性
17 
18 
19         s32Ret = HI_MPI_VB_Init();  //初始化MPP緩存池
20 
21 
22         s32Ret = HI_MPI_SYS_Init(); //初始化MPP系統
23 
24     }
25 
26     .............  
27 }

　　（2）step2為SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel)從用戶事先加載到 buf 中的模型中解析出網絡模型，其函數實現較為複雜，具體的函數參數解析和函數運行過程已經在前面隨筆給出了，需要的話，可以參考隨筆：

　　Hi3559AV100 NNIE開發（1）-RFCN(.wk)LoadModel函數參數解析（//www.cnblogs.com/iFrank/p/14500648.html）

　　Hi3559AV100 NNIE開發（2）-RFCN(.wk)LoadModel及NNIE Init函數運行過程分析（//www.cnblogs.com/iFrank/p/14503482.html）

　　（3）step3為SAMPLE_SVP_NNIE_Cnn_ParamInit，首先給出調用與定義，便於分析：

 1 /*Set configuration parameter*/
 2 stNnieCfg.pszPic= pcSrcFile;
 3 stNnieCfg.u32MaxInputNum = u32PicNum; //max input image num in each batch
 4 stNnieCfg.u32MaxRoiNum = 0;
 5 stNnieCfg.aenNnieCoreId[0] = SVP_NNIE_ID_0;//set NNIE core
 6 
 7 s_stCnnSoftwareParam.u32TopN = 5;
 8 
 9 s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
10 s32Ret = SAMPLE_SVP_NNIE_Cnn_ParamInit(&stNnieCfg,
11                                        &s_stCnnNnieParam,
12                                        &s_stCnnSoftwareParam);
13                                        
14                                        
15                                        
16 static HI_S32 SAMPLE_SVP_NNIE_Cnn_ParamInit(SAMPLE_SVP_NNIE_CFG_S* pstNnieCfg,
17                                             SAMPLE_SVP_NNIE_PARAM_S *pstCnnPara, 
18                                 SAMPLE_SVP_NNIE_CNN_SOFTWARE_PARAM_S* pstCnnSoftWarePara)
19 {
20     ........
21     
22     /*init hardware para*/
23     s32Ret = SAMPLE_COMM_SVP_NNIE_ParamInit(pstNnieCfg,
24                                             pstCnnPara);
25 
26 
27     /*init software para*/
28     if(pstCnnSoftWarePara!=NULL)
29     {
30         s32Ret = SAMPLE_SVP_NNIE_Cnn_SoftwareParaInit(pstNnieCfg,
31                                                       pstCnnPara,
32                                                       pstCnnSoftWarePara);
33                                "Error(%#x),SAMPLE_SVP_NNIE_Cnn_SoftwareParaInit failed!\n",s32Ret);
34     }
35     
36     ........
37 }

　　其中SAMPLE_COMM_SVP_NNIE_ParamInit函數及參數分析可見之前隨筆:

　　Hi3559AV100 NNIE開發（2）-RFCN(.wk)LoadModel及NNIE Init函數運行過程分析（//www.cnblogs.com/iFrank/p/14503482.html），之前的隨筆介紹的很詳細，這裡就不在贅述了。

　　對SAMPLE_SVP_NNIE_Cnn_SoftwareParaInit函數，首先給出定義：

 1 static HI_S32 SAMPLE_SVP_NNIE_Cnn_SoftwareParaInit(
 2     SAMPLE_SVP_NNIE_CFG_S* pstNnieCfg,
 3     SAMPLE_SVP_NNIE_PARAM_S *pstCnnPara, 
       SAMPLE_SVP_NNIE_CNN_SOFTWARE_PARAM_S* pstCnnSoftWarePara)
 4 {
 5     HI_U32 u32GetTopNMemSize = 0;
 6     HI_U32 u32GetTopNAssistBufSize = 0;
 7     HI_U32 u32GetTopNPerFrameSize = 0;
 8     HI_U32 u32TotalSize = 0;
 9     HI_U32 u32ClassNum = pstCnnPara->pstModel->astSeg[0].astDstNode[0].unShape.stWhc.u32Width;
10     HI_U64 u64PhyAddr = 0;
11     HI_U8* pu8VirAddr = NULL;
12     HI_S32 s32Ret = HI_SUCCESS;
13 
14     /*get mem size*/
15     u32GetTopNPerFrameSize = pstCnnSoftWarePara->u32TopN*sizeof(SAMPLE_SVP_NNIE_CNN_GETTOPN_UNIT_S);
16     u32GetTopNMemSize = SAMPLE_SVP_NNIE_ALIGN16(u32GetTopNPerFrameSize)*pstNnieCfg->u32MaxInputNum;
17     u32GetTopNAssistBufSize = u32ClassNum*sizeof(SAMPLE_SVP_NNIE_CNN_GETTOPN_UNIT_S);
18     u32TotalSize = u32GetTopNMemSize+u32GetTopNAssistBufSize;
19 
20     /*malloc mem*/
21     s32Ret = SAMPLE_COMM_SVP_MallocMem("SAMPLE_CNN_INIT",NULL,(HI_U64*)&u64PhyAddr,
22         (void**)&pu8VirAddr,u32TotalSize);
23     SAMPLE_SVP_CHECK_EXPR_RET(HI_SUCCESS != s32Ret,s32Ret,SAMPLE_SVP_ERR_LEVEL_ERROR,
24         "Error,Malloc memory failed!\n");
25     memset(pu8VirAddr, 0, u32TotalSize);
26 
27     /*init GetTopn */
28     pstCnnSoftWarePara->stGetTopN.u32Num= pstNnieCfg->u32MaxInputNum;
29     pstCnnSoftWarePara->stGetTopN.unShape.stWhc.u32Chn = 1;
30     pstCnnSoftWarePara->stGetTopN.unShape.stWhc.u32Height = 1;
31     pstCnnSoftWarePara->stGetTopN.unShape.stWhc.u32Width = u32GetTopNPerFrameSize/sizeof(HI_U32);
32     pstCnnSoftWarePara->stGetTopN.u32Stride = SAMPLE_SVP_NNIE_ALIGN16(u32GetTopNPerFrameSize);
33     pstCnnSoftWarePara->stGetTopN.u64PhyAddr = u64PhyAddr;
34     pstCnnSoftWarePara->stGetTopN.u64VirAddr = (HI_U64)pu8VirAddr;
35 
36     /*init AssistBuf */
37     pstCnnSoftWarePara->stAssistBuf.u32Size = u32GetTopNAssistBufSize;
38     pstCnnSoftWarePara->stAssistBuf.u64PhyAddr = u64PhyAddr+u32GetTopNMemSize;
39     pstCnnSoftWarePara->stAssistBuf.u64VirAddr = (HI_U64)pu8VirAddr+u32GetTopNMemSize;
40 
41     return s32Ret;
42 }

　　函數體內最主要功能是實現s_stCnnSoftwareParam參數的賦值，包含大量賦值語句，其中s_stCnnSoftwareParam結構體各個元素賦值的意義等需要的時候再進行研討，此外函數還實現在用戶態分配 MMZ 內存。通過對兩個函數的分析，step3 SAMPLE_SVP_NNIE_Cnn_ParamInit()完成。

　　（4）step4為HI_MPI_SVP_NNIE_AddTskBuf，為了記錄 TskBuf 地址信息，其作用和注意事項：

　　①記錄 TskBuf 地址信息，用於減少內核態內存映射次數，提升效率；

　　②TskBuf 地址信息的記錄是通過鏈表進行管理，鏈表長度默認值為 32，鏈表長度可通過模塊參數 nnie_max_tskbuf_num 進行配；

　　③若沒調用 HI_MPI_SVP_NNIE_AddTskBuf 預先把 TskBuf 地址信息記錄到系統，那麼之後調用 Forward/ForwardWithBbox 每次都會 Map/Unmap 操作 TskBuf 內核態虛擬地址，效率會比較低。　　

　　給出函數調用和定義:

1 /*SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel);*/     
2 s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
3     
4 s32Ret = HI_MPI_SVP_NNIE_AddTskBuf(&(s_stCnnNnieParam.astForwardCtrl[0].stTskBuf));
5 
6 
7 //定義
8 HI_S32 HI_MPI_SVP_NNIE_AddTskBuf(const SVP_MEM_INFO_S* pstTskBuf);

　　（5）step5為SAMPLE_SVP_NNIE_FillSrcData，實現src數據的填充，此函數十分關鍵，對所給圖像數據：./data/nnie_image/y/0_28x28.y進行處理，為了更好的分析數據處理函數，首先給出函數調用信息：

 1     stNnieCfg.pszPic= pcSrcFile;
 2     stNnieCfg.u32MaxInputNum = u32PicNum; //max input image num in each batch
 3     stNnieCfg.u32MaxRoiNum = 0;
 4     stNnieCfg.aenNnieCoreId[0] = SVP_NNIE_ID_0;//set NNIE core
 5 
 6 /*SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel);*/
 7 s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
 8 
 9     stInputDataIdx.u32SegIdx = 0;
10     stInputDataIdx.u32NodeIdx = 0;
11     s32Ret = SAMPLE_SVP_NNIE_FillSrcData(&stNnieCfg,
　　　　　　　　　　　　　　　　　　　　　　　　　　&s_stCnnNnieParam,
　　　　　　　　　　　　　　　　　　　　　　　　　　&stInputDataIdx);

　　為了更加清楚函數功能，先給出函數定義，方便後面分析（忽略一些次要信息）：

  1 static HI_S32 SAMPLE_SVP_NNIE_FillSrcData(SAMPLE_SVP_NNIE_CFG_S* pstNnieCfg,
  2                                         SAMPLE_SVP_NNIE_PARAM_S *pstNnieParam, 
  3                                         SAMPLE_SVP_NNIE_INPUT_DATA_INDEX_S* pstInputDataIdx)

　　總的來說，函數實現以下功能：

　　①open file fopen(pstNnieCfg->pszPic,”rb”);

 1 //定義文件名
 2 HI_CHAR *pcSrcFile = "./data/nnie_image/y/0_28x28.y";
 3 
 4 stNnieCfg.pszPic= pcSrcFile;
 5 
 6 //函數定義
 7 HI_S32 SAMPLE_SVP_NNIE_FillSrcData(SAMPLE_SVP_NNIE_CFG_S* pstNnieCfg,
 8                                    SAMPLE_SVP_NNIE_PARAM_S *pstNnieParam, 
 9                                    SAMPLE_SVP_NNIE_INPUT_DATA_INDEX_S* pstInputDataIdx)
10 //函數調用
11 SAMPLE_SVP_NNIE_FillSrcData(&stNnieCfg,
12                             &s_stCnnNnieParam,
13                             &stInputDataIdx);
14 
15 fp = fopen(pstNnieCfg->pszPic,"rb");

　　②為後面fread讀取數據量確定u32VarSize大小：

 1     /*get data size   s32Ret = fread(pu8PicAddr,u32Dim*u32VarSize,1,fp);*/
 2     if(SVP_BLOB_TYPE_U8 <= pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].enType &&
 3         SVP_BLOB_TYPE_YVU422SP >= pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].enType)
 4     {
 5         u32VarSize = sizeof(HI_U8);
 6     }
 7     else
 8     {
 9         u32VarSize = sizeof(HI_U32);
10     }

　　③隨即通過pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].enType參數（參數找定義，應該是與輸入模型.wk的模型參數有關，後面可以直接通過printf進行打印輸出，看結果是啥）進行if-lese分支選擇，之後通過fread對fp文件指針讀取數據，確定數據內存地址，並刷新 cache 里的內容到內存並且使 cache 里的內容無效，最後fclose(fp)。

　　先給出enType參數類型：

　　代碼實現：

 1     /*fill src data*/
 2     if(SVP_BLOB_TYPE_SEQ_S32 == pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].enType)
 3     {
 4         u32Dim = pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].unShape.stSeq.u32Dim;
 5         u32Stride = pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Stride;
 6         pu32StepAddr = (HI_U32*)(pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].unShape.stSeq.u64VirAddrStep);
 7         pu8PicAddr = (HI_U8*)(pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u64VirAddr);
 8         for(n = 0; n < pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Num; n++)
 9         {
10             for(i = 0;i < *(pu32StepAddr+n); i++)
11             {
12                 s32Ret = fread(pu8PicAddr,u32Dim*u32VarSize,1,fp);
13                 SAMPLE_SVP_CHECK_EXPR_GOTO(1 != s32Ret,FAIL,SAMPLE_SVP_ERR_LEVEL_ERROR,"Error,Read image file failed!\n");
14                 pu8PicAddr += u32Stride;
15             }
16             u32TotalStepNum += *(pu32StepAddr+n);
17         }
18         SAMPLE_COMM_SVP_FlushCache(pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u64PhyAddr,
19             (HI_VOID *) pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u64VirAddr,
20             u32TotalStepNum*u32Stride);
21     }
22     else
23     {
24         u32Height = pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].unShape.stWhc.u32Height;
25         u32Width = pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].unShape.stWhc.u32Width;
26         u32Chn = pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].unShape.stWhc.u32Chn;
27         u32Stride = pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Stride;
28         pu8PicAddr = (HI_U8*)(pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u64VirAddr);
29         if(SVP_BLOB_TYPE_YVU420SP== pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].enType)
30         {
31             for(n = 0; n < pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Num; n++)
32             {
33                 for(i = 0; i < u32Chn*u32Height/2; i++)
34                 {
35                     s32Ret = fread(pu8PicAddr,u32Width*u32VarSize,1,fp);
36                     SAMPLE_SVP_CHECK_EXPR_GOTO(1 != s32Ret,FAIL,SAMPLE_SVP_ERR_LEVEL_ERROR,"Error,Read image file failed!\n");
37                     pu8PicAddr += u32Stride;
38                 }
39             }
40         }
41         else if(SVP_BLOB_TYPE_YVU422SP== pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].enType)
42         {
43             for(n = 0; n < pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Num; n++)
44             {
45                 for(i = 0; i < u32Height*2; i++)
46                 {
47                     s32Ret = fread(pu8PicAddr,u32Width*u32VarSize,1,fp);
48                     SAMPLE_SVP_CHECK_EXPR_GOTO(1 != s32Ret,FAIL,SAMPLE_SVP_ERR_LEVEL_ERROR,"Error,Read image file failed!\n");
49                     pu8PicAddr += u32Stride;
50                 }
51             }
52         }
53         else
54         {
55             for(n = 0; n < pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Num; n++)
56             {
57                 for(i = 0;i < u32Chn; i++)
58                 {
59                     for(j = 0; j < u32Height; j++)
60                     {
61                         s32Ret = fread(pu8PicAddr,u32Width*u32VarSize,1,fp);
62                         SAMPLE_SVP_CHECK_EXPR_GOTO(1 != s32Ret,FAIL,SAMPLE_SVP_ERR_LEVEL_ERROR,"Error,Read image file failed!\n");
63                         pu8PicAddr += u32Stride;
64                     }
65                 }
66             }
67         }
68         SAMPLE_COMM_SVP_FlushCache(pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u64PhyAddr,
69             (HI_VOID *) pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u64VirAddr,
70             pstNnieParam->astSegData[u32SegIdx].astSrc[u32NodeIdx].u32Num*u32Chn*u32Height*u32Stride);
71     }
72 
73     fclose(fp);

　　（6）step6為SAMPLE_SVP_NNIE_Forward實現NNIE process，便於分析先給出函數的調用及參數的定義：

 1     s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
 2     /* SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel); */
 3     
 4     stInputDataIdx.u32SegIdx = 0;
 5     stInputDataIdx.u32NodeIdx = 0;
 6     
 7     stProcSegIdx.u32SegIdx = 0;
 8     
 9     s32Ret = SAMPLE_SVP_NNIE_Forward(&s_stCnnNnieParam,
10                                      &stInputDataIdx,
11                                      &stProcSegIdx,
12                                      HI_TRUE);
13     
14     static HI_S32 SAMPLE_SVP_NNIE_Forward(
15                         SAMPLE_SVP_NNIE_PARAM_S *pstNnieParam,
16                         SAMPLE_SVP_NNIE_INPUT_DATA_INDEX_S* pstInputDataIdx,
17                         SAMPLE_SVP_NNIE_PROCESS_SEG_INDEX_S* pstProcSegIdx,
18                         HI_BOOL bInstant)

　　SAMPLE_SVP_NNIE_Forward中①SAMPLE_COMM_SVP_FlushCache函數主要實現將內存數據刷新到內存中；②HI_MPI_SVP_NNIE_Forward函數同時對輸入樣本(s)進行CNN預測，對對應樣本(s)進行輸出響應；③HI_MPI_SVP_NNIE_Query函數用於查詢nnie上運行函數的狀態，在阻塞模式下，系統等待，直到被查詢的函數被調用；在非阻塞模式下，查詢當前狀態，不做任何操作。

　　（7）step7為SAMPLE_SVP_NNIE_Cnn_GetTopN實現軟件過程，給出函數調用與參數細節：

 1     s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
 2     /* SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel); */
 3                         
 4     s_stCnnSoftwareParam.u32TopN = 5;
 5     SAMPLE_SVP_NNIE_Cnn_ParamInit(&stNnieCfg, //通過此函數對s_stCnnSoftwareParam進行了賦值操作
 6                                   &s_stCnnNnieParam,
 7                                   &s_stCnnSoftwareParam); 
 8     
 9     s32Ret = SAMPLE_SVP_NNIE_Cnn_GetTopN(&s_stCnnNnieParam,
10                                          &s_stCnnSoftwareParam);            
11 
12     HI_S32   SAMPLE_SVP_NNIE_Cnn_GetTopN(SAMPLE_SVP_NNIE_PARAM_S*pstNnieParam,
13                                         SAMPLE_SVP_NNIE_CNN_SOFTWARE_PARAM_S* pstSoftwareParam)

　　此函數目前基本不修改，函數內部具體實現目前暫不說明，只需注意一點如果改變了網絡結構，請確保SAMPLE_SVP_NNIE_Cnn_GetTopN

函數的輸入數據正確。

　　（8）step8為SAMPLE_SVP_NNIE_Cnn_PrintResult打印blob參數值，給出函數調用與參數細節：

 1     s_stCnnNnieParam.pstModel = &s_stCnnModel.stModel;
 2     /* SAMPLE_COMM_SVP_NNIE_LoadModel(pcModelName,&s_stCnnModel); */
 3                         
 4     s_stCnnSoftwareParam.u32TopN = 5;
 5     SAMPLE_SVP_NNIE_Cnn_ParamInit(&stNnieCfg, //通過此函數對s_stCnnSoftwareParam進行了賦值操作
 6                                   &s_stCnnNnieParam,
 7                                   &s_stCnnSoftwareParam); 
 8     
 9     s32Ret = SAMPLE_SVP_NNIE_Cnn_PrintResult(&(s_stCnnSoftwareParam.stGetTopN),
10                                              s_stCnnSoftwareParam.u32TopN);
11                                              
12       HI_S32 SAMPLE_SVP_NNIE_Cnn_PrintResult(SVP_BLOB_S *pstGetTopN,
13                                              HI_U32 u32TopN)

　　有什麼問題，大家可以提出來，一起討論，後面將給出mobilefacenet的NNIE實現。

Tags: hi3559 Hi35xx項目開發 mobilefacenet.cfg NNIE_SVP_CNN