# 執行模型

Tab Neural Processing Engine
Tab AI Engine Direct

- *class* tabincludedirective

    - ## 使用神經處理引擎部署模型

可透過支援 SNPE 的應用程式 (使用 SNPE C/C++ API 編寫的應用程式) 部署模型 DLC (量化或非量化)。SNPE 提供 API 以載入 DLC、選擇執行階段以執行模型，以及進行推論等。

此外，SNPE 提供了預建的工具 [snpe-net-run](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-2/tools.html#snpe-net-run) ，此工具可載入任意模型 DLC 並執行推論。

- **模型檔案**：  SNPE 轉換工具或 `snpe-dlc-graph-prepare` 工具 (如果在 HTP 上執行) 產生的 DLC 模型檔案。
- **輸入清單**：文字檔案，例如在量化過程中使用的 `input_list.txt` 檔案，但此清單中的輸入原始檔案用於推論。為求簡單，在此範例中，使用與量化相同的輸入清單進行推論。
- **執行階段**：使用者必須選擇特定執行階段，才能對目標執行模型。可用的執行階段選項包括 CPU、GPU 和 DSP (HTP)。

備註

使用以下命令查看工具的幫助內容： `snpe-net-run --help`

### 執行 SNPE .dlc: x86 主機電腦

轉換的 SNPE .dlc 可使用接受 .dlc 及輸入清單作為引數的 `snpe-net-run` 工具執行。在 x86 上執行 SNPE DLC 單純是為了偵錯。

例如：下列命令會產生載入 `inception_v3.dlc` 模型並在 x86 CPU 上執行模型。產生輸出檔案至 `/output_x86/`。

備註

SNPE 中的預設執行階段為 CPU。使用 `snpe-net-run` 執行模型時，不必指定 CPU 執行階段。

${QAIRT_ROOT}/bin/x86_64-linux-clang/snpe-net-run --container ~/models/inception_v3.dlc --input_list ~/models/input_list.txt --output_dir ~/models/output_x86
    Copy to clipboard

![../_images/snpe-run-host.png](data:image/png;base64,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)

### 準備 SNPE 模型以在目標設備上執行

要在目標設備上執行模型，`snpe-net-run` 需要模型的 .dlc、SNPE 執行階段程式庫和輸入清單來執行推斷，進而產生輸出。

備註

在目標設備上執行之前，需將 SNPE SDK 的二進位檔案和函式庫推送至目標設備。

使用以下目錄中的資源：`${QAIRT_ROOT}/lib/aarch64-oe-linux-gcc11.2` 和 `${QAIRT_ROOT}/bin/aarch64-oe-linux-gcc11.2`

使用適用於 Qualcomm 評估套件的正確 DSP Hexagon 架構程式庫。

Tab IQ-8275
Tab IQ-9075
Tab Qualcomm Dragonwing™ RB3 Gen 2

使用 v75 函式庫

使用 v73 函式庫

使用 v68 函式庫

| **檔案** | **來源位置** |
| --- | --- |
| snpe-net-run | ${QAIRT\_ROOT}/bin/aarch64-oe-linux-gcc11.2 |
| libSNPE.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2/ |
| libSnpeHtpPrepare.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2 |
| libSnpeHtpV68Stub.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2 |
| libSnpeHtpV68Skel.so | ${QAIRT\_ROOT}/lib/hexagon-&lt;dsp-arch&gt;/unsigned |
| Inception\_v3\_quantized\_with\_htp\_cache.dlc | ~/models |
| Inception\_v3.dlc | ~/models |
| input\_list.txt | ~/models |
| Inception V3樣本輸入影像 | /tmp/RandomInputsForInceptionV3 |

以下是需要在主機電腦上執行的 scp 命令，用於將 SNPE SDK 函式庫和二進位檔案複製到裝置：

備註

將 `<dsp-arch>` 替換為 `75` (對應 IQ-8275)、`73` (對應 IQ-9075) 和 `68` (對應 Qualcomm Dragonwing™ RB3 Gen 2)。

scp ${QAIRT_ROOT}/bin/aarch64-oe-linux-gcc11.2/snpe-net-run root@[ip-addr]:/opt/
    Copy to clipboard

scp ${QAIRT_ROOT}/lib/aarch64-oe-linux-gcc11.2/libSNPE.so root@[ip-addr]:/opt/
    Copy to clipboard

scp ${QAIRT_ROOT}/lib/aarch64-oe-linux-gcc11.2/libSnpeHtpV<dsp-arch>Stub.so root@[ip-addr]:/opt/
    Copy to clipboard

scp ${QAIRT_ROOT}/lib/aarch64-oe-linux-gcc11.2/libSnpeHtpPrepare.so root@[ip-addr]:/opt/
    Copy to clipboard

scp ${QAIRT_ROOT}/lib/hexagon-v<dsp-arch>/unsigned/libSnpeHtpV<dsp-arch>Skel.so root@[ip-addr]:/opt/
    Copy to clipboard

scp ~/models/inception_v3.dlc root@[ip- addr]:/opt/
    Copy to clipboard

scp ~/models/inception_v3_quantized_with_htp_cache.dlc root@[ip-addr]:/opt/
    Copy to clipboard

scp ~/models/input_list.txt root@[ip- addr]:/opt/
    Copy to clipboard

scp /tmp/RandomInputsForInceptionV3 root@[ip- addr]:/tmp/
    Copy to clipboard

ssh root@[ip-addr]
    Copy to clipboard

cd /opt
    Copy to clipboard

上述步驟讓裝置為模型執行做好準備。以下各節提供關於在可用的執行階段上執行模型的詳細資訊。

### 執行 SNPE .dlc: Arm 架構 CPU

在執行模型之前設置環境變數，以確保 SNPE 二進位文件和函式庫可以訪問以運行模型。

備註

SNPE 中的預設執行階段為 CPU。使用 `snpe-net-run` 執行模型時，不必指定 CPU 執行階段。

export LD_LIBRARY_PATH=/opt:$LD_LIBRARY_PATH
    Copy to clipboard

export PATH=/opt/:$PATH
    Copy to clipboard

snpe-net-run --container inception_v3.dlc --input_list input_list.txt --output_dir output_cpu
    Copy to clipboard

![../_images/snpe-run-cpu.png](data:image/png;base64,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)

### 執行 SNPE DLC：GPU

要在GPU執行環境運行模型，使用 `inception_v3.dlc` 並將輸出保存到 `output_gpu` 中。

備註

要使用 GPU，需使用 `--use_gpu` 參數指定執行環境。

export LD_LIBRARY_PATH=/opt:$LD_LIBRARY_PATH
    Copy to clipboard

export PATH=/opt:$PATH
    Copy to clipboard

snpe-net-run --container inception_v3.dlc --input_list input_list.txt --output_dir output_gpu --use_gpu
    Copy to clipboard

### 執行 SNPE DLC：HTP

要在 HTP 後端上執行模型，使用 `inception_v3_quantized_with_htp_cache.dlc` 並將輸出儲存在 `output_htp` 中。

備註

要使用 HTP，需使用 `--use_dsp` 參數指定執行環境。

export LD_LIBRARY_PATH=/opt:$LD_LIBRARY_PATH
    Copy to clipboard

export PATH=/opt:$PATH
    Copy to clipboard

export ADSP_LIBRARY_PATH="/opt;/usr/lib/rfsa/adsp;/dsp"
    Copy to clipboard

snpe-net-run --container inception_v3_quantized_with_htp_cache.dlc --input_list input_list.txt --output_dir output_htp --use_dsp
    Copy to clipboard

![../_images/snpe-run-htp.png](data:image/png;base64,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)

### 驗證輸出

對於每個輸入的原始檔案透過 input\_list 文件提供給 snpe-net-run，將生成一個輸出資料夾，其中包含輸出張量，這些張量保存為原始檔案，其大小將與模型的輸出層相符，如下圖所示。

備註

使用 [Netron](https://netron.app/) 工具來視覺化模型。

![../_images/snpe-validate-output.png](data:image/png;base64,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)

在 `inception_v3` 範例中，輸出的原始檔案是一個二進位檔案，其中包含 1000 個分類類別的機率。

我們使用 Python 指令碼將檔案讀取為 NumPy 陣列，以執行後處理和輸出驗證。以下範例檢查 HTP 預測與 CPU 預測是否相同。

1. 將輸出檔案從目標裝置複製到主機電腦，以便驗證輸出結果。

ssh root@[ip-addr]
        Copy to clipboard

cd /opt/
        Copy to clipboard

scp -r /opt/output_cpu user@host-ip:<path>
        Copy to clipboard

scp -r /opt/output_htp user@host-ip:<path>
        Copy to clipboard

備註

在以下 (`compare.py`) Python 腳本中使用上述命令 `<path>`。
2. 一旦來自裝置推論的輸出被複製到主機電腦，準備一個腳本，載入保存在 `output_htp` 和 `output_cpu` 的輸出張量進行比較。

    在這個例子中，`output_htp` 和 `output_cpu` 都會被複製到 `~/models` 目錄。

    以下範例指令碼用來比較使用的範例輸入之一的輸出。輸出儲存至主機電腦上的 `/opt/` 目錄。可使用 `numpy.fromfile(…)` API 將 `snpe-net-run` 的輸出載入至 NumPy `ndarrays`。

備註

預設情況下， `snpe-net-run` 會將輸出張量以 **float32** 格式保存為 NumPy 文件。

# python postprocessing script (compare.py) import numpy as np
        
        import numpy as np
        
        htp_output_file_path = "<path>/output_htp/Result_1/875.raw"
        cpu_output_file_path = "<path>/output_cpu/Result_1/875.raw"
        
        htp_output = np.fromfile(htp_output_file_path, dtype=np.float32)
        htp_output = htp_output.reshape(1,1000)
        
        cpu_output = np.fromfile(cpu_output_file_path, dtype=np.float32)
        cpu_output = cpu_output.reshape(1,1000)
        
        # np.argmax gives the cls_id with highest probability from tensor.
        
        cls_id_htp = np.argmax(htp_output)
        cls_id_cpu = np.argmax(cpu_output)
        
        # Let's compare CPU output vs HTP output
        
        print("Cpu prediction {} \n Htp Prediction {}".format(cls_id_cpu, cls_id_htp))
        Copy to clipboard

輸出

![../_images/snpe-cpu-htp-compare.png](data:image/png;base64,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)

### 使用 SNPE API 部署模型

SNPE SDK 提供 C/C++ API 以建立/開發在所選的加速硬體 (CPU、GPU 或 HTP) 上執行模型的應用程式。請參閱示範 SNPE C/C++ API 的 [範例應用程式](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-2/cplus_plus_tutorial.html)。

### 使用 Qualcomm 智慧多媒體 SDK (IM SDK) 部署模型

為了改善建置整個使用案例管道 (從攝影機串流、對影像進行預處理、執行推論等) 時的開發者體驗，已將 SNPE 整合至 Qualcomm IM SDK 中作為 [qtimlsnpe](https://docs.qualcomm.com/bundle/publicresource/topics/80-70020-50/qtimlsnpe.html) 外掛程式。

此外掛程式以 SNPE C API 為基礎，並提供 SNPE 功能 (載入和執行模型)。透過 `qtimlsnpe` 外掛程式，您可以在 Qualcomm IM SDK 管道中使用轉換後的模型 DLC 來實現使用案例。

請參閱 [開發自己的 AI/ML 應用程式](https://docs.qualcomm.com/doc/80-70020-15BT/topic/develop-your-own-application.html)，了解如何使用 Qualcomm IM SDK 部署 SNPE DLC 的說明。

- *class* tabincludedirective

    - ## 使用 AI Engine Direct 部署模型

可透過支援 QNN 的應用程式 (使用 QNN C/C++ API 編寫的應用程式) 部署模型 .so (量化或非量化)。QNN 提供 API 以動態載入模型 .so，並使用所選的後端在硬體上執行模型。

QNN 提供一個預編譯工具 ([qnn-net-run](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/tools.html#qnn-net-run)) ，該工具可動態加載 .so 模型並在選定後端上使用提供的輸入進行推論。

在 CPU、GPU 或 HTP 上執行時，qnn-net-run 需要以下三個參數：

- **模型檔案**：`qnn-model-lib-generator` 產生的 .so 檔案
- **後端檔案**：用於目標後端的 .so 檔案

> 
> 
> - CPU 後端：`libQnnCpu.so`
>     - GPU 後端：`libQnnGpu.so`
>     - HTP 後端：`libQnnHtp.so`
- **輸入清單**：文字檔案，例如在量化過程中使用的 input\_list.txt 檔案，但此清單中的輸入原始檔案用於推論。為求簡單，此範例使用與量化相同的輸入清單進行推論。

### 執行 QNN .so: x86 主機電腦

轉換的 QNN .so 可使用接受模型 .so、後端程式庫 .so, 與輸入清單作為引數的 `qnn-net-run` 工具執行。

例如：下列命令會載入 `libinception_v3.so` 模型並在 x86 CPU 上執行模型。產生輸出檔案至 `~/models/output_x86/`。

${QAIRT_ROOT}/bin/x86_64-linux-clang/qnn-net-run --model ~/models/libs/x86_64-linux-clang/libinception_v3.so --backend ${QAIRT_ROOT}/lib/x86_64-linux-clang/libQnnCpu.so --input_list input_list.txt --output_dir ~/models/output_qnn_x86
    Copy to clipboard

### 準備 QNN 模型以在目標設備上執行

要在目標設備上執行模型，`qnn-net-run` 需要模型的 .so、QNN 二進位檔案與執行階段程式庫和輸入清單來執行推斷，進而產生輸出。

備註

在目標設備上執行之前，需將 QNN SDK 的二進位檔案和執行階段程式庫推送至目標設備。

使用以下目錄中的資源：`${QAIRT_ROOT}/bin/aarch64-oe-linux-gcc11.2` 和 `` ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2/``

使用適用於 Qualcomm 評估套件的正確 DSP Hexagon 架構程式庫。

Tab IQ-8275
Tab IQ-9075
Tab Qualcomm Dragonwing™ RB3 Gen 2

使用 v75 函式庫

使用 v73 函式庫。

使用 v68 函式庫。

| **檔案** | **來源位置** |
| --- | --- |
| qnn-net-run | ${QAIRT\_ROOT}/bin/aarch64-oe-linux-gcc11.2 |
| libQnnHtp.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2/libQnnHtp.so |
| libQnnCpu.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2/libQnnCpu.so |
| libQnnGpu.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2/libQnnGpu.so |
| libQnnHtpPrepare.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2 |
| libQnnHtpV68Stub.so | ${QAIRT\_ROOT}/lib/aarch64-oe-linux-gcc11.2 |
| libinception\_v3\_quantized.s o | ~/models/libs/aarch64-oe-linux-gcc11.2 |
| libinception\_v3.so | ~/models/libs/aarch64-oe-linux-gcc11.2 |
| libQnnHtpV68Skel.so | ${QAIRT\_ROOT}/lib/hexagon-&lt;dsp-arch&gt;/unsigned |
| libqnnhtpv68.cat | ${QAIRT\_ROOT}/lib/hexagon-&lt;dsp-arch&gt;/unsigned |
| libQnnSaver.so | ${QAIRT\_ROOT}/lib/hexagon-&lt;dsp-arch&gt;/unsigned |
| libQnnSystem.so | ${QAIRT\_ROOT}/lib/hexagon-&lt;dsp-arch&gt;/unsigned |

以下需要在主機上執行的 `scp` 命令，用於將 QNN SDK 函式庫和二進位檔案複製到裝置。

備註

將 `<dsp-arch>` 替換為 `75` (對應 IQ-8275)、`73` (對應 IQ-9075) 和 `68` (對應 Qualcomm Dragonwing™ RB# Gen 2)。

scp ${QAIRT_ROOT}/bin/aarch64-oe-linux-gcc11.2/qnn-* root@[ip-addr]:/opt/
    Copy to clipboard

scp ${QAIRT_ROOT}/lib/aarch64-oe-linux-gcc11.2/libQnn*.so root@[ip-addr]:/opt/
    Copy to clipboard

scp ${QAIRT_ROOT}/lib/hexagon-v<dsp-arch>/unsigned/* root@[ip-addr]:/opt/
    Copy to clipboard

scp ~/models/libs/aarch64-oe-linux-gcc11.2/* root@[ip-addr]:/opt/
    Copy to clipboard

scp ~/models/input_list.txt root@[ip-addr]:/opt/
    Copy to clipboard

scp /tmp/RandomInputsForInceptionV3 root@[ip-addr]:/tmp/
    Copy to clipboard

ssh root@[ip-addr]
    Copy to clipboard

cd /opt/
    Copy to clipboard

上述步驟讓裝置為模型執行做好準備。以下各節提供關於在可用的執行階段上執行模型的詳細資訊。

### 執行 QNN .so: Arm 架構 CPU

在基於 Arm 的 CPU 上運行模型時，`qnn-net-run` 需要模型 .so 檔案、後端 .so 檔案及輸入清單來進行推論並生成輸出。

對於 RB3Gen2 目標，.so 必須使用 `aarch64-oe-linux-gcc11.2` 工具鏈進行交叉編譯。

例如，以下命令將輸出檔案寫入 `/opt/output_cpu/`。

export LD_LIBRARY_PATH=/opt/:$LD_LIBRARY_PATH
    Copy to clipboard

export PATH=/opt:$PATH
    Copy to clipboard

qnn-net-run --model libinception_v3.so --backend libQnnCpu.so --input_list input_list.txt --output_dir output_cpu
    Copy to clipboard

### 在 GPU 上執行 QNN 模型

在 Adreno GPU 上執行模型時，`qnn-net-run` 需要模型 .so 檔案、後端 .so 庫 (如 `libQnnGpu.so`) 及輸入清單來執行推論並生成輸出。

例如，以下命令將輸出檔案寫入 `/opt/output_gpu/`。

export LD_LIBRARY_PATH=/opt/:$LD_LIBRARY_PATH
    Copy to clipboard

export PATH=/opt:$PATH
    Copy to clipboard

qnn-net-run --model libinception_v3.so --backend libQnnGpu.so --input_list input_list.txt --output_dir output_gpu
    Copy to clipboard

### 在 HTP 後端執行 QNN 模型

要在 HTP 後端上執行模型，需使用 `libquantized_inception_v3.so` 模型檔案及 `libQnnHtp.so` 後端函式庫，並將輸出儲存在 `output_htp` 目錄中。

export LD_LIBRARY_PATH=/opt/:$LD_LIBRARY_PATH
    export PATH=/opt:$PATH
    Copy to clipboard

export ADSP_LIBRARY_PATH="/opt/;/usr/lib/rfsa/adsp;/dsp"
    Copy to clipboard

qnn-net-run --model libinception_v3_quantized.so --backend libQnnHtp.so --input_list input_list.txt --output_dir output_htp
    Copy to clipboard

### 驗證輸出

對於每個輸入的原始檔案透過 input\_list 文件提供給 `qnn-net-run`，將生成一個輸出資料夾，其中包含原始輸出檔案，其大小將與模型的輸出層相符，如下圖所示。

![../_images/qnn-validate-output.png](data:image/png;base64,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)

在這個 `inception_v3` 範例中，原始輸出檔是一個二進位檔案，包含了 1000 個分類類別的概率。

您可以使用 Python 指令碼將檔案讀取為 NumPy 陣列，以執行後處理和輸出驗證。以下範例指令碼檢查 HTP 預測與 CPU 預測是否相同。

ssh root@[ip-addr] cd /opt/
    Copy to clipboard

scp -r /opt/output_cpu user@host-ip:<path>
    Copy to clipboard

scp -r /opt/output_htp user@host-ip:<path>
    Copy to clipboard

備註

在以下 Python 腳本中使用上述命令中的 `<path>`。

一旦將 `qnn-net-run` 工具的輸出從裝置複製到主電腦，請創建一個簡單的 Python 腳本以載入 CPU 和 HTP 執行的輸出並使用 NumPy 進行比較。

#python postprocessing script (compare.py)
    
    import numpy as np
    
    htp_output_file_path = "<path>/output_htp/Result_1/875.raw"
    cpu_output_file_path = "<path>/output_cpu/Result_1/875.raw"
    
    htp_output = np.fromfile(htp_output_file_path, dtype=np.float32)
    htp_output = htp_output.reshape(1,1000)
    
    cpu_output = np.fromfile(cpu_output_file_path, dtype=np.float32)
    cpu_output = cpu_output.reshape(1,1000)
    
    cls_id_htp = np.argmax(htp_output)
    cls_id_cpu = np.argmax(cpu_output)

    # Let's compare CPU output vs HTP output
    print("CPU prediction {} \n HTP prediction {}".format(cls_id_cpu, cls_id_htp))
    Copy to clipboard

輸出

![../_images/qnn-cpu-htp-compare.png](data:image/png;base64,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)

### 使用 QNN API 部署模型

Qualcomm AI 引擎 Direct SDK 提供 C/C++ API 以建立/開發可載入已編譯的 .so 模型並在所選的加速硬體 (CPU、GPU 或 HTP) 上加以執行的應用程式。請參閱示範 QNN C/C++ API 的 [範例應用程式](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)。

### 使用 Qualcomm IM SDK 部署模型

為了改善建置整個使用案例管道 (從攝影機串流、對影像進行預處理、執行推論等) 時的開發者體驗，已將 QNN 整合至 Qualcomm IM SDK 中作為 [qtimlqnn](https://docs.qualcomm.com/bundle/publicresource/topics/80-70020-50/qtimlqnn.html) 外掛程式。

此外掛程式以 QNN C API 為基礎，並提供動態載入和執行模型的功能。透過 `qtimlqnn` 外掛程式，開發者可在 Qualcomm IM SDK 管道中使用經過轉換和編譯的模型函式庫來實現使用案例。

請參閱 [開發自己的 AI/ML 應用程式](https://docs.qualcomm.com/doc/80-70020-15BT/topic/develop-your-own-application.html)，了解如何使用 Qualcomm IM SDK 部署 QNN 模型的說明。

Last Published: Dec 23, 2025

[Previous Topic
移植模型](https://docs.qualcomm.com/bundle/publicresource/80-70020-15BT/topics/port-models.md) [Next Topic
執行 AI/ML 範例應用程式](https://docs.qualcomm.com/bundle/publicresource/80-70020-15BT/topics/ai-ml-reference-apps.md)