Default Avatar

Ethon Lin

Ethon Lin

About

Username
Ethon Lin
Joined
Visits
1,325
Last Active
Roles
Member, ModeratorsKL520, ModeratorsKL720, ModeratorsAIModel

Comments

  • 若模型是自行訓練的話就要考慮到模型所需要的輸入格式來修正對應的scpu, ncpu 一般會建議先透過PLUS中的gereric範例來對模型做推論，因gereric範例會將模型透過usb傳入NPU中推論，可以避免在fw上有相容性問題，若能正確拿到推論答案，再開始著手準備對應的fw，以便將fw…
    in KL520 board ncpu error Comment by Ethon Lin June 2022
  • Hello, For KL720, you can use the example "solution_kdp2_hico_mipi" in <KL720_SDK_1.5.0>. You can fine the usage about using mipi sensor to send image and inference with NPU. Here is the repository of the example project file: /KL72…
    in Inference using the Firmware SDK Comment by Ethon Lin June 2022
  • 你誤會了，雖然電腦有抓到USB to TTL (UART)的線，但我指的是在執行flash_programmer.py的時候程式抓不到該USB port的連線，下列幾種情況都可能導致抓不到 沒有連接到電腦上 (由你提供的訊息可以排除此項) 設定的USB port number有誤，因此請你檢查ttyUSBx的x是多少 該…
    in Flash upgrade failed Comment by Ethon Lin June 2022
  • 看起來可能是沒有抓到對應UART的usb port，可以先輸入下面指令來列出所有的裝置 ls /dev 連接UART的usb線應該會是 ttyUSBx (x為數字) 即可對應修改setup.py中的ID，改為ttyUSBx
    in Flash upgrade failed Comment by Ethon Lin May 2022
  • 從InvalidProgramInput來推斷，有可能是輸入的參數有誤，如input op's name, input data等，可以再參考文件確認你使用的參數 http://doc.kneron.com/docs/#toolchain/command_line/#appendix
    in 在量化模型時出現錯誤 Comment by Ethon Lin May 2022
  • 確認一下，所以onnx轉換問題在切除不支援的op後就可以轉換了嗎? 另外，實際硬體與原始模型運行的比較，這邊指的是onnx模型的推論結果與實際上在520 / 720上執行嗎? 所以已經有成功在520 / 720上執行了嗎? 請問要使用的是520還是720呢? 執行ktc.kneron_inference 運行時…
    in 模型轉換問題 Comment by Ethon Lin May 2022
  • Hello, 想請問一下GT指的是什麼?是最後的答案嗎? 在使用bypass時最常見的問題是輸入資料除了要先做完模型的preprocessing之外，還需要把資料排成NPU所需的格式後依序傳入，可以參考連結的說明，將輸入資料依序排列成R -> G -> B -> A後再輸入，若為3 channel …
    in 使用KL520推論與量化結果不符 Comment by Ethon Lin May 2022
  • There is no specific size limitation for model. But to implement model inference on Kneron's device, there is memory limitation on chip. All of following functions cost memory, model data(NEF), input image, inference output data, npu working buffer…
    in Tflite to Onnx Comment by Ethon Lin May 2022
  • Hello, 請問是使用管理者權限執行嗎? sudo ./DFUT_console 若還是有問題的話，可以提供完整的錯誤訊息截圖嗎? 希望可以包含從輸入指令開始的完整訊息
    in 樹莓派連接問題 Comment by Ethon Lin May 2022
  • Hello, 想請問你使用的是KL520還是KL720呢? 初步看起來模型中含有不支援的operator，這可能會導致後續的錯誤，請參考下圖將紅線以下的部分移除，屆時你會拿到四個output data( 51x28x28, 34x14x14, 34x7x7, 51x3x3)，再將這四個output node的資料在PC上做後續reshape, …
    in 模型轉換問題 Comment by Ethon Lin May 2022
  • Hello, 目前有遇過因為樹梅派版本問題，導致編譯好的DFUT執行檔無法使用，為了避免環境相容性問題，建議可以自行編譯plus v1.3中的c example來生成DFUT執行檔，詳細的操作可以參考下面討論連結 https://www.kneron.com/forum/discussion/comment/905/#Comment_905
    in 樹莓派連接問題 Comment by Ethon Lin May 2022
  • Have you tried to update firmware on windows instead of ubuntu on VM. Or neither on ubuntu nor windows couldn't work well? During flash is programming, the system would reset KL720 and reconnect the usb between KL720 and PC. Sometime user encounter…
    in Inference error and firmware update error for KL720 Comment by Ethon Lin May 2022
  • Hello, You mentioned that you can get the correct result with dongle but got wrong abnormal output with host mode on 96 board. Is my understanding right? I wonder whether are this two results are using different input image format on each platform. …
    in Customized model using DevKit with KL520 and SDK Comment by Ethon Lin May 2022
  • Hello, 因為模型的channel為1，而目前的範例是針對image inference設計，所以都是3channel架構，建議可以試著參考bypass process的範例，並且將你的input修改成rgba 4channel格式來試試。 for example: 若原本輸入資料為 0, 1, 2, 3, 4, ... 則修改為 0 0 0 0, 1 0 0 0…
    in kl720推理出現TIMEOUT問題 Comment by Ethon Lin May 2022
  • Hello, It's different between IMAGE_FORMAT_PARALLEL_PROC and IMAGE_FORMAT_RAW_OUTPUT. User can set the bit IMAGE_FORMAT_PARALLEL_PROC to enable or disable parallel mode, for instance, img_run[i].img_cfg.image_format |= IMAGE_FORMAT_PARALLEL_PROC; …
    in Customized model using DevKit with KL520 and SDK Comment by Ethon Lin May 2022
  • Hello, If you are using a channel(R, G, B) model with RGBA8888 input, it's ok to set channel A = 0 to fit the format. For the sequence of RGBA data, please refer to the comment of the link. https://www.kneron.com/forum/discussion/comment/760/#Commen…
    in Customized model using DevKit with KL520 and SDK Comment by Ethon Lin May 2022
  • Hello, Are you using DFUT tool built by Kneron? (Kneron_DFUT_ubuntu_1.3.0.zip) Or are you using the tool built by your self? (inside PLUS v1.3.0) Or maybe you can tried to update firmware(/kneron_plus_v1.3.0/res/firmware/KL720) in Windows with our …
    in Inference error and firmware update error for KL720 Comment by Ethon Lin May 2022
  • Hello, 影響量化結果的可能性很多，可以先檢查下面兩點 用來做量化分析的圖片是否皆與訓練模型或是推論場景有關? 且圖庫中應盡可能的涵蓋各種使用場景的圖片 增加量化用的圖片
    in 量化後模型結果與原本的差很多 Comment by Ethon Lin May 2022
  • 如同你的截圖，你可以選擇透過列出的兩種方式來將dongle中原本host_lib的fw更新成PLUS專用的fw use command line use GUI
    in KL520 使用自訓練模型遇到問題 Comment by Ethon Lin April 2022
  • 抱歉，但因為host_lib已經是去年的版本了，建議是更換到新架構PLUS上使用。 若堅持要使用host_lib的話，可以試著參考 https://www.kneron.com/forum/categories/innoserve-area 內的文章，有過去大家所遇到問題的紀錄 常見的是訓練後的模型參數沒有正確的設定到host_li…
    in KL520 使用自訓練模型遇到問題 Comment by Ethon Lin April 2022
  • 是的，截圖訊息顯示 updating fw reading file failed <= 失敗 reading model failed <= 失敗 report status succeeded 讀取版號成功，但上面的燒錄fw與模型都失敗
    in KL520 使用自訓練模型遇到問題 Comment by Ethon Lin April 2022
  • Because you are using on VM, please check whether the KL720 reconnect to PC after reset process. ( The OS will reset usb device with some commands in processing.) And did you still encounter firmware updating issue after adding permission setting? T…
    in Inference error and firmware update error for KL720 Comment by Ethon Lin April 2022
  • 問題看起來是當host lib在對firmware下達inference命令時出錯，推測應該是你目前燒錄的fw不支援該應用 但因為host_lib的firmware已經接近一年沒有在更新了 ，可能得請你確認一下你手邊是否有該應用所搭配的firmware可以使用。
    in KL520 使用自訓練模型遇到問題 Comment by Ethon Lin April 2022
  • Academy 僅供特定的模型框架使用，一般的NEF僅能在PLUS(sdk)上運行，沒辦法透過Academy執行。 Academy框架的模型轉換，可以參考這篇討論內的連結 https://www.kneron.com/forum/discussion/comment/958/#Comment_958
    in KL520 使用自訓練模型遇到問題 Comment by Ethon Lin April 2022
  • You can try the following link.  http://services.kneron.com.tw:7070/
    in Would you share the link that how to transfer .tflite file to ota.bin? (about Teachable Machine) Comment by Ethon Lin April 2022
  • It seems like Academy got some compatible issue on your PC. Please check the connection status of dongle and camera. Academy utility can only get the first camera(index 0) on your PC, we don't provide the selection of camera on tool.
    in Kneron Academy can' t access my notebook(Lenovo P14s) camera module(Realtek). Comment by Ethon Lin April 2022
  • Yes, they are only compatible on Win/Ubuntu now. But we will arrange some patches for you to make they work on Raspberry Pi. It should cost some time and we will inform you by mail after the patches are ready .
    in Register Raspberry PI device Comment by Ethon Lin April 2022
  • Hi Mishra, KL520 only can output the result of last layer in model. And due to the memory limitation, python or opencv is too large to process on KL520 (firmware). But the SDK "PLUS" we provided can do that easily. PLUS is a software SDK f…
    in Implement GRADCAM Comment by Ethon Lin April 2022
  • 一般來說，直接使用編譯好的tool是有可能遇到系統不相容，但目前尚未收過在同一個環境底下編譯後無法執行回報。 segmentation fault錯誤是因為不正確存取記憶體導致，可能的原因很多，建議可以找塊新的記憶卡重新安裝Raspbian後再試，耐能的網站上也有提供相關檔案下載…
    in KL520 使用自訓練模型遇到問題 Comment by Ethon Lin April 2022
  • Hi Myron, Now the applications on KNEO are only compatible with Windows and Ubuntu. Please run these applications on PC with these two OS. For your case, we will re-evaluate whether should we release the libraries for Raspberry Pi or not in the futu…
    in Register Raspberry PI device Comment by Ethon Lin April 2022

© Kneron, Inc.