![Альберт Иванов](/img/default-banner.jpg)
- 126
- 103 947
Альберт Иванов
Registrace 10. 10. 2012
monodepth (midas) on cpu and NCS 2(MYRIAD) on raspberry pi 4b
docs.openvino.ai/2022.3/notebooks/201-vision-monodepth-with-output.html
6.692309379577637 seconds CPU
0.17417049407958984 seconds MYRIAD
6.692309379577637 seconds CPU
0.17417049407958984 seconds MYRIAD
zhlédnutí: 14
Video
arducam64mp - Failed to allocate buffers: Cannot allocate memory - fix the error
zhlédnutí 75Před měsícem
arducam64mp cant capture at full resolution with picamera2. even with libcamera the error persists. so there is a solution provided. see also: forum.arducam.com/t/unable-to-request-2-buffers-cannot-allocate-memory-with-2-64mp-cameras/6143/15 sudo nano /boot/cmdline.txt add at the end: alloc_in_cma_threshold=16 sudo nano /boot/config.txt change: dtoverlay=vc4-kms-v3d to: dtoverlay=vc4-kms-v3d,cm...
how to install Intel® Neural Compute Stick 2 on Raspbian x64 with openvino
zhlédnutí 38Před měsícem
main guide: docs.openvino.ai/2022.3/openvino_docs_install_guides_installing_openvino_raspbian.html python 3.9 the last version supported at the moment. mkdir intel && cd intel wget storage.openvinotoolkit.org/repositories/openvino/packages/2022.3.2/linux/l_openvino_toolkit_debian9_2022.3.2.9279.e2c7e4d7b4d_arm64.tgz -O openvino_2022.3.2.tgz sudo tar -xf openvino_2022.3.0.tgz mv l_openvino_toolk...
не работает вывод звука на наушники (jack) на raspberry pi - чиним
zhlédnutí 3Před měsícem
не работает вывод звука на наушники (jack) на raspberry pi - чиним
face_detector dlib python vs mediapipe face_detector python
zhlédnutí 38Před měsícem
speed tests face detectors on live stream github.com/zoldaten/dlib_and_mediapipe
EAST_text_detection tflite model speed test on raspberry pi 4b python 3.9 bullseye х64
zhlédnutí 23Před měsícem
download model from here - www.kaggle.com/models/spsayakpaul/east-text-detector/tfLite/int8 int8 or float32 (cant remember exactly) code: github.com/zoldaten/PINTO_model_zoo/blob/main/051_East_Text_Detection/text_detection_video_tflite.py this is adopted working code (original from PINTO is deprecated)
paddleocr on gpu gtx 3050 speed test
zhlédnutí 103Před měsícem
import time,cv2 from paddleocr import PaddleOCR,draw_ocr import paddle #paddle.device.is_compiled_with_cuda() paddle.utils.run_check() ocr = PaddleOCR(use_gpu=True, use_angle_cls=False, rec_model="ch_PP-OCRv2_rec_infer_opt.nb",drop_score=0.8,\ warmup=False,show_log=False,use_onnx=False, lang='en') # need to run only once to download and load model into memory #img_path = 'img_12.jpg' #img_path ...
M.2 google accelerator on ubuntu pc
zhlédnutí 107Před měsícem
1.echo "deb [amd64] packages.cloud.google.com/apt coral-edgetpu-stable main" | sudo tee /etc/apt/sources.list.d/coral-edgetpu.list 2.curl packages.cloud.google.com/apt/doc/apt-key.gpg | sudo apt-key add - 3.sudo apt-key adv keyserver keyserver.ubuntu.com recv-keys E88979FB9B30ACF2 4. sudo apt-get update 5.sudo apt-get install gasket-dkms libedgetpu1-std 6.sudo apt-get install python3-pycoral ad...
Phi-3-mini-4k-instruct-onnx (onnxruntime-genai) test on raspberry pi4b bullseye
zhlédnutí 78Před měsícem
1. cmake and g update needed. remove them. build cmake from source. g : add deb ftp.de.debian.org/debian bookworm main to /etc/apt/sources.list (packages.debian.org/bookworm/arm64/g -11/download) sudo apt update sudo apt install g 2. install genai - onnxruntime.ai/docs/genai/howto/install ./build.sh skip_tests config Release download model: huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx/t...
edge-tpu-silva+picamera2 example
zhlédnutí 67Před měsícem
see modified example - github.com/zoldaten/edge-tpu-silva/blob/main/example/object_detect.py original - github.com/DAVIDNYARKO123/edge-tpu-silva net works very fast but badly on detections (in compare with original yolo)
google coral usb accelerator + picamera2 classify_example on Raspbian Bullseye 64
zhlédnutí 44Před měsícem
реализация picamera2, которая работает с libcamera вместо legacy raspicam c моделями классификации с небольшими моделями inference 12-15мс, а вот с imagenet 1,1 сек. но это и понятно - в imagenet 1000 классов.
how to install coral usb accelerator for video stream use with csi camera raspberry pi 4 Bookworm
zhlédnutí 78Před 2 měsíci
1. usb coral need python 3.9. it doesnt support 3.11 which is native for Raspbian Bookworm. one way - install python 3.9 with pyenv. 2. install coral packages inside python 3.9 see github.com/google-coral/pycoral/issues/140 Download pycoral-2.0.0-cp39-cp39-linux_aarch64.whl and tflite_runtime-2.5.0.post1-cp39-cp39-linux_aarch64.whl from github.com/google-coral/pycoral/releases/tag/v2.0.0 pip in...
google coral USB Accelerator test on raspberry pi 4b Bookworm
zhlédnutí 53Před 2 měsíci
coral.ai/docs/accelerator/get-started/#requirements need python3.9 to install. python 3.11 (which is native for Bookworm) is not working with. build from source for 3.11 ? may be. 1. pyenv install -v 3.9.9 2. install wheels: Download pycoral-2.0.0-cp39-cp39-linux_aarch64.whl and tflite_runtime-2.5.0.post1-cp39-cp39-linux_aarch64.whl from github.com/google-coral/pycoral/releases/tag/v2.0.0 pip i...
phi3 vs gemma2b on raspberry pi cm4 tests
zhlédnutí 47Před 2 měsíci
original: github.com/microsoft/Phi-3CookBook/blob/main/md/02.QuickStart/Ollama_QuickStart.md тест небольших моделей через ollama на Raspbian bullseye x64 cm4. выводы: - phi3 дает более емкие ответы - phi3 прогружается быстрее - на русском языке модели выдают практически одинаково плохие результаты попросту переводя англ. вариант ответа.
тестирование Gemma 2B через ollama на raspberry pi 4b
zhlédnutí 65Před 2 měsíci
тестирование Gemma 2B через ollama на raspberry pi 4b
llamafile test with OmniFusion - вопросы llama на русском
zhlédnutí 72Před 2 měsíci
llamafile test with OmniFusion - вопросы llama на русском
OmniFusion - speed test on raspberry pi 4b Raspbian Bookworm
zhlédnutí 46Před 3 měsíci
OmniFusion - speed test on raspberry pi 4b Raspbian Bookworm
libjxl (compress image from Google) test on windows
zhlédnutí 17Před 3 měsíci
libjxl (compress image from Google) test on windows
создание субтитров на русском с помощью whisper
zhlédnutí 306Před 3 měsíci
создание субтитров на русском с помощью whisper
making subtitles with whisper fix article code for windows
zhlédnutí 13Před 3 měsíci
making subtitles with whisper fix article code for windows
VK16E G28U7F GPS module test on raspberry pi 4b
zhlédnutí 40Před 3 měsíci
VK16E G28U7F GPS module test on raspberry pi 4b
install craft-text-detector for opencv up 4.5 version
zhlédnutí 165Před 4 měsíci
install craft-text-detector for opencv up 4.5 version
yoloV9 yolov9-c.pt test on windows on cuda 12.1
zhlédnutí 55Před 5 měsíci
yoloV9 yolov9-c.pt test on windows on cuda 12.1
how to remove nsfw filter in facefusion 2.2.1
zhlédnutí 15KPřed 5 měsíci
how to remove nsfw filter in facefusion 2.2.1
Is this the time spent both for inference and writing the output image, or only inference?
inference
2.6.1 doesn't work
Can I start server from any llamafile? If I will start like ./llava-v1.5-7b-q4.llamafile -ngl 9999 -m OmniFusion-1.1-Q5_K_M.gguf --mmproj mmproj-model-f16.gguf is it suppose work correct?
they say so. try.
@@user-os1xg1rh4c I tried, when I send there an image, it breaks up, starting always give some strange bullshit like” you, re you don’t you are need you are you …” but without image, it talks normal
thanks for this video . I have problem when I ran the code I gets the error " import libcamera ModuleNotFoundError: No module named 'libcamera' " I use python 3.9.2
How did you achive to install picamera2 in Python 3.9.2? I tried a lot but I cannot use that python version and install picamera in my virtual env. I'm struggling a lot.
@@abelreybarreiros1425 picamera2 have no problem with python3.9 and 3.11. coral has.
I have same problem . are you find solution for this problem
I am trying to configure raspberry pi 4b with edge tpu silva (google coral for tflite models) but I am struggling because I am using python 3.9 in virtual env but I cant install picamera2 with pip. By the way the picamera2 package is native installed in my raspberry but I cannot use it for my scripts for ML purposes. As I have the Pi camera v3 model, I cannot use another library to obtain the image/video from my camera to use it with opencv or whatever. Can you help me or do you know some choice to use my camera with tflite and the Gcoral to apply my models ?
@@abelreybarreiros1425 better choice is to move to Bullseye. picamera2 based on libcamera and one’s should keep in mind that coral only supports python3.9 (the last version). in Bookworm python3.11 is native and only way there - use gstreamer or the something similar.
Funciona en la version 2.6.0 de: PROBABILITY_LIMIT = 0,80 a: PROBABILITY_LIMIT = 1,00
Не работает. Пишет "Не удалось подключиться"
Thought that phi3mini onnx was actually not that bad…
perhaps, compilation without optimisation. they wrote would think about it.
Is actually that shitty? Fuck
有用
does it work on 2.6
Really?
does the onxx file in the download contain all of the data/images from the objects365 dataset or did you train it on just those 8 classes shown in the video?
all of the data/images from the objects365 dataset.
@@user-os1xg1rh4c thanks bro
thank you it's worked
Guys this was a funny read lol and its much easier than your attempts on here lol any kind of bypass is set to command line args in user.bat just run line in there set as args bypass to tile. just as you would for - - Bypass Torch - - cuda maddock etc
hi can u show mw how u do that
Hi, it is not opening the window when i do the same as you do ?
you use python before? the idle is from python and always open
but if you cant for me it show me the option "edit" and edit like a .txt, and if you cant too, try to install an programming editor like notepad
Как на макбуке отключить? Очень надо)
😂
Can you tell me how to get an entire sentence/line as one polygon with craft?
u may work with text_threshold in craft initial settings, i.e.: craft = Craft( output_dir=None, rectify=True, export_extra=False, text_threshold=0.7, if it doesnt help - look at refinet as part of craft as it responsible for link words together after craft work done.
work in 2.5.2
работает или не работает?
@@K-A_Z_A-K_S_URALA у меня работало на 2.5.2
круто!
но долго.
Интересно, какая задержка обработки, и raspberry pi3b потянет ли whisper?
I never managed to get any usable image from the Radxa 4k camera with the rock5b board. Would you be able to explain how to to install the software and how to configure the camera to get colour images?
github.com/Joshua-Riek/ubuntu-rockchip/issues/288
Please help me on 2.4 colab! 😊
I need help with pinokio ff 2.4.1
Gracias, funciona correctamente ❤😂😅
Очень подробный обзор. Спасибо. Подскажите какой температуры ставить термопредохранитель?
v 2.4.0 it works only on preview (
but it works for pc installation? if I use Pinokio browser?
works on Pinokio 2.4.1
@@hieuluu2 thanks but not understood how to edit that commands on browser version
For all versions of facefusion, change line 28 as follows: from: PROBABILITY_LIMIT = 0.80 to: PROBABILITY_LIMIT = 1.00
can you confirm it working for the latest version (2.4.1) ?
@@saikomatirikis3520 yes...
not working for me@@saikomatirikis3520
@@saikomatirikis3520pas pour moi
@@saikomatirikis3520 Just tested it. It works!
doesn't work for 2.4
Worked for me, thanks!
Great demonstration, can I ask you a few questions about your config. Did you use SSD or SD memory card for booting and etc, what about cooling and also other stuff like increasing swap file size (not very effective in this particular project but it can help to reduce the pressure of the ram a little I think). Would you kindly share some details about the whole process, thanks in advance.
sd card used. it does not matter it this case ssd or sd actually. raspberry pi with 8gb ram. its enough for mistral. but the modei is quite huge for raspberry. try this one - czcams.com/video/Vx2YFaM7tSs/video.html
hi,bro ,I use Mac m1 and use Pinokio to setup fasfusion 2.3.0 ,I can't bypass nsfw, need help🤣
Did you find the solution yet?
bro the file also exist even if you have pinokio you can change the propability limit !!
@@cgiguy hey, when i click on content_analyser then edit with idle nothing happens
Sorry i didn't understand what have you changed? Please detail?
thanks bro
Спасиба брат
Also fine with 2.3
Спасибо тебе! Теперь наконец то все VPN подключения работают
sposibo drug. vsjo robotajet 👍👍👍
Sorry i didn't understand what have you changed
MAX_PROBABILITY
Учим теорию и ставим 1080i разрешение, получаем нагрузку почти в 2 раза меньшую, стик не перегревается и все ок
How to use the navigation stack with the Pioneer robot (with rosary) and the rplidar a3 sensor (Hector Slam or Gmapping library) Hey, My question is how to use the ros Navigation Stack library to perform autonomous navigation on the pioneer robot that I am using the rosaria library for control, I am also using the rplidar_03 sensor with the hector slam library in the project. If anyone can help me :)
Thank you, this worked.
Спасибо тебе добрый человек!
А что нельзя было нприпаять кулер прям к разъёму мини юсби?
У меня такой сгорел через 5 месяцев, сейчас заказал второй и тоже буду ставить радиатор
У меня тоже такой стик , тупит нереально пользоваться невозможно, не стал ничего колхозить , заказал новую приставку тока уже в полном формате и думаю там с охлаждением должно быть куда лучше
смотрится максимально колхозно... сопли из проводов, отдельные блоки питания, вентиляторы. Не хватает еще синей изоленты.
4к работает год без нареканий. Включен 24/7